What's in IT for Us? - Schlumberger/media/Files/resources/oilfield_review/ors97/... · Schlumberger Wireline & Testing, Aberdeen, Scotland. ClientLink, InterACT, MAXIS (Multitask

What’s in IT for Us?

The WorldWide Web is accumulating some 200,000 new pages daily; an

even greater number of electronic messages are sent to individuals or

posted on specialized bulletin boards every day; databases proliferate.

These facts proclaim the amazing growth in the use of information

technology (IT). But they also beg the key question: “With so much

information milling about the system, what’s in IT for us?”

2 Oilfield Review

Guillermo ArangoMontrouge, France

Nick ColleyBritish Gas International Exploration & ProductionReading, England

Christine Connelly Kent GreenesKeith PearseBP Exploration Operating Co. Ltd.Sunbury on Thames, England

Jerome DenisPeter HighnamAustin, Texas, USA

Charles DurbecClamart, France

Larry GutmanDavid SimsSugar Land, Texas

Stuart JardineTim Jervis Reid SmithCambridge, England

Richard MilesUniversity of CambridgeCambridge, England

For help in preparation of this article, thanks to Joe Amlin, Schlumberger Limited, Sugar Land, Texas,USA; Harry Barrow, Schlumberger Cambridge Research, Cambridge, England; Eddie D’Souza, Omnes, Houston,Texas; Douglas Gray-Stephens, Chris Kenyon, Bill MacGregor, David Scheibner and Claire Vishik,Schlumberger Austin Product Center, Austin, Texas; Per Helgaker, Geco-Prakla, Oslo, Norway; Jerry Huchital, Dowell, Sugar Land, Texas; Khemarata Kunsuik-Mengrai and Alain Michel, Schlumberger Wireline & Testing, Montrouge, France; and Neill Wylie,Schlumberger Wireline & Testing, Aberdeen, Scotland.

ClientLink, InterACT, MAXIS (Multitask Acquisition andImaging System), MDT (Modular Formation DynamicsTester), PEPTEC, SuperVision and TCV are marks ofSchlumberger. TWS (Trusted Web Service) and INCA(Intranet Node Corporate Access) are marks of Omnes.Lotus Notes is a mark of Lotus Development Corp.Netscape Enterprise is a mark of Netscape Communications Corporation. PGP is a mark of PrettyGood Privacy, Inc. BackWeb is a mark of Interad Ltd.Pointcast is a mark of Pointcast, Inc. Marimba is a markof Marimba, Inc. AppleShare, Macintosh and AppleTalkare marks of Apple Computer, Inc.

1. Cursus Publicus may be literally translated to mean“Public Road.”

2. The cost of sending a package was $5 per half-ounce.

It is wrong to think of information networksas a new phenomenon. In the 2nd centuryAD, the Roman Emperor Septimius Severusruled an empire that covered about onethird of the world. As ever, accurate infor-mation was vital, and his network was theCursus Publicus.1 It centered on personalmessengers called Nuntii who handcarriedinformation to and from Rome.

Improving on an existing system, Septim-ius Severus established “service stations”every 20 miles or so along all the mainroutes through the empire. By simply show-ing a pass, the Nuntii were able to exchangehorses or use a bed for the night. In this way,

Autumn 1997

the efficient—though not always secure—flow of information across thousands ofmiles was ensured through a state-of-the-artnetwork that survived until the fifth century.

More than a thousand years later in 1860,the speed and bandwidth of informationtechnology were virtually unchanged, andechoing the Romans, the Pony Expressestablished its own smaller version of theCursus Publicus in the USA—this time on acommercial basis.2 But by then, technologydevelopment was beginning to accelerateand just eighteen months later the firsttranscontinental telegraph line precipitatedthe end of the Pony Express.

Since then, speed and bandwidth havecontinued to increase and in the last fewyears the rate of increase has accelerateddramatically. Today, electronic informationtechnology (IT) is delivering data and infor-mation to individuals and corporations at arate and volume never before achieved. Thisarticle reviews some of the contemporarychallenges and opportunities presented bymodern IT and also looks ahead at howknowledge management is becoming anincreasing concern within the oil industry.

3

■From data to information toknowledge. Datamust be integratedto change them intoinformation andthen organizedusing experience tocreate knowledge.

Data

Information

Knowledge

Simple observations or measurements

Data endowed with relevance andpurpose by humans

The most valuable information is given context, meaning or a particularinterpretation that includes humanwisdom, reflection and synthesis

Easy to structure

Easy to capture on machines

Easy to transfer

Often quantified

Difficult to structure

Difficult to capture on machines

Difficult to transfer

Often tacit

Requires unit of analysis

Needs a consensus meaning or definition

Requires human mediation

■Contemporary network. The Schlumberger intranet currentlyserves about 25,000 users in more than 65 countries.

First, it is important to understand howdata, information and knowledge arerelated. Data may be defined as raw facts,information as data endowed with rele-vance and purpose, and knowledge asinformation enhanced by context, meaningand interpretation.3

For example, the data in a timetabledescribing a train departure might read“Camb-07.39-2.” Information consists of dis-crete pieces of data that are ordered andorganized by the mind into various patterns.Therefore, the information described bythese data is that the Cambridge train departsfrom Platform Two at 7:39 in the morning.

If someone then told you that: “The 7:39always leaves five minutes late because ithas to wait for a connection with anothertrain,” he or she is imparting their knowl-edge to you—although, if you were notinterested in the train, this constitutes noiserather than knowledge. The problem then iswhat value to put on this knowledge. If the

4 Oilfield Review

advice came from the manager of the trainstation, the knowledge is more likely to becredible than if it came from a stranger youhad just met in the station. Unless of coursethe stranger turned out to be the driver ofthe train.

In short, data alone cannot tell you whatyou need to know. And even after the datahave been transformed into information, thisin turn must be organized and integratedusing experience, memory and insight tocreate knowledge (above left).

Since its genesis, the oil industry has beenone of the most geographically dispersedand multicultural businesses. As such, thedemands made on those charged with find-ing and delivering hydrocarbons havealways been high. For example, when the

pioneering oilfield engineers of the 1930swaded through the swamps of Venezuela todiscover and develop resources there, theyhad no practical means to rapidly commu-nicate with their headquarters or technicalcenters. For all intents and purposes, theywere the sole repositories of their branch ofknowledge on that rig at that time.

Although communications have improveddramatically, this effectively remained thesituation until relatively recently. It was onlythen that new ways of using expertise—within a wider organization and beyond theconfines of the rig—started to be deployed.Yet even today, it is virtually impossible todeliver the full scope of corporate know-how to every corner of industry activity. Butit is this objective that is motivating opera-tors and contractors alike. The goals are toimprove decision making, raise servicequality and cut operational costs by apply-ing the full weight of corporate knowledgeeverywhere all the time.

Using an organization’s knowledge andstored information to the fullest possibleextent has some prerequisites. First, thetechnology needs to be in place—namely, asecure communications infrastructure, thenecessary repositories of data and informa-tion and the software to extract the desiredinformation. However, knowledge startsand ends with people; putting knowledgeto work across an organization requiresnew processes and individuals who are pre-disposed to communicate their knowledgeto others.

Disasters 13%

Insider misuses 19%

Outsider attacks 3%

■Information security—a people issue.Only about 3% of breaches in informationsecurity come through attacks from theoutside. Nevertheless, technology and anincident response team must be in place torespond to such breaches and any disas-ters that may occur.

Creating a Secure InfrastructureDevelopments over the past decade or sohave put in place an infrastructure—a seriesof connected networks—that can reach intoalmost every corner of the world. In mostcases, only political problems prevent fullcoverage (see “The Links Behind the Desk-top,” next page).

The growth in bandwidth and speed ofthese networks is creating new ways of doing business. For example, theSchlumberger intranet—SINet—first offeredthe transmission of basic e-mail over a directlink in 1985. Today, the network may beused to deliver a 25 million trace, 3D seis-mic survey in a few working days, helping tosubstantially reduce the turnaround time forprocessing seismic data (previous page, top).

There are of course key differencesbetween the networks of today and a sim-plistic system of messengers such as theCursus Publicus. These differences concernnot only the dramatically higher bandwidthand speed, but also accessibility. Networkssuch as SINet deal not just in traffic to andfrom the center; everyone on the networkcan access everyone else. Achieving thisuniversality and developing communica-tions networks that can meet expandingneeds required the establishment of stan-dards that allow interoperability (below).

Simply stated, interoperability is the abilityof different systems, products and servicesto work together. Therefore, no matter whatthe computing and communication plat-form, SINet offers equal accessibility. Thekey to interoperability is the developmentand implementation of open interfaces.

Autumn 1997

With different Schlumbergeroperations coveringalmost theent i reworld,needstended tovary. Theinteroper-ability wasachieved, andis maintained,through a col-laborative envi-ronment via aforum that can beused to receive andalso propagate opin-ions and ideas.

A similar approach of col-laboration and discussionaround the world was neededto ensure that broad standardshave been adopted by the net-work community as a whole.Through the work of the InternetEngineering Task Force, the Internet hasbeen able to grow far beyond its originallylimited bounds.

Emperor Septimius Severus made greatefforts to keep secret the information hetransmitted. Similar concerns face many oftoday’s network users. Indeed, one of thestrongest barriers to increased use of IT issecurity concerns. In Roman times, securitywas provided by a unique seal on the pack-

Bugs,mistakes, negligence65%

■New technolo-gies bring newchallenges. In 19th centuryEurope, the prob-lem of interoper-ability quicklytook root wherecountries andcompanies chosedifferent gaugesfor their railways,preventing easyconnection ofrival networks.The issue of inter-operability is onethat still concernsIT managersaround the world.

age being transmitted. For this seal to bebroken required the negligence or collabo-ration of the messenger.

The picture today remains fundamentallythe same. Industry statistics reveal thatonly a small percentage of all securityincidents are due to “outsider attacks.” Inmost cases, security breaches are due toignorance—like misuse of technology—oremployee negligence (above).

For this reason, corporate security policiesusually start with the work force and extendthrough the establishment of internal secu-rity policies and audit procedures. Theseencompass access, authentication and dataintegrity procedures (see “Security Policiesin Focus,” page 9).

However, securing an open networkingarchitecture does pose new technical chal-lenges. There are two major concerns. Thefirst is the protection of the interfacebetween a closed private network—intranet—and the open Internet, which is

5

3. Davenport TH: Information Ecology—Mastering theInformation Knowledge Environment. New York, New York, USA: Oxford University Press, 1997.

(continued on page 8)

The Links Behind the Desktop

For most of us, a network is visible only through

the applications it supports—e-mail, file transfer,

shareware and more sophisticated links. Behind

the screen on the desktop there is wiring or fiber-

optic cables, satellite facilities, switches, routers

and hubs, together with specialized computers

hidden from view. Binding all this together are

rules and protocols.1

When two or more computers are linked

together they create a network. Typically, when the

computers are close together they form a local-

area network (LAN). Devices may be connected by

twisted-pair wire, coaxial cables, or fiber-optic

cables. In general, LANs are confined to a single

building or group of buildings. LANs are capable of

transmitting data at very fast rates, much faster

than data may be transmitted over a telephone

line; but the distances are limited. However, one

LAN may be connected to other LANs over any dis-

tance by telephone lines or radio waves, creating

a wide-area network (WAN).

Networks are like the plumbing system of a city.

The LANs are equivalent to the plumbing system in

the house. The WAN is equivalent to the main

water pipes and pumping stations linking houses

together. Expansion and redevelopment constantly

change the requirements of the system.

Most LANs connect workstations and personal

computers. Each computer—or node—in a LAN

has its own central processing unit with which to

execute programs. However, it is also able to

access data and devices anywhere else on the

LAN. This means that many users may share data

and devices, such as laser printers. Users may also

communicate across LANs and WANs using e-mail.

6

The computers communicate with each other

according to sets of rules—protocols—that may be

implemented either in hardware or in software.

These ensure that the whole network responds uni-

formly, such as using the same type of error check-

ing and data compression, and recognizing that the

sending device has finished sending a message

and the receiving device has received a message.

There are a variety of standard protocols which

platforms and networks must support. For exam-

ple, token-ring networks, ethernets and ARCnets

are the most common for PCs. Most Apple net-

works are based on the AppleTalk network system,

which is built into Macintosh computers. Each has

particular advantages and disadvantages. From a

user’s point of view, the only interesting aspect

about protocols is that your computer or device

must support the right ones if you want to commu-

nicate with other computers.

The way that a network is configured is called

its architecture. In a client-server architecture,

each computer or process on the network is usu-

ally either a client or a server. Clients are less

powerful PCs or workstations on which users run

applications. Clients rely on servers for resources,

such as files, devices and even processing power.

Servers are powerful computers or processors

that are often dedicated, meaning that they perform

no other tasks besides their server function. A file

server is a computer and storage device that stores

files for any user on the network; a print server

manages one or more printers; a network server

manages network traffic; and a database server

processes database queries. On multiprocessing

operating systems, however, a single computer

may execute several programs at once. A server in

this case could refer to the program that is manag-

ing resources rather than the entire computer.

Another type of network architecture is known as

a peer-to-peer architecture because each node has

equivalent responsibilities, acting as a server and

client—for example, AppleShare or Network

Neighborhood. Both client-server and peer-to-peer

architectures are widely used.

From Internet to IntranetCompany LANs and WANs are increasingly con-

nected via routers to the outside world in the

shape of the much talked about Internet. The Inter-

net provides individuals with many different ways

to disseminate and retrieve information. The Inter-

net’s underlying communications concept is to

connect networks of information together so that a

user with his or her personal computer or worksta-

tion can connect, either directly or through a suc-

cession of intermediary computers, to a remote

computer that acts as a server of information. This

connection permits the flow of data and informa-

tion, typically at the request of the user.2

The Internet may best be pictured as a system

that transports discrete packets of information. The

Internet protocol (IP) takes care of ensuring that

the routers know where to send the data being

transported in the system—each location carries

its own IP number. For practical reasons, informa-

tion sent across IP networks is broken up into man-

ageable packets that may not be transported

through the network in one block. This prevents any

one user from monopolizing the system. And at any

one time, all sorts of different packets from many

different sources are moving through the system.

Oilfield Review

1. A comprehensive set of definitions and explanations maybe found at the PC Webopædia: http://www.pcwebopae-dia.com/index.html.

2. A more detailed discussion of the Internet’s genesis andsome of the protocols that control it may be found at:http://www.ladas.com/NII/NII_Technology.html.

3. In fact such links already exist. For example, to aid in the development of new production logging technology, cooperation between researchers in Schlumberger Cambridge Research, Cambridge, England and BP Exploration in Sunbury on Thames, England, was greatlyaided through joint project Web pages. These were accessible through both company intranets, yet access was limited to those working on the project.

4. Clark R, Danti B, Guthery S, Jurgensen T, Kennedy K, Keddie J and Sims D: “Building a Global Highway for Oilfield Data,” Oilfield Review 5, no. 4 (October 1993): 23-35.

Once the packets reach their IP destination, they

need to be reconstructed in the right order to form

the original file. That is why the transmission con-

trol protocol (TCP) is employed. TCP is used to

break the initial file into pieces that are usually no

longer than 1500 characters long. It numbers each

piece, so the receipt of the full document may be

authenticated, and correct reconstruction may be

carried out.

Many established protocols have been inte-

grated and enhanced using tools—such as Web

browsers—that access the WorldWide Web. The

WorldWide Web allows organizations to use

graphical “front ends” to provide remote users

with point-and-click access to information stored

on their servers, as well as access through links to

information stored on other remote servers. Web

browsers are programs that run on a personal

computer or workstations that enable a user to

establish connections to these graphical front

ends, and to view, retrieve and manipulate data

provided by those remote servers.

At the root of the Web are three protocols: the

hypertext markup language (HTML), a file format

for embedding navigational information in graphi-

cal and text-based documents; the hypertext

transfer protocol (HTTP), a communications proto-

col for communicating navigational information

and other data between the remote server and the

requesting computer; and the uniform resource

locator (URL) scheme for identifying locations of

Web-accessible documents.

A Web site or group of Web sites belonging to

an organization and accessible only by the organi-

zation’s members, employees or others with

authorization constitute an intranet. An intranet’s

Web sites look and act just like any other Web

Autumn 1997

sites, but the firewall surrounding an intranet

fends off unauthorized access (for more informa-

tion about network security, see “Security Policies

in Focus,” page 9).

The next step is to link some intranets

together, so that companies may communicate

more efficiently with their partners, suppliers,

customers and contractors without having to go

through the Internet. We will then have reached

the age of the extranet.3

The Schlumberger IntranetIn 1984, long before the term “intranet” had been

coined, Schlumberger connected research labs in

France, Japan, the USA and England through a

series of proprietary networks.

But by the late 1980s, use of multiple propri-

etary networks proved unsatisfactory. In 1991,

Schlumberger decided to chart an aggressive

Internet-based strategy using the TCP/IP protocol

for its international network.4 Between 1989

and 1996, Schlumberger invested just under

$100 million in the Schlumberger intranet—

SINet—which is now one of the world’s largest

private networks and is the backbone of the com-

pany’s business infrastructure.

Today, SINet serves more than 25,000 users at

450 locations in more than 65 countries. This adds

up to a formidable collection of hardware with over

500 routers plus 150 access servers, 1750 ether-

net hubs, 2000 modems, and more than 25,000

desktops systems of all types. Satellite technology

provides access from remote sites, such as

those in West Africa, Jakarta and Balikpapan in

Indonesia, and elsewhere. Depending on the loca-

tion, this gives a network bandwidth ranging from

4.8 kBps to 45MBps.

In 1995, the joint venture Omnes was formed

by Schlumberger Limited and Cable and Wireless

plc. Omnes acquired SINet, which it now adminis-

ters. Since its formation, Omnes has served the

communications and IT needs of more than 100

customers based in over 30 countries. In 1996,

the Omnes quality management system at its

Service Management Center (SMC), Houston,

Texas, USA, was granted ISO 9002 certification

by Lloyd’s Register Quality Assurance. The SMC

provides global telecommunications customer

support including helpdesk, network management

and fault management services.

7

Encrypt Decrypt

Cipher textPlain text Plain text

■Encryption usinga shared key.

the main gateway to the outside world. Sec-ond, if partners and suppliers choose toimprove communications by linking theirintranets—creating so-called extranets—each link will have to be protected andsecured. Thus, the need for technical solu-tions such as firewalls, encryption, andauthentication is increasing.

Firewalls—Designed to block the progress ofmost kinds of unauthorized access to orfrom a private network, firewalls police theintranet-Internet doorway. All traffic enteringor leaving an intranet that passes throughthe firewall is examined. Traffic that doesnot meet specific security criteria is rejected.There are several techniques: • A packet filter looks at each packet of data

or information entering or leaving the net-work and accepts or rejects it based onnetwork administrator-defined rules

• An application gateway applies securitymechanisms to specific applications, suchas those used to transfer files

• A proxy server intercepts all messagesentering and leaving the network. Theproxy server effectively hides the true net-work addresses.

8

Encrypt

Cipher text

Jane's public k

John

Plain text

■Asymmetric encryption using public and pri

In practice, many firewalls use two ormore of these techniques in concert. Forgreater security, data may be encrypted.

Encryption—In its most basic form, encryp-tion amounts to the scrambling of data usinga mathematical program that may bereversed to unscramble data to their usableformat. Encryption and decryption areenabled by the possesion of the appropriatekey. The keys that control access to the dataare actually strings of alphanumeric digitsthat are plugged into the mathematical algo-rithm that scrambles the data. Anyone withthe key can decrypt the data to yield theoriginal sequence of binary digits that com-prise the file (above).

There are two main types of encryption:symmetric and asymmetric—often calledpublic-key. Symmetric-key systems use asingle key that both the sender and recipienthold. However, key management is easier inpublic-key—asymmetric—systems, which iswhy most messaging systems such as PGPPretty Good Privacy and S/MIME SecureMIME use public-key algorithms to dis-tribute keys through the Internet. The public

Decrypt

ey

Jane's private key

Jane

Plain text

vate keys.

key is disseminated to anyone from whomyou wish to receive a message. The ownerof the private key keeps it secret and uses itto decrypt the messages sent to him andencrypt messages he sends (below).

When John wants to send a secure mes-sage to Jane, he uses his key and Jane’s pub-lic key to encrypt the message. Jane thenuses her private key to decrypt it.4 In addi-tion, Jane is certain that it was John whosent the message since it required John’s keyfor successful encryption and decryption.

Although the public and private keys arerelated, it is computationally impractical todeduce the private key even if the publickey is known. The only difficulty with pub-lic-key systems—as with symmetric key sys-tems—is that message senders need therecipient’s public key to encrypt a messageto him or her, and distribution and mainte-nance of public keys require some kind ofrepository. This is important because Johnneeds to be sure that he has Jane’s publickey when sending a message. For example,if Kate were to publish her key under Jane’sname and John used it to send a message,Jane would be unable to decrypt it and Katemay even be able to read the message—thisis a type of denial-of-service attack. How-ever, to avoid this, digital certificates can beused to bind Jane’s name to her public key.

Today, the private key may be stored on ahard or floppy disk which in turn mightneed a personal identity number (PIN) toopen. However, an increasingly popularoption is for the encryption algorithms to beheld on the chip of a smart card that is thenused in a card reader attached to the com-puter. A smart card also requires a PIN toactivate it. Storing the algorithms on a cardis desirable because the secret key neednever leave the card, avoiding the possibilityof a rogue terminal storing users’ keys.

Authentication—When mathematical algo-rithms are used to combine information froma data object, such as a word-processing doc-ument or image, with a user’s private key, adigital signature is created. The signature canbe created only by the holder of the privatekey. Anyone with the appropriate public keycan then verify that the object was signed asclaimed. Verifying the signature on a dataobject is one example of authentication.

Oilfield Review

4. A description of cryptography may be found at:http://www.sandybay.com/pc-web/public_key_cryp-tography.htm.

Security Policies in Focus

Information security is not a technology. Rather it

is a discipline, an attitude, on the part of every-

one—essentially it requires the same disciplines

as quality, health, safety and environment

(QHSE). Also, being “secure” is not something

that can be taken for granted for any period of

time, as threats shift continually. People are at

the center of any effort to achieve and perpetuate

information security.

Information is the foundation of most aspects of

business carried out by Schlumberger companies.

As such, all employees and subcontractors have

ethical and legal obligations to protect proprietary

information owned by or under the custody of

Schlumberger, and to maintain the confidentiality

of this information.

To help achieve this, the priority is on building

awareness in and offering training to employees.

High-risk behavior is identified and simple and

practical recommendations are made available to

all employees. This is then supported by on-line

training material. Site security officers ensure that

guidelines are followed and that information secu-

rity is assured.

Information security within Schlumberger is

monitored by the company’s QHSE organization,

rather than its IT or financial departments as is

the case in many other large organizations. QHSE

is responsible for raising awareness, and for

training, monitoring and reporting information

security issues because it already has established

processes and a network of people in place to cre-

ate the required continuous improvement and

safety cultures. A scale of the severity of informa-

tion security incidents has been developed

according to QHSE guidelines based on the classi-

fication of information, the financial impact and

Autumn 1997

the time to undo the damage. An incident

response team is always available to investigate

serious problems and provide the technical exper-

tise to find appropriate solutions.

An information classification standard is a

foundation for information security because it

makes clear what needs protection and how to

protect it. Information classification also reduces

the cost of security by establishing controls that

are commensurate with the value of the informa-

tion being protected.

Information security incidents are defined as

loss of control over information assets—such as

hardcopy documents, electronic files, equipment,

passwords and decryption keys. Incidents are pre-

vented by defining the level of control appropriate

to each type of information and then deploying the

necessary mechanisms to implement that level of

control. The Schlumberger information classifica-

tion standards establish three categories of sensi-

tive information—secret, confidential and private.

Secret information provides the organization

with a significant competitive edge, shows specific

business strategies or organizational directions, or

is essential to the technical or financial success of

a product or service. Unauthorized disclosure

would cause serious damage to the interests or

reputation of the company. Examples of secret

information include that which could affect the

share price, relates to significant acquisitions and

divestments or details R&D information about new

tools prior to testing the engineering prototype. All

client information and data in Schlumberger cus-

tody are classified as secret.

Strict control mechanisms—procedures and

tools—are established for protecting information

classified as secret. For example, it may only be

located on the intranet if protected by TWS Trusted

Web Server technology; if it is shipped on CD-ROM

through insecure channels, it must be encrypted

with a strong encryption method; when the media

are disposed of, a secure destruction method must

be used.

Confidential information would be prejudicial to

the interests of the company, or would cause

embarrassment or difficulty for the company or its

employees if disclosed. For example, such infor-

mation includes confidential agreements and con-

tracts, tool maintenance manuals, engineering

files or personnel documents.

Appropriate control mechanisms must also be

applied when the information is created, modified,

stored, communicated or disposed of. These con-

trols are less strict than those applied to informa-

tion classified as secret. For example, encryption

of the information during transmission on the

intranet may be optional, while in the case of

secret information it is mandatory. Even so, TWS

Trusted Web Server protection for Web-based infor-

mation and strong encryption are recommended

when communicating it through insecure media

such as CD-ROMs.

Private information is available to company

employees only as part of routine business. For

example, it includes intracompany e-mail that

does not contain confidential or secret informa-

tion, and equipment catalogs. Private information

is protected by using less stringent security mech-

anisms. For example, weak encryption is used

instead of strong encryption, and for information

posted on the intranet, firewalls are considered to

offer sufficient protection.

9

1

geaTttf

bmnthtodvIdse

nsWsiipoare

noBsnps

cAhTvimcmdpomba

Geco-Prakla vessel or land crew

Data access via web serverOptional:

Data analysis via Geco-Prakla

or client software

Acquisition Secure web server Client office

■Desktop access with state-of-the-art security. The Geco-Prakla SuperVision serviceenables clients to monitor the progress of acquisition and processing projects while pre-venting unauthorized access to the information.

Generating a digital signature uses crypto-raphic techniques, but does not necessarilyncrypt the work; the work may still beccessed and used without decryption.hus, digital signatures identify the origina-or of a particular file, and also verify thathe contents of the file have not been alteredrom what was originally distributed.“Digital watermarking” methods have alsoeen developed that encode digitized infor-ation with attributes. These attributes can-ot easily be detected or disassociated fromhe file containing that information. Thus,idden messages may be embedded in digi-ized visual or audio data, making breachesf copyright easier to detect. This watermarkoes not degrade, yet does not affect theisual or audible quality of the work.

nstead, the embedded information can beetected only if specifically sought out byomeone who knows what to look for—forxample, the copyright owner.There is, therefore, a wide range of tech-iques available to an IT solutions provider—uch as Omnes, the Schlumberger/Cable and

ireless joint venture—to deliver a range ofecure services. For example, Omnes offersts INCA Intranet Node Corporate Accessntegrated Internet and intranet solutionslatform to distribute the Internet through anrganization while ensuring security through fully managed firewall capability. Omnesecently implemented this for the Qatar Gen-ral Petroleum Corp.Firewalls may be likened to a hard exter-al shell of an egg; once broken, the wholef the egg inside may be accessed.ecause firewalls may in some circum-tances be bypassed or breached, compa-ies need additional defenses for thosearts of their intranets that contain the mostensitive information.Combining Web-based accessibility withase-hardened security, Schlumbergerustin Product Center, Austin, Texas, USA,as developed with Omnes the TWSrusted Web Service secure documentault. For this, a Netscape Enterprise servers equipped with state-of-the-art securityeasures—firewalls, personal digital certifi-

ates and tools to authenticate, encrypt andonitor all traffic. The server is also on a

edicated machine that constantly monitorsossible intrusions and checks the integrityf its configuration. This opens the way forore effective on-line communications

etween partners, or between contractorsnd operators (above).

0

For example, by using TWS technology,Geco-Prakla has launched its SuperVisionservice. This enables an oil company tomonitor the progress of seismic acquisitionand data processing projects. Authorizedpersonnel are able to access a site dedi-cated to a given project. They may acquirehourly, daily or monthly updated informa-tion on the progress of acquisition and pro-cessing. Images and quality control plotswill be available, as will daily productionreports, spreadsheets and correspondence.Over time, the updated information createsa project archive.

Digital certificates are already in use foraccess to TWS protected sites. Currentlythese are loaded onto a computer’s harddisk, and cannot be tampered with or transferred to another machine.Schlumberger is experimenting with theuse of smart cards as physical carriers ofpersonal digital certificates.

Using Infrastructure: Building Better TeamsSince the introduction of photocopiers,nothing has influenced the way that infor-mation flows through an organization morethan computer networks. The simplest bene-fit comes from the replacement of phone,fax and telex messages with electronic mailon private and public networks. However,more sophisticated changes are possibleand are being harnessed to improve the waythat information and knowledge are propa-gated and accumulated. This section looksat how improving communications tech-niques has yielded exciting results.

When BP Exploration Operating CompanyLimited was reorganized into 42 separatebusiness assets in 1994, a “federation ofassets” was created. Each asset was giventhe freedom to develop processes and solu-tions to meet local needs.

Good communications were clearly essen-tial to making this federation work. Head-quarters needed a means of coordinationand the business units needed ways of com-municating the creative results of their rela-tive independence. One potential solutionwas the installation of video conferencingcapability across the organization.

At the outset, it was decided that the pro-ject be led not by a group of IT experts, butby a special core team. There were severalreasons for this. One was a determinationthat the initiative should not be seen as an

Oilfield Review

■The BP Andrew platform brought into pro-duction six months early, in June 1996. Theproject of developing the Andrew field inthe UK North Sea is a good example of thepositive effects of virtual teamworking (VT).The goal of this team was to complete theconstruction of the new platform on timeand VT proved to be one of the major inno-vations on the project. (Courtesy of BritishPetroleum)

IT initiative, but as a business initiative. Asecond was that the emphasis was being puton the need to change work behavior, ratherthan technology. Finally, the aim was to usevirtual teamworking (VT) to cross organiza-tional boundaries. It was decided that theseobjectives would best be served by a groupdrawn from different parts of the company.The team of five people began work inDecember 1994. From the start, the focuswas on connecting people with comple-mentary expertise.

In 1996, building on the success of the VTpilot, BP Exploration took interoperability astep further than many companies, stan-dardizing the desktop and server environ-ment across the organization. This meantthe choice of just two types of PC—desktopor laptop. Each was supplied with core soft-ware plus additional applications from astandard suite. The infrastructure was alsostandardized through common file serversand network protocols. In just six months,the fully-supported changeover for 7500users in 27 locations was achieved.

Moving VT from pilot to operational statusmeant that 500 of these PCs were fitted witha package that enabled use of desktop videoconferencing, multimedia e-mail, applica-

Autumn 1997

5. Integrated services digital network (ISDN) is an

tion-sharing shared chalkboards, videorecording, groupware exchanges, Webbrowsing and document scanning.

Connections were made using ISDN overstandard telephone lines and, when neces-sary, through satellite links.5 The first unitswere put in place in August 1995. A key ele-ment of the implementation revolvedaround coaching. A subgroup of the coreteam—the coaching team—took on the roleof showing participants how to use the tech-nology. More importantly, the coachesdemonstrated how teams could use thetechnology to improve their work processesand enhance personal interaction. Of thefive pilots chosen for the first wave, the pro-ject of developing the Andrew field in theUK North Sea proved the worth of VT.

The goal of the Andrew team was to com-plete the construction of the new platformon time. The job of building the platform fellto an alliance of BP Exploration and severalother companies, including Brown & RootLimited, a design and engineering firm withoffices in Colliers Wood, near London, Eng-land; and Trafalgar John Brown Oil & GasLimited (now a member of the KvaernerGroup), a construction firm based inTeesside, northeast England. So the trial wasnot just one of linking geographically sepa-rate operations; it also linked different orga-nizations. In the end, the project team suc-cessfully rolled out the Andrew platform inrecord time (above). There are many rea-sons for this success, but here is how PhilForth, an independent consultant from York,England, who worked with the Andrewteam to help them use VT effectively, pin-points the advantages of VT.

Errors are removed from the conversation:“I have come to understand that conversa-tion is rich in visual clues and is poor inaural clues. I have listened to phone conver-sations whose main achievement was irrita-tion and confusion for both parties. I havealso been involved in video conferencesthat have used the visual clues to steerthrough misunderstandings and move theproject forward.”

Individuals establish their integrity: “It is fareasier to establish credibility when there is aphysical presence. We now have someexcellent examples of how video conferenc-ing has lead to one individual giving respon-sibility to another because of the rapportthat has been established through VT.”

Time to completion is dramatically reduced:“VT has allowed an action to happen atonce. Momentum, a precious commodity ina large organization, is maintained.”

VT gets the right information to the rightpeople at the right time, enabling them totake the right action: “We are seeingstartling examples of how the right thingshappen because VT gets people the infor-mation they need.”

VT stops paper shuffling and ties thingsdown: ”The potential for VT to remove themass of paper that crawls around an organi-zation is becoming increasingly apparent.”

This latter example includes the effectiveuse of groupware such as Lotus Notes. Inthe case of Andrew, Lotus Notes softwarewas used only towards the end of the pro-ject in the completion phase. However, thepotential to decrease paperwork may beappreciated by looking at the design changenotices (DCN) that were normally raised inthe engineering department in ColliersWood and construction change notices(CCN) raised in Teesside.

In total, the project generated about 650DCNs and 1000 CCNs. Each CCN was cir-culated to four or five disciplines and couldlead to changes in perhaps 30 drawings thatwould be attached to the CCNs for circula-tion. Each CCN would then be forwarded tosome six people for comment or for them toaccept. The end result was described as a“flood of paper” that was carried by courierdaily between Colliers Wood and Teesside.Most of this could have been avoidedthrough the effective use of groupware.

Using Infrastructure: Improving Service Quality The advent of effective networks eases theaccess of employees to stored information.Daily forms and reports that used to beneeded may be consolidated and their flow through an organization optimized.Furthermore, electronic commerce andimproved management systems will simplifythe interfaces between service companiesand their clients.

Put simply, the daily clutter of paperworkthat obstructs the delivery of services maybe removed. Thus, people may concentrateon productive tasks while being able tomore clearly identify past mistakes in order

11

international communications standard for sendingvoice, video and data over digital telephone lines. An ISDN line supports data transfer rates of 64 kBps(64,000 bits per second).

Process

Clutter

Tim

e, d

ays

1996

0

20

40

60

80

100

Today

PrivateLAN/WAN

PrivateLAN/WAN

Human interface

PTSN

Client PC

Office

Wellsite MAXIS unit

Well

Dial-upaccess

Directnetwork

Publicphone lines

■Removing clutterand improvingresponse to fieldrequests for support.By initiating anelectronic system toreport the need formodifications toequipment,Schlumberger Wireline & Testinghas reduced thepaper chase andensured that theresponse time isnow 15 days ratherthan 100.

■Bringing the wellsite tothe client office. By usingits IT infrastructure,Schlumberger Wireline &Testing can deliver data toa client’s desktop whilelogs are being acquired.

6. Edmonds P: “Linking Solutions to Problems,” Oilfield Review 8, no. 4 (Winter 1996): 4-17.

to avoid them in the future. These two fac-tors alone are significant steps towardimproving service quality (right).

Similarly, electronic documentation maybe used to reduce office clutter and thelength of company bookshelves. Firstattempts to transfer documents such as man-uals to an electronic format centered on sim-ply mimicking the paper format: ElectronicDocument Management meant creatingelectronic books on CD-ROM or the Web.

Oilfield service companies are increas-ingly turning to electronic documentationfor manuals and service support literature.The technology offers scope to produce sig-nificantly more than simple scanned manu-als. The documents themselves may be mul-tilayered. The links between the document’sproducer, shipper, reader and maintainermay be two-way, allowing for rapid modifi-cation and redistribution.

A key element in effective service deliverywithin a fast-changing technological envi-ronment is the training and development ofemployees. Here too, companies areincreasingly turning to network-deliveredtraining programs that incorporate a grow-ing number of elements of interactionbetween the learner and the lesson (see“On-Line Electronic Training,” next page).

While IT offers service providers opportu-nities to optimize their internal processesand to better train their engineers, networkconnectivity also allows service deliveryitself to be reengineered. For example, theSchlumberger ClientLink initiative allowsclient needs and problems to be better gath-ered, propagated and ultimately dealt with.6

Another example of this is the real-timetransfer of data from rigs to oil and gascompany offices. The Schlumberger Wire-line & Testing InterACT program, forinstance, enables office-based customers tofully engage with the wellsite logging engi-neer, ensuring optimal use of a logging sur-vey operation. The InterACT service offersreal-time transmission of log files, two-waycommunication and data compression.Optical and digital data are conveyed inreal time from the MAXIS Multitask Acqui-sition and Imaging System logging unit sothat they may be viewed and stored onclient desktop computers. In this way, rig-based decisions may be made faster and infull cooperation with the company expertswho do not need to be on the rig during the

12

job. Supervisors can remotely oversee well-site operations, and geoscientific expertscan respond in a coordinated manner dur-ing logging operations.

At the core of the InterACT service is a PC-based file transfer system (FTS) developed atthe Schlumberger Austin Product Center. TheFTS uses standard transmission control pro-tocols (TCP) and Internet protocols (IP) witha read-while-write capability, adaptablecompression and quick recovery (see “FromInternet to Intranet,” page 6). Data compres-sion minimizes the bandwidth allocation,allowing use of conventional phone lines.During field tests, networks with throughputsas low as 10 kBps were used to transmit awide variety of imaging logs.

The transmit-while-acquire function makesit possible to transfer very large files duringthe several hours typically involved in a data

acquisition program. In certain situations,these files could be transferred to a process-ing center before the logging tools havebeen fully rigged down. The largest suchoperation to date in the North Sea involvedtransmission of a file of 320 MB in 11 hoursover a nominal 64 kBps link.

The fully automated system is designed sothat neither the logging engineer or theoffice-based client face technical barriers toits use, nor is anyone distracted from theoperation by it (above).

Oilfield Review

On-Line Electronic Training

It should go without saying that to offer the highest

possible standard of service requires training. In

the past, this has often meant periodic visits to

training centers to upgrade skills. However, the

pattern of the oil business has evolved. The need

for greater flexibility and faster training points to

more self-training. Today using advanced simula-

tion and other IT methods, the culture is changing

from “spoon-fed” teaching to proactive learning

while on the job.

For example, Schlumberger Wireline & Testing

has developed a new engineer training program—

the PEPTEC program—to replace one that had

been in use for over twenty years. This new pro-

gram places greater responsibility on individuals

to ensure that their training assignments are com-

pleted. Further, there is also a greater responsibil-

ity for field management to assist trainees with

their assignments. However, the new methodology

as well as the new training aids make this much

easier to achieve than in the past.

The program uses simulators and interactive

technology to help field engineers better under-

stand the basics. Training is now more focused on

each discipline. Yet there is also formal instruction

in “soft skills” such as communication.

The evolution of IT allows much greater access

to contemporary technical information. The advan-

tage of electronic documentation is that it is eas-

ier to maintain and update. Connectivity around

the world often allows training programs to be

accessible via intranets. In areas where connec-

tivity is poor or nonexistent, programs are con-

veyed on CD-ROMs.

Autumn 1997

Whether on-line or on CD-ROM, the training

modules take advantage of being in an electronic

medium. Many of the training modules now

involve interactive multimedia sections—each

section lasts about three minutes. At the end of

each section of training, self-assessment is used

to chart progress.

Simulators have also significantly changed the

training process to enhance logging or testing

training. Through simulation, new engineers may

experience logging and testing conditions that are

more realistic than can be achieved using 800-ft

[244-m] plastic-cased training wells. Today, many

new tools are developed using simulation code

that will be easily adaptable for training purposes.

Field engineers also use simulations of drill-

stem or surface testing to practice their skills. The

surface testing module is designed to operate

either alone or in conjunction with surface testing

hardware for both single-well and multiple-well

models. There is also a drillstem testing simulator

that comes complete with drawworks.

To ensure improved quality control, a data bank

of good and bad log examples has been compiled

as interactive multimedia files to help emphasize

the differences between good and bad logs. This

will enhance familiarity with basic logs in common

types of fields around the world and help in under-

standing some basic interpretation concepts.

Quality control and interpretation basics will be

emphasized and taught hands-on.

The PEPTEC program is not unique in

Schlumberger. Anadrill and Dowell, for example,

are both developing IT-based learning organiza-

tions. In both cases, the aims are similar—to

encourage involvement and communication

between field engineers and their supervisors,

organizations and clients. Training may be tar-

geted both to individual and client requirements,

answering specific needs of the field. At the same

time, an engineer acquires the necessary skills as

part of a continuous development program.

To ensure this significant corporate investment

remains “evergreen,” a group of IT specialists and

instructional designers located in Austin are inves-

tigating technologies and standards that will

enhance the integration of training, ease mainte-

nance, and improve relevance to the end-user.

One key to success is ensuring the longevity and

compatibility of training material through use of a

vendor-independent data format—such as the

Standard Generalized Markup Language (SGML)—

instead of proprietary data formats.

Material authored using SGML is not limited to

a single publishing medium and may be dis-

tributed on the Web, CD-ROM or paper, if neces-

sary, from a single source. Database publishing,

as this is called, also allows for information to be

reused across collections. Thus, a technical pro-

cedure in a reference manual can be reused in a

training manual. This greatly reduces the cost of

maintenance and reduces the chance of errors

because only one version of the information need

be held, instead of multiple copies pasted into

many different locations.

13

14

7. Beham R, Brown A, Mottershead C, Whitgift J, Cross J,Desroches L, Espeland J, Greenberg M, Haines P,Landgren K, Layrisse I, Lugo J, Moreán O, O’Neill Dand Sledz J: “Changing the Shape of E&P Data Man-agement,” Oilfield Review 9, no. 2 (Summer 1997):21-33.

8. One search engine to try is the FAQ Finder at the University of Chicago. This is a WordNet-based appli-cation—a different approach to natural langauge pro-cessing: http://faqfinder.cs.uchicago.edu:8001/.

9. NOEMIE is a project in the European Union’s Espritprogram, where the partners are, from France:Schlumberger Limited, Matra Cap Systémes, Univer-sité Dauphine and Acknosoft; from Norway: SINTEFand Norsk Hydro; and from Italy: JRC Ispra.

Data analysis

"What-if" analysis

QA controlled database

Operations

Decision supportinformation

Risk analysisand simulation

Probabilitydistributions

Field data

X X+6

100

0

50

100

40

60

80

X+12 X+18

X X+6 X+12 X+18

HH

P u

tiliz

atio

n, %

TCV

util

izat

ion,

%

Total pumpers

Total pumpers

Average TCVutilization

Average pumperutilization

■Analysis of hydraulic horsepower (HHP)and TCV treatment control van utilizationin West Texas. The addition of extra HHP(top) reduces pumper utilization andincreases TCV use (bottom) to 100%.

■Risk analysis. How field data may beused to support subsequent operationaldecisions.

British Gas (BG) International Exploration& Production has used the InterACT serviceto relay logs in real time from the North Seaand Trinidad to the company’s headquartersin Reading, southern England. For example,when testing wells, MDT Modular Forma-tion Dynamics Tester pressure measure-ments were relayed to Reading. There, Dr.Nick Colley, BG Principal Petrophysicist,Petroleum Engineering Department, wasable to judge when sufficient data had beengathered for a given zone. At his request, thetool’s probes were then closed and movedto the next zone of interest. It was almost asif he had actually been on the rig at thetime. “The InterACT facility has all the mak-ings of a significant change in petrophysicalworking practice. One can exert nearly allthe control as if one was in the unit itself,”explains Dr. Colley.

Simulating a Better FutureWe have seen how IT-based processes andsystems may be used to clear away theunnecessary ephemera of day-to-day opera-tions and to develop engineers who candeliver world-class services right into clientoffices. However, parallel to these activities,simulation—a recognized tool for risk anal-ysis and process optimization—is beingdeployed so that accumulated informationand data may be better used to improvefuture activities.

For example, costs are influenced by sev-eral sources of uncertainty. The process ofquantifying the effects of these uncertaintiesthrough simulation is called risk analysis(above). Elsewhere, simulation techniquesare being used to model new equipment ortools, accelerating their introduction to thefield. Models of the earth are also being cre-ated so that similar geologies may be com-pared and contrasted to improve log andseismic data interpretation.

Additionally, service company operationsmay be simulated to improve safety or effi-ciency. A key challenge facing an industrywith expanding activity is the need to meetdemand. Capital investment in high-costequipment is one way of meeting medium-or long-term demand. By analyzing globaloperational statistics, it is possible to iden-tify significant trends and develop improvedmarket understanding. This informationmay then be used to better specify whatequipment to acquire. In this way, clientdemand for services may be met cost effec-tively and quickly.

However, short-term needs may often bebetter met through increased use of currentassets. Techniques that help optimize equip-ment levels and new equipment allocation,or ones that identify bottlenecks and suggestresource management strategies, are alreadyin use.

An example of this comes from the USA.To deliver its fracturing services, Dowelloperates thousands of trucks and other vehi-cles. Work-flow analysis of three districts inWest Texas has suggested a potentialincrease in utilization of stimulation trucks.

To examine what stimulation jobs wererequested and which were carried out, datawere gathered from standard reports—ser-vice orders, well treatment reports, supplyservice receipt data, service quality reportsand driver trip reports. Researchers based inAustin modeled equipment allocation usingseveral parameters: the hydraulic horse-power (HHP), pressure and rate of availablepump units; the usage of TCV treatmentcontrol vans; the job requirements and theirlocation; and the number of jobs per dayand the time required for each job.

The modeling process was able to showthat the brake on increasing the utilizationof the pumpers was the number of TCVunits. Increasing the number of pumperswould reduce overall pumper utilization,while average TCV usage would quickly riseto 100% (left). Then, by proposing anincrease in TCV usage, new pumper-sharingpolicies, and a new dispatching protocol—where jobs were allocated in groups ratherthan individually as they were received—the researchers were also able to show theability to meet a significantly increaseddemand, over recent historical averages.

What this relatively simple example indi-cates is that operations research model-ing—which has commonly been used byairlines and in other industries—has a roleto play in improving equipment utilizationand speeding up service company responseto client requests.

Oilfield Review

Experience database

Generalknowledge

Stored data

Stored data

Selection and sampling

Preprocessingand cleaning

Previouscases

Reuse relevantcases

Verify/revisesolution as

required

Retainsolution as experience

Newcase

New case(problem)

Retrieverelevantcases

Applying data-mining techniques

Transformationand reduction

Mergedatabase

Stored data

Stored data

Stored data

Generic databases

Company databases

Data Mining Case-Based Reasoning

■Data mining enables the grouping of scattered and apparently independent data intoclasses or clusters based on some similarities. Case-based reasoning then provides a “user-centered” view, attempting to link a user’s current situation to previous similar situations,from which the data and information relevant to the user’s problem may be identified.

Keeping Track of Data and InformationBetter business processes and effective sim-ulation work using IT depend on access tothe necessary data and information. A clas-sic response to this need has often been tocreate a database. Because of this, the cor-porate information of large industrial com-panies is usually scattered among manysources—databases, spreadsheets and Webpages—distributed over many different geo-graphical locations. Additionally, legacydata repositories are usually constructed tostore and manage dedicated administrativeor operational data, and are targeted for spe-cific uses and needs. For instance, thepumper-truck example above required datagathered from at least six different sources.

Considerable efforts are being made by theIT industry to develop ways of “drillingdown” different data repositories, extractingthe required data and information.7 So-called search engines usually locate relevantdocuments of any format in databases anddata repositories on the basis of matchingquery terms with the documents.8 However,most conventional search engines deliver tothe petitioner unvalidated and unsortedselections. This means that relevant datamay not be available when and where theyare needed—at least not for an organizationas a whole.

Autumn 1997

There is an alternative approach thatmany of us may have benefited from with-out necesarily knowing it. The philosophyof the most effective help-desk services maybest be summarized by the maxim thatwhen faced with a problem, it is a goodidea to find out whether it has happenedbefore. And if it has, to know what solved itlast time. The challenge for many busi-nesses is to gather that data and informationfrom multiple sources and to weight theresponse to take into account the personmaking the request.

For example, the expected answer to thequery, “What is the efficiency of equipmentXYZ?” may differ depending on whether thechief accountant or the technical managerasked the question. The answer would alsohave to use data from many sources.Because the integration of the existingdatabases into one single repository is usu-ally not desirable for cost and technical rea-sons, an alternative is to federate them.

In 1996, groups in France, Norway andItaly—including Norsk Hydro andSchlumberger—embarked on the NOEMIEproject, which has European Unionsupport.9 NOEMIE uses data-mining (DM)techniques—using search engines—toextract data from multiple sources. Then ituses case-based reasoning (CBR) methods toorient the way in which the data are han-dled according to who makes the query.

Using data delivered from severaldatabases by the DM process, CBR com-pares the user’s current problem with previ-ous specific situations—or cases—stored inan experience base. This experience basealso provides the necessary general informa-tion—concept hierarchies, relationships,associations and decision rules—needed todeliver the response. Between the user andthe CBR stage, a further interface relates theprocess to the user’s situation, orienting theresponse to user needs (above).

15

■NOEMIE cases. In order to make this search possible and efficient,a past case needs to be described by a set of generic attributesrelated to the problem. This example lists attributes describing anunwanted event on a platform.

Attribute Examples

Event category Accident, Incident, Failure, Combination

Identification Industry, Company, Failure, Combination,Topside, Subsea, Location, System class orEquipment class

Severity Persons, Environment, Production, Emergencyshut down, Trip, Equipment maintenance

Activity Design, Installation, Testing, Drilling, Workover,Production, Maintenance, Training

Main observed cause Action, Circumstances, Technical, Combination

Remedial actions Human, Operating, Management, Technical,Logistics

Effect of remedialactions

Factors similar to above. Must normally beevaluated based on statistics of the recurrenceof such events after the initial event.

Contributing factors Human, Management, Procedures,Work environment, Electrical, Mechanical,Operations

To assess the feasibility of the methodol-ogy, the NOEMIE team is building a demon-strator that incorporates existing methods foreach part of the chain. The project isexpected to be completed in 1999. Becauseone of the end users involved is NorskHydro, one pilot project is concentrating onevents that have a negative influence onsafety, production reliability, and the cost ofrepairs to oil and gas production facilities.The type of subjects addressed range fromthe specific, “How to improve a specificfailure-prone unit or component,” to thegeneral, “How to optimize the layout of anew production platform.”

When an unwanted event has occurred, adecision needs to be made by managementon how to reduce the likelihood of or pre-vent such incidents in the future. To makethis decision, all cases with similar charac-teristics need to be located with NOEMIE,and the parameters describing those eventsused as a guide for resolving the current sit-uation (right). Such incidents will thenbecome cases in the NOEMIE experiencedatabase and be available to aid future deci-sion-making.

The NOEMIE database aims to resolve theunsorted nature of data and informationdelivered by search engines, through anautomated case-based approach. An alter-native is to create a network of company-wide experts to assess, filter and order thedata and information delivered by searchengines and other sources.

This approach has been adopted by theSchlumberger Technology Watch (TW) pro-gram, designed to harness external knowl-edge and to include all available technologyinto new products and services. The aim is toshorten the product development cycle byenabling in-house research and developmenteffort to concentrate on core competencies.

The TW program also leverages externaldatabases and information feeds. To avoidinformation overload, a filtering agent—therefinery—was developed. The refineryuploads information from relevant sources,such as Petroleum Abstracts and the USpatent database, and indexes it. Anyone inthe company can subscribe to existing pro-files of interest or create his or her own to

16

receive an e-mail alert when new informa-tion is available. This decentralized manage-ment reduces to a minimum the administra-tion of the system.

An example of TW in action is the deliveryof improved downhole battery technology.Development costs for new batteries is high,while time to market may be long. Due tothe oilfield market’s low volume, batteryvendors do not tend to be interested in cus-tomization. Using TW, the state-of-the-art ofbattery technology was determined and therole of existing packaging challenged. Whilesurveying battery technology, Marvin Mile-wits, senior development engineer with theSchlumberger Perforating & Testing Center(SPT) in Rosharon, Texas, met with represen-tatives of a relatively small technology com-pany. Development of a long-term relation-ship with this company helped SPT todesign battery cells that are adapted to someof the most stringent requirements in size,shape and temperature.

Access to key vendor expertise provedvaluable in implementing new chemistry forthe cells and in experimenting with shapesbetter adapted to downhole tools. By lever-aging outside resources, SPT was able toconcentrate on internal requirements, suchas proximity to the drillstring for measure-

ments-while-drilling tools, temperature andlongevity for drillstem test measurements.

Corporate Knowledge: Retention and DeliveryIn the NOEMIE and TW cases describedhere, attempts are being made to organizeand integrate information and data usingexperience, memory and insight, which wasour definition for knowledge creation at thestart of this article. The techniques point to arelatively new application for IT—oftendescribed as knowledge management. But aswe have seen, such techniques may varyfrom essentially automatic systems that bor-der on artificial intelligence to those thatrequire a high degree of human intervention.

For its part, BP defines knowledge manage-ment as an attempt to recognize what is anessentially human asset, and to turn it intoan organizational asset that may be accessedand used by a broader set of individuals onwhose decisions its assets depend.

Oilfield Review

Building on the successful introduction ofvirtual teamworking, BP is now strengthen-ing its abilities in both knowledge captureand knowledge transfer. In practice, thismeans discovering in a systematic waywhat and where the knowledge is, makingspecific knowledge visible, capturing expe-rience and creating the knowledge productspeople want to use.

This has led to a number of projects. ManyBP Exploration assets and interest groupshave established home pages on the BP-Web. These home pages are designed as liv-ing documents to provide up-to-date infor-mation. In some cases, detailed descriptionsof staff competencies have been included.These entries may be interrogated by searchengines to discover the best source of keyknow-how when it is needed. In the longterm, every graduate-entry employee willbuild and maintain his or her personalhome page. This will reinforce personalresponsibility for knowledge management.

Also on the Web is the Einstein page. Thisprovides contemporary information aboutBP experience and training opportunities. Itaims to accelerate the learning process byensuring that staff members have access tothe necessary training resources, includingcomputer-based material for learning at adistance.

Moving forward, BP has a number of pro-jects now in the pipeline. For example, onescheme recognizes the changing expecta-tions regarding employee mobility and isseeking to change the traditional oilfieldpattern and move more work to people,rather than the opposite. Although, somejobs have to be carried out on the spot,more work may be carried out by fullyeffective sub-teams, located remotely from aproject. Work is now under way to optimizethis approach.

Like every other operator, BP Explorationstrives to reduce drilling costs and is seekingto accelerate the speed of learning from onewell to another. The stakes are high as drilling

Autumn 1997

often makes up the lion’s share of develop-ment costs. To enhance learning betweenwells, the company is using video clips of thedrilling team dis-tributed around thecompany on CD-ROM. Immediatelyafter completing thewell, team membersrecord what they didand, more impor-tantly, what theywould do differently.These CD recordingsare facilitated andedited by a profes-sional producer.

In fact, this idea isborrowed from theUS Army, which hasdeveloped tech-niques to retainexperience. After every major training exer-cise or operation, the unit involved immedi-ately holds an after-action review. Held onlocation if necessary, this systematic, facili-tated process is designed to review whatwas supposed to happen, what did happen,why there were differences and who didwhat. The actions of commanders, special-ists and soldiers are all exposed.

Of course, propagating this acquiredknowledge by CD-ROM is just one tech-nique. Another element in enhancing learn-ing is real-time accessibility to previousexperience. To improve drilling perfor-mance, researchers in Schlumberger Cam-bridge Research, Cambridge, England, havebuilt an experimental, Web-based system todiagnosis the causes of stuck pipe duringdrilling and then to suggest remedies.

Stuck pipe was chosen because, in addi-tion to there being an urgent need to reducethe number of toolstrings that are lost inhole, there already exists an industry-accepted database on how to map stuck-pipe decisions.10

Attempts made to oand integinformatiodata usingexperiencmemory ainsight.

Using this database as a starting point,researchers have built a prototype that inter-actively interrogates the user about what

may be observed—for example, stand-pipe pressure trends,changes in overpulland the type of shaleobserved on theshakers. The systemstores a number ofrepresentative cases,and uses them todetermine the likeli-hood of various typesof sticking mecha-nism—such as differ-ential sticking or keyseating. Based on theobservations input bythe user, the systemdisplays the proba-

bilities of various mechanisms causing theproblem. The greater the number of obser-vations, the better refined the predictions ofthe model will be.

More cases will be stored as they areexperienced, allowing a local knowledgebase to be developed. And, by following upcase studies, the decision map may berefined. This system will be used to ensuremore accurate diagnosis, for better earlywarning and improved training (next page).In short, it will accelerate the learningcurve. It is then planned that the local expe-rience be fed back to form a global knowl-edge base. This resource will enable drillersin new areas to more quickly solve stuck-pipe problems.

The system will not stop at just one aspectof drilling. It will also be linked to downholemeasurements, drilling parameters andother elements of the drilling process, offer-ing a web-based, cross-platform drillingknowledge base at the drill floor. Progress indelivering state-of-the-art knowledge glob-ally will then have been made.

are beingrganize

raten and e, nd

17

10. Bailey L, Jones T, Belaskie J, Orban J, Sheppard M,Houwen O, Jardine S and McCann D: “Stuck Pipe:Causes, Detection and Prevention,” Oilfield Review3, no. 2 (October 1991): 13-26.

■Stuck-pipe experience. To improve drilling performance and reduce the incidence ofstuck pipe, researchers at Schlumberger Cambridge Research have developed a Web-based system to diagnose the causes of stuck-pipe incidents and to suggest solutions.

Is circulation restricted?

Gather Information More information

Is overpull in new section?

Smooth

Erratic

Wellbore geometry, formation ledges

Reactive mobile formationsor unconsolidated formations

Unconsolidated, fractured, faultedor geopressured formations

Unconsolidated, fracturedor faulted formations

Fractured, faulted formationsCementblocks, junk

Cementblocks, junk

0%

Differential sticking

Mobile formations

Undergauge hole

Poor hole cleaning

Unconsolidated formations

Key seating

Other

100%

Cementblocks, junk

Key seating

Inadequate hole cleaning

Inadequate hole cleaning

Reactive mobileor unconsolidated formations

Wellbore geometry,fractured or faulted formations

Mechanism

Local database

Wellbore geometry,formation ledges

Observations

Rea

ctiv

e m

obile

form

atio

ns

Tripping out

Yes

Yes

Yes

Yes

Yes

YesKey seating

Yes

Yes

No

No

No

Yes

Yes

Yes Yes

Yes

Yes

Yes

Are known problem formations exposed in new hole section?

Does rotating string allow obstruction tobe passed?

Are known problem formations exposed in hole section drilled by previous bits?

Are known problem formations exposed in hole section drilled by previous bits?




Is downward motion possible?

No

No

No

No

No

NoNo

No

No

No

No

No


Can BHA be rotated free?

Is downward motion possible?

Push Me—Pull YouBecause IT improves communications, theopportunities for experts to collaborateacross international boundaries have dra-matically increased. The picture of commu-nities of interest forming regardless of geog-raphy and company status is one that mostpeople wholeheartedly welcome. However,some reservations have been expressed. For

18

example, a paper published in Scienceraises an interesting prospect of what it callsthe “Balkanization of science.”11

At the heart of this is the notion—which isequally valid for disciplines outside purescience—that limitations on people’s timewill mean that local contact between peo-ple of different specialization decreases astheir efforts to communicate within theirspecialization increases. The possible result

is the creation of a series of highly special-ized, yet unrelated, communities of interest(next page).

One way of avoiding this has beenaddressed by BP, with attempts to createwhat it calls “the virtual coffee machine,” toencourage random exchanges over its VT

Oilfield Review

Geographic communities

Discipline communities

■Balkanized science. Time limitations can lead to fragmenting ofgeographic communities and cross-specialization communities.

equipment. In this way people who mayotherwise never talk, can meet and findcommon interest. The geographically dis-persed Andrew organization held a virtualmeeting every morning and randomexchanges were found to be beneficial.

There is also a need to link schemes likeNOEMIE—call them the information man-agement approaches—with the newlyknowledge-motivated employees of an orga-nization. The objective of these links mustbe to ensure that the information peopleneed is made available efficiently.

In the beginning there was PUSH technol-ogy—applications like e-mail allow thesender to write a message and push it to thepersons of choice. Then came PULL tech-nology—people interested in a particularissue may enter a Web page, access a file ofinterest and pull it to their desktops.

Today, one area of debate centers on whatinformation should be pushed and whatpulled. For example, it is possible, usingapplications like Pointcast, BackWeb,Marimba and other new entries into thismarket, for individual users to register inter-est in certain areas, leaving the software tofind and present this customized informa-tion as required, often in a visually com-pelling way.

The ability to push information to people’sdesktops means that some information, suchas important computer virus updates or inter-esting company news, could enjoywidespread distribution. But not all informa-tion need go to all employees, and liketelexes, faxes and e-mails before, overuse ofthis technology will reduce its effect. The keylies in finding ways of targeting informationto those people who will most benefit fromit. In the jargon of our age: contextual push.

This ability to target will offer boundlessopportunities to better propagate informa-tion through organizations. And, as we entera new millennium, developments in thisarea will shape the way many of us willreceive the information we need. If thesedevelopments bear fruit, we will becomepart of a well-informed work force armedwith the tools to boost productivity, improveservice quality and gain competitive advan-tage. To do this, we will require inclusive,comprehensive and easily accessible corpo-rate memories. —CF

Autumn 1997 19

11. Van Alstyne M and Brynjolfsson E: “WideningAccess and Narrowing Focus: Could the InternetBalkanize Science?” Science 274, no. 5292 (November 29, 1997): 1479-1480.

Clear Fracturing Fluids for IncreasedWell Productivity

Hydraulic fracturing treatments can greatly

improve well productivity by decreasing wellbore

skin. Residue from conventional polymer-base

fluids, however, may clog pore spaces in the

proppant pack, reducing fracture permeability. A

new fracturing fluid offers a solution by not using

polymers. This viscoelastic surfactant fluid requires

no chemical breakers, yet cleans up better than

crosslinked polymer fluids, leading to a lower skin

and greater well productivity.

20 Oilfield Review

Bill ChaseImperial Oil Resources Ltd.Norman Wells, Northwest Territories,Canada

Walt ChmilowskiRichard MarcinewChuck MitchellCalgary, Alberta, Canada

Yen DangHouston, Texas, USA

Kevin KraussErik NelsonSugar Land, Texas

Tom LantzPhillips Petroleum CompanyLafayette, Louisiana, USA

Chuck Parham (consultant)Coastal Oil & Gas Corp.Houston, Texas

Jerry PlummerLafayette, Louisiana

For help in preparation of this article, thanks to VinceDeBonis, David Norman and Paul Price, Dowell, NewOrleans, Louisiana, USA; Chuck Ebinger, CompletionEngineers, Lafayette, Louisiana; Larry Foster and HalRiordan, Dowell, Lafayette, Louisiana; Terry Greene,Dowell, Sugar Land, Texas, USA; David Pierce, Dowell,Houston, Texas; Rob Tinis, Dowell, Red Deer, Alberta,Canada; and Mathew Samuel, Dowell, Longview, Texas.ClearFRAC, DataFRAC, PERMPAC and STIMPAC aremarks of Schlumberger.

1. Martinez SJ, Steanson RE and Coulter AW: “FormationFracturing,” in Bradley HB (ed): Petroleum EngineeringHandbook. Richardson, Texas, USA: Society ofPetroleum Engineers (1987): 55-1–55-12.

2. Skin is a zone of reduced permeability near the well-bore. Skin damage causes an excess pressure droparound the wellbore and reduces formation fluid flowinto the wellbore.

3. Stewart BR, Mullen ME, Howard WJ and NormanWD: “Use of a Solids-Free Viscous Carrying Fluid inFracturing Applications: An Economic and Productiv-ity Comparison in Shallow Completions,” paper SPE30114, presented at the SPE European FormationDamage Control Conference, The Hague, The Netherlands, May 15-16, 1995.

4. Ely JW: Stimulation Engineering Handbook. Tulsa,Oklahoma, USA: PennWell Publishing Company(1994): 79-97.

5. Armstrong K, Card R, Navarrete R, Nelson E, Nimer-

A successful hydraulic fracturing treatmenthas too long been defined as one that waspumped without problems. Rather, the truemeasure of a successful fracture treatment isincreased production or injectivity.1 Theobjective is to improve fluid communicationbetween the reservoir and the well. One ofthe most significant developments inhydraulic fracturing has been the realizationthat many stimulation treatments might notproduce a negative skin or decrease the skinas much as desired.2 Laboratory testing indi-cates that unbroken residue from solids-basepolymer fluids can plug the pores in theproppant pack. Some of this polymerresidue can remain in the well indefinitely,hindering production.3

The ideal fracturing fluid should showminimal pressure drop in the pipe duringplacement, have adequate viscosity tocarry proppant effectively and degradeafter the fracture closes so as not to leaveresidual material.

From a historical perspective, fracturingfluids have evolved significantly since thefirst fracture stimulation was performed in1947. Early stimulation treatments used sur-plus napalm added to gasoline to create aviscous fluid capable of initiating and propa-gating a fracture. During the 1950s, viscous

Autumn 1997

ick K, Samuelson M, Collins J, Dumont G, Priaro M,Wasylycia N and Slusher G: “Advanced FracturingFluids Improve Well Economics,” Oilfield Review 7,no. 3 (Autumn 1995): 34-51.

oils and gelled oils were the fluids of choice,as it was generally believed that water intro-duced into oil reservoirs would cause forma-tion damage. With the realization that waterdid not cause as much damage as originallythought, engineers began pumping watergelled with guar and guar derivatives (lineargelled water) in the 1960s.

As fracturing grew in popularity in the1970s, wells were also being drilled deeperand hotter formations were encountered.There was an increasing need for fluid vis-cosities greater than those offered by lineargels. To attain sufficient viscosity andincreased thermal stability in higher temper-ature reservoirs, linear gels were crosslinkedwith borate, zirconate or titanate ions.4

In the 1980s, foamed fracturing fluidsgrew more popular as engineers becamemore aware of the permeability damagecaused by polymeric fluids. The use offoams decreased the amount of guar intro-duced into the fracture, thereby reducingthe amount of residue and, hence, theextent of damage. In a foam fluid, the gasphase typically occupies more than half ofthe total fluid volume, so less liquid, andhence less guar, is pumped into the well.Foams also enhance cleanup after a fractur-ing treatment. The liquid volume is lower,and the entrained gas offers significantlymore energy to evacuate the fracturing fluidfrom the well.

The quest for cleaner fluids continued intothe 1990s, when advanced breaker technol-ogy and lower polymer concentrationsbecame effective tools for reducing and lim-iting damage from guar.5

The next step was the development of apolymer-free aqueous fracturing fluid. Thisfluid is unlike guar or hydroxyethyl cellulose(HEC) fluid systems; rather, it belongs to anew class of fracturing fluids—those basedon viscoelastic surfactants. This article dis-cusses the new polymer-free fracturing flu-ids and, with specific case studies, detailshow they may be used most effectively toincrease well productivity.

21

■Micelle structure.The micellar struc-ture of the vis-coelastic surfactantin brine appearsrod-shaped orworm-like as themicelles becomeentangled. Theworm-like micellesexceed 100 micronsin length and con-tain several thou-sand molecules.When the viscoelas-tic surfactant fluidbreaks, the micellesbecome spherical,with a diameterroughly equal tothe width of therod-shapedmicelles. Thisimage was takenwith a cryo-trans-mission electronmicroscope at theUniversity of Minnesota.

Hydrocarboncore

Surfactantmolecule

ActivatorSurfactantmolecules

Activator

Brine environment Oil environment

■Breaking ClearFRAC fluids. No additional chemicals are usually needed to break ClearFRAC fluids. Dilution by formation water or contact with hydrocarbons will disruptthe rod-shaped micelles, breaking the fluid.

1µm

Scale

Viscoelastic Surfactant DevelopmentIn 1983, The Dow Chemical Companyintroduced fatty amine quaternary ammo-nium salts containing alkyl groups longerthan C14 as thickeners in consumer prod-ucts, such as bleach, liquid dishwashingdetergent and topical cosmetics. These vis-coelastic surfactants belong to a class ofcompounds that form micelles in an aque-ous system containing certain anions.6 Thedeformation of such systems is time depen-dent; the system acts as a solid unless a suf-ficient amount of shear has been appliedfor a certain length of time. When the sys-tem deforms, the rheological behavior isnearly Newtonian.

A viscoelastic surfactant fluid providesexcellent particle suspension. Dowellapplied this fluid technology first as agravel-pack fluid, PERMPAC fluid. The sur-factant is added to common completionbrines—potassium chloride, ammoniumchloride, calcium chloride or calcium bro-mide—to suspend gravel effectively. The sur-factant concentration varies from 2.5 to 6%by volume, depending on the anticipatedtemperature in the well. The main advantageof this fluid, unlike polymer-base systemssuch as guar or HEC, is that little residue isleft upon breaking. This type of treatmentresults in a gravel pack with significantlyhigher conductivity.7

The principal advantages of viscoelastic-surfactant fluids are ease of preparation,minimal formation damage and highretained permeability in the proppant pack.The fluids are typically prepared by continu-ous mixing of the surfactant into the brinebefore it passes through a high-shearblender. The blender provides sufficientshear for complete dispersion of the surfac-tant and fluid viscosification. Viscoelasticsurfactant fluids can also be used in fracpack and in conventional high-permeabilityfracturing treatments.8 The originalPERMPAC surfactant works well in theseapplications, but cost and temperature limi-tations—less than 140°F [60°C]—preventedwidespread use in hydraulic fracturing appli-cations. Recent modifications to the chemi-cal structure of the surfactant have reducedthe fluid cost and increased the temperaturelimit to 200°F [93°C], opening the door tohydraulic fracturing applications.9

The ClearFRAC surfactant is a blend of aquaternary ammonium salt, erucyl bis (2-hydroxyethyl) methyl ammonium chloride(derived from rapeseed oil), with iso-propanol. ClearFRAC fluid is a mixture ofthis surfactant in brine. The preferred brinecompositions include 3% by weight of

22

ammonium chloride and 4% by weight ofpotassium chloride solutions. At tempera-tures greater than 150°F [66°C], sodiumsalicylate is added as a stabilizer. The sur-factant concentration varies from 0.7 to 4%by volume.

The surfactant molecule consists of ahydrophobic tail that is 22 carbon atomslong. The head group is hydrophilic, and iswhere the quaternary ammonium group islocated. When the surfactant is added to thebrine, the surfactant molecules aggregateinto structures in which the hydrophobic

Oilfield Review

Autumn 1997

Equipment and Materials for ClearFRAC Fluids

Frac tank

Brine

Proppant

ClearFRAC surfactant

Pump truckBlender

Well

Proppant

Guar,breakers andsurfactants

Fracturing Equipment and Materials

Continuousmixer

Frac tank

WaterBactericide

KCI orsubstitute

Pump truck

Blender

Breakers andfluid-loss additives

Well

Crosslinker,buffers and

breakers

■Mixing procedures. A typical crosslinked guar-base fracturing fluid may use up to 13different additives to provide the desired rheological properties. Complex metering andpumping equipment is needed to ensure accurate delivery of the chemical concentra-tions (top). In the simplified ClearFRAC mixing system, the surfactant is added to a brinesolution, then mixed with proppant and pumped downhole (bottom). Fewer chemicalsand less equipment reduce the likelihood of problems during a job.

tail groups are on the inside, and thehydrophilic groups are on the outside. Suchstructures are called micelles. In the case ofClearFRAC fluid, the micelles are rod-shaped or worm-like (previous page, top). Ifthe surfactant concentration is above a criti-cal concentration, the micelles entangle andhinder fluid movement. Such interactionsproduce the fluid’s viscosity.

The viscosity of ClearFRAC fluids is brokenby two mechanisms: contact with hydrocar-bons or dilution by formation water (previ-ous page, bottom). Because one or both ofthese conditions occur in fractured wells, noadditional breaker chemicals are required;however, there are some common additiveswhich can contribute to the break mecha-nism. Produced oil, condensate or dry gasaffects the electrical environment in thefluid, disrupting the micelles. The micelleschange shape from rods to spheres, andfluid viscosity is lost because the micellescan no longer become entangled. In thecase of formation water, dilution of ClearFRAC fluid reduces the surfactant con-centration, and the rod-shaped micelles nolonger entangle with one another.

The field application of these fluids hasbeen successful. Present applicationsinclude wells in which fracture conductiv-ity or fracture length is important, mobi-lization of complex mixing equipment isdifficult, or situations where cleanup is anoverriding concern.

Operationally, the preparation of Clear-FRAC fluids is simple (left). Because nopolymer hydration is required, the surfactantconcentrate can be metered continuouslyinto the brine for easy mixing. No crosslink-ers, breakers or other chemical additives arenecessary. The mixing of the fracturing fluidis simplified by elimination of variances dueto polymer hydration and breaker effectsand the need for extensive metering andpumping systems. Moreover, there is lesswaste due to elimination of tank bottoms,the unpumpable residual fluid remaining inthe bottom of the containers used in batch-mixed jobs.

23

6. A micelle is a molecular aggregate.7. Parlar M, Nelson EB, Walton, IC, Park E and DeBonis

VM: “An Experimental Study on Fluid-Loss Behaviorof Fracturing Fluids and Formation Damage in High-Permeability Porous Media,” paper SPE 30458, pre-sented at the SPE Annual Technical Conference andExhibition, Dallas, Texas, USA, October 22-25, 1995.

8. Stewart et al, reference 3.9. Samuel M, Card RJ, Nelson EB, Brown JE, Vinod PS,

Temple HL, Qu Q and Fu DK: “Polymer-Free Fluid forHydraulic Fracturing,” paper SPE 38622, presented atthe SPE Annual Technical Conference and Exhibition,San Antonio, Texas, USA, October 5-8, 1997.

24

Fluid-LUnlikeClearFRas a reConseqtially cpolymeaqueouleavingClearFRmatrix.about cous flthe por

As a fluid, wthat of fluid. ClearFloss adin fract

ProppaThe cotranspofluid srate of170 sederivedpolymelogicapower-thumb

Viscolike Neent-visspectruthat otems. shear pletelydegradhigh-shtant flproppa

At a gcosity with teperaturincreaadjustinventioncosity dtemperwith hywater,

100,000

10,000

1000

100

10

Vis

cosi

ty, c

p

0.01

Shear rate, sec-1

Formation and fracture Tubulars and perforations

0.1 1 10 100 1000

2.5% viscoelastic surfactant

40-lbm/1000 gal hydroxyethyl cellulose

Viscosity profileat 75°F [24°C]

■Viscosity profile. At high shear rates—above 170 sec-1—the viscosity of the viscoelasticsurfactant fluids is less than that of HEC fluids, giving the viscoelastic surfactant fluidslower friction pressures during pumping.

4% ClearFRAC fluid through brine core20-lbm/1000 gal borate-crosslinked guar through brine core4% ClearFRAC fluid through hydrocarbon core20-lbm/1000 gal borate-crosslinked guar throughhydrocarbon core

■Leakoff-rate. ClearFRAC fluids have a more predictable leakoff rate than guar-base flu-ids. In these tests, 1-in. Berea sandstone cores were saturated with either brine or hydro-carbons. Guar-base and ClearFRAC fracturing fluids were pumped through the cores.The collected fluid volume was measured relative to the core pore volume to determinethe rate at which each type of fluid leaks off into a formation. The leakoff for ClearFRACfluids is not drastically different in both hydrocarbon-saturated and brine-saturated cores.In contrast, the leakoff of guar fluids is affected by the type of fluid in the formation. Thus,fracturing design with ClearFRAC fluids is less complex because the leakoff is more pre-dictable and less affected by the nature of the formation fluids.

oss Control polymer-base fracturing fluids, AC fluid does not form a filter cakesult of leakoff into the formation.uently, the fluid-loss rate is essen-

onstant with time (left). Also, unliker-base fluids where a lower viscositys phase enters the formation matrix, most of the solids behind, wholeAC fluid with full viscosity enters the

At formation permeabilities less than5 mD, it is difficult for an elastic, vis-uid such as ClearFRAC fluid to entere throats.result, the leakoff rate of ClearFRACith no fluid-loss additives, is less thana 20-lbm/1000 gal crosslinked-borateIn high-permeability formations,RAC fluid is compatible with fluid-ditives, and significant improvementsuring fluid efficiency are observed.

nt Transportnventional guideline for proppantrt is that the viscosity of a fracturing

hould be at least 100 cp at a shear 100 sec-1 or 50 cp at a shear rate ofc-1 (below left). This guideline was from experience with conventionalr-base fracturing fluids whose rheo-l behavior generally follows thelaw rheological model. This rule-of- may not apply to ClearFRAC fluid.elastic surfactant fluids behave morewtonian fluids, with a flatter appar-cosity profile across the shear-ratem. This response is different from

f most aqueous-base polymer sys-The viscoelastic surfactant fluid isthinning, but its rheology is com- reversible and has no permanentation of viscosity when exposed toear conditions. Viscoelastic surfac-

uids provide ample viscosity fornt transport in the fracture.iven surfactant concentration, the vis-of a ClearFRAC fluid will decreasemperature (next page, top). This tem-e-related thinning can be reduced bysing the surfactant concentration org the salt concentration. Unlike con-al polymer systems, however, the vis-oes not degrade with time at a given

ature. Until the fluid is contaminateddrocarbons or diluted with formation

Oilfield Review

the viscosity will remain stable.

Autumn 1997

500

400

300

200

100

075

Fluid temperature, °F

3% ammoniumchloride base fluid

100 125 150 175 200

1.0% 1.5% 2.0% 3.0% 4.0%

Viscoelasticsurfactantconcentration

Vis

cosi

ty a

t 100

sec

,-1cp

■Typical rheological performance. Viscoelastic surfactant fluids generate high viscosityat low temperatures but thin with increasing temperature, making them excellent frac-turing fluids at temperatures below 200°F and especially in low-temperature wells.

Sur

fact

ant c

once

ntra

tion,

%

Fluid temperature, °F

50 75 100 125 150 175 200

5

4

3

2

1

0

■Proppant transport. With increasing bottomhole temperature, a larger percentage ofClearFRAC surfactant is required for adequate proppant transport.

The behavior of surfactant-base fluids dif-fers substantially from that of guar-base flu-ids, and laboratory observation and fieldexperience have suggested that viscoelasticsurfactant fluids with viscosities below theconventional guideline are efficient and arecapable of placing proppant as per design.These observations led to a series of prop-pant-transport tests at STIM-LAB, an inde-pendent laboratory in Duncan, Oklahoma,USA. ClearFRAC fluids were tested at vari-ous flow rates, proppant concentrations andtemperatures to correlate fluid propertieswith proppant transport capability (see“Laboratory Testing,” page 27).

Effective proppant transport was demon-strated in a large-scale fracture simulator atfluid viscosities as low as 30 cp at 100 sec-1.This result was due to the fluid’s elasticityand high viscosity at low-shear rates.

The proppant transport tests proved thatviscoelastic surfactant fluids provide ade-quate proppant transport throughout the 75to 175°F [24 to 79°C] fluid temperaturerange. Even when small amounts of prop-pant settling occurred, the perforation areawas kept clear at all times, and more than90% of the proppant remained in suspen-sion throughout the fluid volume. It isimportant to mention that when a nonvis-cosified brine/sand slurry is pumpedthrough the slot, the sand immediatelydrops and plugs the perforation.

The tests demonstrated that the conven-tional viscosity versus proppant transportguideline derived for polymer-base fluidsmay not apply to ClearFRAC fluids. At ambi-ent temperature, a 42-cp fluid providedexcellent proppant transport (no apparentsand settling), yet at 175°F, a 100-cp fluidallowed some sand to settle. A better guide-line might be to use the minimum concen-trations that were empirically estimatedduring this experimental program (left).

Fluid viscosity, as calculated by the power-law model, does not adequately predict theproppant transport capability of ClearFRACfluids. Further work is needed to determinerheological parameters that better describethe proppant transport behavior of ClearFRAC fluids.

25

Liquid outlet

Collectionvessel

Lightsource

Lightsource

Videocamera

Visual cell: 1-in. diameter

Pressure taps, 10 cm apart

Bubble size analyzer

Visual cell: 1-in. diameter

Digital processindicator

Precision pressuretransmitter

Backpressure

Gas outlet

Gas inlet

Transparent sand pack cell:5.00-cm OD4.12-cm ID100-cm length30- to 50-mD permeability

Check valve

Check valveValve

Thermal massflow controller

Antisurge valve

Microprocessorand control equipment

Pump

N2 BottleFoam generator

Precision pressuretransmitter

Safe container

Relief valve

■SCR foam test equipment.

Foam Stability and RheologyTo help reduce fluid cost, improve fluid effi-ciency and accelerate cleanup, ClearFRACfluids have been mixed with nitrogen to pro-duce foamed fracturing fluids. Experimentsto quantify the stability and rheologicalbehavior of these systems at elevated tem-peratures and pressures were performed atSchlumberger Cambridge Research (SCR),Cambridge, England, and at STIM-LAB.

Initial experimental work at ambient tem-perature and pressure showed that stablefoams with a half-life—the time at whichhalf of the liquid phase has separated—exceeding 12 hr could be obtained withClearFRAC fluids containing only the surfac-tant itself as the foaming agent.

To determine the suitability of ClearFRACfoams for use in the field, it was necessaryto conduct experiments to quantify their sta-bility and rheological behavior at elevatedtemperature and pressure. The initial testswere conducted with the SCR test appara-tus, which uses a heated syringe pump tofeed the liquid phase into a foam generator.A nitrogen bottle with a digital flow con-troller feeds the gas into the foam generator.The foam passes by a sight glass, allowingobservation of foam texture and bubble sizedistribution. A collection vessel with a glasswall is then filled. The cell is heated by awater bath to control temperature. The foamgenerator and collection cell can be pres-sured up to 1200 psi [8273 kPa] (right).

Four base fluids were tested, all preparedfrom 3% ammonium chloride brine andvarying concentrations of ClearFRAC surfac-tant. The foams were generated at about1100 psi [7583 kPa] and ambient tempera-ture, then pumped into the preheated collec-tion vessel. The foams were evaluated atthree test temperatures: 110°F [43°C], 150°F[66°C] and 190°F [88°C]. Each experimentwas videotaped to show the texture of thefoam, foam during filling of the collectionvessel and the condition of the foam afteraging in the collection vessel. The half-livesof foam prepared from ClearFRAC fluidsrange from greater than 12 hr at fluid tem-peratures less than 150°F to 40 min at 190°F.

At STIM-LAB, the rheological behavior ofClearFRAC foams was tested with base fluidscontaining 0.2 to 3.0% ClearFRAC surfac-tant. At a fluid temperature of 75°F, the rheo-logical behavior of ClearFRAC foams wasevaluated at various concentrations andfoam qualities (next page). In most cases, themaximum viscosity was achieved at a foamquality of approximately 80%. The finest andmost uniform texture was observed at a 70%foam quality. There is little apparent morpho-

26

logical difference between the foams pre-pared from different brine compositions.

The results of these experiments showedthat ClearFRAC surfactant is an excellentfoaming agent in its own right, and no addi-tional surfactants are necessary to producestable foamed fracturing fluids from 50 to

90% quality up to a temperature of at least175°F. In addition, in the 70 to 80% qualityrange, viscosities far greater than those ofthe base fluids can be attained. This effectreduces the surfactant concentrationrequired to prepare useful ClearFRAC fluids.

These findings have been borne out by asuccessful series of more than 100 experi-mental foamed fracturing treatments in

Oilfield Review

Autumn 1997

Laboratory Testing

1. In a power-law fluid, viscosity is a function of shear rate:µ=Kγn-1.

2. The abbreviation ppa indicates pounds of proppant addedto 1 gal of fluid.

Viscoelastic surfactant

concentration

1.5% 0.5% 0.4% 0.2%

550

500

450

400

350

300

250

200

150

100

50

0

Vis

cosi

ty a

t 100

sec

,-1cp

Foam quality, %

50 60 70 80 90

3% ammoniumchloride base fluid

■Foam rheology. Mixed in a 3% ammonium chloride brine at 75°F, ClearFRAC foamreached maximum viscosity at 80% foam quality.

Canada and Kansas, USA. Use of nitrogenor carbon dioxide to prepare energized—less than 52% foam quality—or true foamedClearFrac fluids with foam quality greaterthan 52% has been established.

To date, nearly all of the foamed ClearFRACjobs have used ammonium chloride as thebrine salt, partly due to the low foam viscosi-ties at low temperatures observed in the labo-ratory experiments.

In Canada, ammonium nitrate is now usedas the brine component in many wells. Mostof the spent fracturing fluid is disposed of bylandfarming, so brines with chlorides mustbe carefully monitored to remain withinenvironmental limits. Because ammoniumnitrate is a fertilizer, landfarming thesebrines is advantageous to the environment.

In proppant transport tests conducted at STIM-LAB

in Duncan, Oklahoma, the laboratory equipment

consisted of two 50-gal [0.2-m3] mixing tanks and

a heated section of tubing. The fluid was agitated

with mixing blades and then pumped through the

1-in. (0.89-in. ID) tubing, which was heated by gas

burners. Shear was simulated by pumping the

fluid through the tubing. Once the fluid left the tub-

ing, it traveled through a series of four sizes of

pipe ranging from 1/2 in. to 1 in. Pressure drops

were measured over each section of pipe and a

rheogram of shear stress versus shear rate was

generated to calculate the flow behavior index, n,

and consistency coefficient, K, values.1 Changes

in pressure were measured in duplicate on all

pipes and slots to allow a backup measurement in

the event of transducer line plugging by proppant.

Flow rate and pressure data were recorded contin-

uously throughout the test runs, and the data were

processed to plot shear stress versus shear rate.

The capability of a fluid to transport proppant

was assessed visually when the fluid system was

pumped through a slot 1 ft [0.3 m] high by 8 ft

[2.4 m] long with a 5/16-in. [0.8-cm] gap width.

One panel of the slot was transparent to allow

visual inspection. The diameter of the perforation

in the center of the slot entrance was 5/16 in.

Dynamic transport of proppant in the slot was

recorded on videotape. A grid system was placed

on the slot to observe proppant-settling velocities

as the slurry traveled across the 8-ft length. The

fluid was stopped at the end of a run to record the

static settling time. The pumping rate of fluid

through the slot was varied from 1 to 3 gal/min [4

to 11 L/min]. The shear rate at the slot varied from

20 to 60 sec-1, and the shear rate at the perforation

varied from 1300 to 3900 sec-1.

The proppant used in all tests was 20/40 Ottawa

sand, and tests were performed at 4 and 8 ppa.2

The goal of the tests was to determine the mini-

mum concentration of surfactant necessary to pro-

vide adequate proppant transport. The tests were

run at ambient temperature, 150°F and 175°F.

The first test fluid consisted of 0.5% ClearFRAC

surfactant in 3% ammonium chloride at 75°F

[24°C] with a viscosity of 22 cp at 100 sec-1. Before

the proppant was added, the fluid was allowed to

circulate through the system. At 3 gal/min

[11 L/min], significant turbulence was observed

around the perforation opening. The fluid exhibited

a chunky texture upon exiting the perforation,

showing a quick recovery after experiencing the

high-shear environment.

When red dye was added to the fluid to observe

flow behavior, it was immediately apparent that

the fluid in the center of the slot was moving more

quickly than at the edges. Such laminar flow

behavior may explain why lower friction pressures

are observed in the fluid compared to polymer-

base systems when ClearFRAC surfactant is used.

When sand was added at 4 ppa, the perforation

area was kept clear at all times; however, some

minor amounts of sand settled at the bottom of the

slot. Nevertheless, the settled proppant continued

to be dragged along the bottom of the slot to the

exit perforation. An in-line densitometer showed

that greater than 90% of the proppant was in circu-

lation. Since the area around the perforation open-

ing never plugged, this can be considered as ade-

quate proppant transport. This finding is supported

by field results.

The test was repeated with a doubling of the

ClearFRAC concentration to 1%; the fluid viscosity

increased to 42 cp at 100 sec-1. When dye was

shot into the fluid prior to the addition of proppant,

laminar flow behavior was again apparent; how-

ever, the fluid displayed a much tighter structure.

After exiting the perforation, the fluid instantly

recoiled to form a gelatin-like mass. When sand

was added at 4 ppa, there was no evidence of set-

tling. The fluid appeared to grab the sand grains at

the perforation exit and carry them across the slot.

The same effect was seen at higher concentra-

tions. Pumping was stopped, and sand settling

was not evident after 30 min. Further tests were

conducted with varying concentrations of ClearFRAC

surfactant and at varying temperatures.

27

10. Barree R and Mukherjee H: “Engineering Criteria for

Fracturing Fluid FlowbackTo maximize well productivity, it is essentialto maximize fracture cleanup. Polymerresidues that stay in the fracture contributesignificantly to a lowered proppant-packpermeability, leading to a loss in treatmenteffectiveness. Even a small amount of poros-ity loss can cause major loss in retained per-meability.10 Parameters such as types andconcentrations of gelling agent, crosslinker,breaker, reservoir temperature, flowbackrate and shutdown time can affect thedegree of permeability damage. To under-stand the relationship of these parameters tofracture cleanup, quantification of the poly-mer in the flowback fluid is crucial.11

A basic assumption is that a cleaner frac-ture will produce reservoir fluids at a higherrate. But how is fracture cleanup related toproduction? A reasonable analysis is that agiven mass of returned polymer produces agiven volume of pore space available forflow in the proppant pack. Therefore, underequivalent reservoir conditions, a directrelationship should exist between returnedpolymer and production. The conventionalmethod of quantifying cleanup from ahydraulic fracture has been to report load-water recovery. This amount may be affectedby produced formation water, and hencemay be inaccurate. Instead, a colorimetricmethod that involves a phenol-sulfuric acidreaction is used to accurately test thereturned fluids for guar or HEC.12

28

Pol

ymer

retu

rned

, %

70

60

50

30

10

0

40

20

East Texas Co

ETX

04E

TX01

ETX

02E

TX03

ETX

05E

TX10

ETX

08E

TX06

ETX

07E

TX09

CO

-HC

O-A

CO

-FC

O-I

CO

-KC

O-E

■Flowback analysis. An analysis of 150 hydran average return of 35 to 45% of the polymeresidue remaining in the fracture could impe

Analysis of the fracturing fluid returned tothe surface after hydraulic fracturing indi-cates that only 35 to 45% of the guar-basepolymer that is pumped during the treat-ment flows back out of the well during the1- to 10-day flowback period (below). Theremaining polymer stays in the fracture anddecreases well productivity. Thus, there is aneed for a fracturing fluid that can bebrought back to the surface more efficiently.

One method of assessing damage involvescore-flow tests. Leakoff tests were con-ducted on 12-in. [30-cm] Berea sandstonecores at a differential pressure of 1000 psi[6890 kPa]. Two core permeabilities weretested: 230 mD and 1000 mD. The coreswere flooded with 40-lbm/1000 gal borate-crosslinked guar, an 80-lbm/1000 gal HECpolymer and 4% ClearFRAC solution,respectively (next page, top). After theleakoff test, the cores were left in a fluid-losscell, and brine was injected in the oppositedirection. Steady flow rates were reached todetermine the retained permeabilities. In thecore flooded with ClearFRAC fluid, flowbegan immediately as the brine diluted andbroke the fluid. The guar- and HEC-floodedcores had significantly lower permeabilities,even after 24 hr of cleanup. These core-flood tests clearly indicate that polymerresidues can decrease the core permeability.

Well

lorado South Texas Central Texas

CO

-LC

O-C

CO

-JC

O-B

CO

-GC

O-D

STX

01S

TX02

STX

03S

TX04

STX

05S

TX06

STX

07C

TX07

CTX

08C

TX01

CTX

02C

TX03

CTX

04

aulic fracturing treatments in the USA showsr pumped. The large volume of polymerde fluid flow into the well.

When wells treated with ClearFRAC fluidare initially flowed back, the tail of theslurry may still have significant viscosity ifit has not yet contacted hydrocarbons orformation water. To help clean up the frac-turing fluid during the initial flowback,especially in underpressured reservoirs,small amounts of some polar organic com-pounds can be added to the tail slurry toaccelerate breakdown. Used in conjunc-tion with proper flowback techniques, thisprocedure can minimize or eliminate earlysand production.13

Shallow Gas Foam FracturesIn September 1996, PanCanadian PetroleumLtd. conducted a field test of the ClearFRACsystem in 10 wells in a shallow gas field inthe Princess East field in Brooks, Alberta,Canada. Like many stimulated wells, thesewells were in poor-quality reservoir areas.Five wells were fractured with this fluid, andfive were fractured with a low-guar systemas a control. Each well had four formationstreated. These low-pressure gas wells werefractured with foamed ClearFRAC fluid; thenitrogen in the foam provided the energy toassist the wells back to production andimprove cleanup. The wells treated withClearFRAC fluid flowed back a greater vol-ume of fracturing fluid—based on fluid vol-umes and colorimetric analysis—than didthe control wells. This improved initial

Oilfield Review

Fracture Flowback Procedures,” paper SPE 29600,presented at the SPE Rocky Mountain Regional/Low-Permeability Reservoirs Symposium, Denver, Col-orado, USA, March 19-22, 1995.

11. Armstrong et al, reference 5.12. Pope D, Britt L, Constien V, Anderson A and Leung L:

“Field Study of Guar Removal from Hydraulic Frac-tures,” paper SPE 31094, presented at the SPE For-mation Damage Control Symposium, Lafayette,Louisiana, USA, February 14-15, 1996.

13. Anderson AJ, Ashton PJN, Lang J and Samuelson ML:“Production Enhancement Through Aggressive Flow-back Procedures in the Codell Formation,” paper SPE36468, presented at the SPE Annual Technical Con-ference and Exhibition, Denver, Colorado, USA,October 6-9, 1996.Card RJ, Howard PR and Féraud J-P: “A Novel Tech-nology to Control Proppant Backproduction,” SPEProduction & Facilities 10, no. 4 (November 1995):271-276.Willberg DM, Card RJ, Britt LK, Samuel M, EnglandKW, Cawiezel KE and Krus H: “Determination of theEffect of Formation Water on Fracture Fluid CleanupThrough Field Testing in the East Texas Cotton Val-ley,” paper SPE 38620, presented at the SPE AnnualTechnical Conference and Exhibition, San Antonio,Texas, USA, October 5-8, 1997.

40-lbmborate-

crosslinkedguar

Retainedpermeability

HEC 80


Greater than 24-hr

cleanup

Immediatecleanup

1000-mD permeability

1 in. 2 in. 4 in. 5 in.

12 in.

12 in.

52% 82% 95%

47%

73%

20%

ClearFRACfluid


1000-psi differential pressure

Greater than 24-hr

cleanup

■Core-flow test results. Berea sandstone cores treated with ClearFRAC fluid cleaned upmore quickly and regained more permeability than similar cores treated with borate-crosslinked guar or HEC.

1400

1200

1000

800

600

400

200

300

250

200

0

50

100

150

Pro

duct

ion

rate

, m3 /d

Tota

l pro

duct

ion,

thou

sand

m3 /d

June

ClearFRAC fluidControl fluid

0July

ClearFRAC fluidControl fluid

■ClearFRAC test wells. PanCanadian Petroleum Ltd. tested ten wells in the Princess Eastgas field in Alberta, Canada. Five wells were fractured with ClearFRAC foam fluid andfive with a crosslinked low-guar fluid. After 11 months of commingled production, thewells were tested individually over a three-month period. Analysis of each group of fivewells shows the wells treated with ClearFRAC fluid produce about 10% more than thecontrol wells.

cleanup and flowback of the fracturing fluidare believed to have contributed to higherproduction rates.

An initial comparison of well productionduring the first few months following thefracture stimulations was impossiblebecause fluid from all zones in each welland from all the wells were commingledinto a common pipeline and not monitoredindividually. From June through August1997, PanCanadian conducted flow-provertests on these wells to quantify the long-term differences in production attributableto the different fracturing fluids.

Before the wells were tested, each wascleaned out with coiled tubing to removeany water and sand fill. The formations inthis part of the Princess East gas field are ofpoor quality; hence, overall production perwell is low. In addition, the wells tend toload with water over time if water is pre-sent, as was the case here. The gas rates foreach group of five wells were plotted dur-ing the test period (below right).

The results were encouraging. A year later,the wells treated with ClearFRAC fluid stillaveraged about 9 to 10% more production.The curves converged slightly after onemonth as certain wells had water loadingand slugging. It is possible that the quality ofthe reservoirs in the specific wells mayaccount for the incremental production;however, statistical averaging of the 20zones in each group of wells in the samearea indicates this cause is unlikely. Assum-ing an average rate difference of 500 scm/d[17 Mscf/D] for the five wells over time—not integrated for specific decline curves buttaken from the cumulative difference duringthe first 51 days of the test shown in the fig-ure—ClearFRAC wells produced a cumula-tive volume of 150,000 m3 [5300 Mscf]more than the control wells. The incremen-tal revenue may be small from these low-producing wells, but the favorable trend hascaused PanCanadian to analyze the poten-tial long-term production improvement inother wells in Brooks and Drumheller,Alberta, with 500 to 600 wells slated forClearFRAC treatments next year.

Acid FracturesSeveral cases involving acid fractures illus-trate the importance of using a nondamag-ing fracturing fluid. Imperial Oil ResourcesLimited’s Norman Wells field is an oil fieldnow under waterflood in the Northwest Ter-ritories, Canada. The field has 325 wells—166 producers and 159 injectors arrangedin a five-spot pattern on six man-made and

Autumn 1997

three natural islands in the MackenzieRiver. The field’s economic life is expectedto run through the year 2010. At present,some areas of the field are marginally prof-itable, and a solution is needed to improverecovery rate.

The formation is the Kee Scarp, a micriticlimestone with natural fractures, vertical andhorizontal, induced by the tectonics of themountain ranges on both sides of the river.

Because the overburden stress is the leastprincipal stress, any induced fracture will behorizontal, not vertical. Thus, to increaseproduction, each low-rate layer in each wellhas to be treated. The frequency of naturalfractures varies throughout the field.

The formation has up to 18 distinguishablelayers in four separate reef or shoalsequences, with an average reservoir heightof 80 m [262 ft]. The respective layer heights

29

Well A

Well B

Well C

4

12

22

5

22

34

3

1

2

4

3

5

Oil

Before fracturing

Flow rates, m3/d

After fracturing

Water Oil Water

■Norman Wells field production data. In the Norman Wells field in the Northwest Territories, Canada, the first wells fractured with ClearFRAC fluid immediately began producing more oil.

vary from 2 to 10 m [7 to 33 ft], and not alllayers are in every well. The average truevertical depth is 450 m [1476 ft], and theaverage bottomhole temperature is 25°C[77°F]. The permeability is typically greaterat the top of the reservoir and decreaseswith depth. Average formation permeabilityis 4 mD, but there are few data on verticalpermeability. The higher permeability layersat the top of the Kee Scarp do not requirestimulation because many are already pro-ducing with increasing water cuts withinexpected production rates.

Historically, stimulation of the producersin the Norman Wells field has involvedselective acid treatments, and more recently,foam-diversion acidizing treatments. Neithertechnique has yielded consistent results onthe producers, although the foam-diversionacidizing treatments have had excellentresults on the injectors. A side effect of theearly acid treatments has been a weakeningor loss of wellbore isolation between layers.Thus, to induce a fracture in lower layers, aremedial isolation strategy was needed.

Acid-fracturing treatments were done inthe past with no attempt to direct theinduced fracture into the lower permeabilitylayers. Thus, the fractures went into the topof the zone, with no increase in production.The treatments pumped 50 to 100 m3

[13,000 to 26,000 gal] of fluid and hadexpected fracture lengths of about 50 m[164 ft]. The only net result directlyattributable to these treatments, however,was accelerated water breakthrough. At thispoint, it was determined that the only wayto increase production in the poorer per-forming wells was with a small inducedfracture, 10 to 15 m [33 to 49 ft], directed atthe tighter permeability zones.

30

The isolation problem was solved, and aninitial fracturing program was proposed withalternating stages of a low-pH guar gel, forfluid leakoff control, with 25 m3 [6600 gal]of 15% hydrochloric acid. This gel systemwas selected because of its compatibilitywith acid at low temperature and its ease ofpreparation, with minimal manpower andequipment. A crosslinked guar was deemedbetter for leakoff control than gelled acidbecause the bottomhole temperature inthese wells was below or at the minimumrecommended for these fluids. One of themain concerns in using a guar-base fluidwas suspected loss of whole polymer intosecondary fissures and fractures and theinability to recover this fluid. Unrecoveredpolymer could remain in these pathways,thereby limiting production.

ClearFRAC fluid became available at thebeginning of the stimulation program in thefield. The fracturing fluid system waschanged because the viscoelastic surfactantfluid is easy to mix, fracturing fluid viscositycan be altered simply by changing surfactantconcentration, crosslinking times are not anissue, and no breakers or extra chemicalsare needed. A viscoelastic surfactant fluidwas most desirable because it would notplug the secondary fissures and fractures.

The first treatment was on a well in thesoutheast end of the field where the wellsintersect the lower, tighter layers of the KeeScarp. The ClearFRAC fluid was mixed in a4% potassium chloride brine. The treatmentcalled for alternating pads of ClearFRACfluid and acid, followed by an overdisplace-ment with brine. A DataFRAC analysis indi-cated the pad volumes of ClearFRAC fluidcould be reduced by 60% from the originaldesign because of a low leakoff rate. It is notknown whether the low leakoff rate was a

local formation phenomenon or a propertyof the viscoelastic surfactant fluid, becausethe original leakoff rates were determinedfrom matrix treatments using gel-base fluidsin other parts of the reservoir.

In total, seven fracturing treatments withstaged ClearFRAC fluid and acid were doneon three wells. The production results so farare encouraging (left). In all three wells, oilflowed to surface within 30% of the recov-ered injected volume with no signs of emul-sion or other fluid incompatibility problems.If a polymer fluid had been used, with littlefluid returned, there would be the potentialfor a large amount of unrecovered polymerto plug the secondary fractures and fissuresand impede production.

Frac PackingHydraulic fracturing has long been consid-ered a stimulation treatment for low-perme-ability formations in hard-rock areas. Inthese treatments, the goal is to create a long,thin fracture with a large surface area. Frac-ture half-lengths, or wings, can reach 500 to1000 ft [150 to 300 m] and have widths intenths of an inch or less.

In contrast, frac packs typically have shortwing lengths of 10 to 50 ft [3 to 15 m] andfracture widths of 1 to 2 in. [3 to 5 cm]. Afrac pack, or STIMPAC treatment, is a frac-ture created using high-viscosity fluid,pumped above fracture pressure, to placesand outside the annulus between the cas-ing and downhole screen and a short dis-tance into the formation. The aim is to cre-ate a high-conductivity sand pack extendinga sufficient distance from the wellbore,beyond any wellbore damage, to create aconduit for the flow of reservoir fluids atlower pressure differentials.14

In late 1991, Dowell and BP Explorationperformed the first true frac pack in the Gulfof Mexico.15 Since then, the incidence offrac packing has skyrocketed, with some600 frac-pack treatments conducted lastyear out of roughly 1200 sand-control com-pletions in the Gulf. The trend is continuingas more operators are considering frac packsin areas that were once gravel packed.

Frac packs are ideal completion tech-niques to bypass formation damage. Manymechanisms can cause wellbore damage—crushing due to stress, fluid and solids inva-sion during drilling, perforating damage,fines migration or precipitation of paraffinor scale. Some of this damage is stressinduced and, therefore, may not be uniform

Oilfield Review

Fluidflow

Fluidflow

Fluid, solids andperforating damage

Borehole

Radial Flow

■Stress and radial flow. Damage fromdrilling fluids, drilling-induced stress, perfo-rating damage and fines migration all con-tribute to nonuniform damage in the near-wellbore region. The damaged area shownabove is an area of increased stress. Higherstress on an unconsolidated formation willreduce permeability. Fluid no longer flowsequally in all directions, increasing fluidvelocities and fines migration, which fur-ther damage the near-wellbore region.

Bilinear Flow

Borehole

Fracture wings extending past damage

Fluidflow

Fluidflow

Stress-induceddamage

Fluid, solids andperforating damage

■Fracture flow. With a frac pack extendingpast the near-wellbore damage, flowbecomes bilinear. Fluid flows almost per-pendicular to the fracture and then downthe clean proppant pack toward the well.Fluid velocities are then lower in the forma-tion for the same or even higher flow rates.The result is a lower drawdown and lessmigration of fines into the wellbore area.

14. DeBonis VM, Rudolph DA and Kennedy RD: “Expe-riences Gained in the Use of Frac-Packs in UltralowBHP Wells, U.S. Gulf of Mexico,” paper SPE 27379,presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana,USA, February 7-10, 1994.Fan Y and Economides MJ: “Fracture Dimensions inFrac&pack Stimulation,” paper SPE 30469, presentedat the SPE Annual Technical Conference and Exhibi-tion, Dallas, Texas, USA, October 22-25, 1995.Ebinger CD: “Frac Pack Technology Still Evolving,”Oil & Gas Journal 93, no. 43 (October 23, 1995): 60-70.

15. McLarty JM and DeBonis V: “Gulf Coast Section SPEProduction Operations Study Group—TechnicalHighlights from a Series of Frac Pack Treatment Sym-posiums,” paper SPE 30471, presented at the SPEAnnual Technical Conference and Exhibition, Dallas,Texas, USA, October 22-25, 1995.

16. Brannon HD and Pulsinelli RJ: “Breaker Concentra-tions Required to Improve the Permeability of Prop-pant Packs Damaged by Concentrated Linear andBorate-Crosslinked Fracturing Fluids,” SPE Produc-tion Engineering 7 no. 4 (November 1992): 338-342.Samuelson ML and Constien VG: “Effects of HighTemperature on Polymer Degradation and Cleanup,”paper SPE 36495, presented at the SPE Annual Tech-nical Conference and Exhibition, Denver, Colorado,USA, October 6-9, 1996.

17. Efficiency is the volume of the fracture divided bythe total volume pumped. With very permeable rock,more fracturing fluid can leak off to the formation,decreasing fluid efficiency.

18. The kh product is the permeability times the heightof the fracture.

around the wellbore. Only a few of thesedamage mechanisms can be corrected withacid treatments. Because the damage isoften nonuniform, flow into the wellboremay not be equal in all directions. Even aperfect gravel pack will produce throughless than 100% of the available flow area insuch cases.

Radial flow toward the wellbore causeshigher velocities and pressure drops adja-cent to the well (above). These high ratesand pressure drops can cause formationminerals to become mobile and bridge nearthe wellbore, in the perforation tunnels, orin the annular gravel-pack region. This dam-age, along with declining reservoir pressure,causes production from gravel-packed wellsto decline over time. As wells age, operatorsoften open the wells up, increasing draw-down to maintain production; however,such actions also increase fines migrationand damage.

A frac pack addresses these problems bycreating a conduit perpendicular to the mini-mum principal stress, extending beyond anynear-wellbore damage. The flow area intothe wellbore increases, reducing pressuredrop and fluid velocities in the formation,thereby eliminating the causes of finesmigration. Formation fluids establish a some-

Autumn 1997

what bilinear flow at lower velocities (right).The key property of the fracture is that itmust be highly conductive.

Perhaps the most critical factor to ensurefuture production in a frac pack is keepingthe proppant pack clean. With polymer frac-turing fluids, polymer residue can be signifi-cant on the small fracture face in a frac-packcompletion. At the end of a treatment, asthe formation begins to close, a polymerfluid will have no place to go but into theformation. The fluid leaks off into the forma-tion, and the polymer residue can becomehighly concentrated as a filter cake. Part ofthis problem can be mitigated by usinggreater concentrations of breakers or encap-sulated breakers, which are crushed by theclosing formation and release the breaker inthe appropriate location. These mechanismswork well, but are still imperfect.16 Someestimates put the efficiency of these break-ers at less than 50%. The breaker itselfcould leak off into the formation, bypassingdehydrated polymer.

HEC has often been used as a frac-packfluid because of its small particle size. Thedisadvantage, however, is that too muchpolymer is needed to create the viscosity tofracture high-permeability zones; a viscouspolymer membrane may form and requirehigh volumes of breakers. Borate-crosslinked low-guar systems are used inmany frac packs worldwide, especiallywhere temperatures exceed 200°F. The useof crosslinked fluids cut frac-pack costs andexpanded use of the technique. The fluids,however, can still produce a positive skinfactor in the completion.

Viscoelastic surfactant fluids are particlefree and behave like linear fluids, makingthem ideal for frac-pack operations. Vis-coelastic fluids typically have lower fluidefficiency than normal fracturing fluids,but, in this application, that is a desirablefeature.17 Viscoelastic surfactant fluids havea nearly constant leakoff response to pres-sure. In addition to being solids free andnondamaging, viscoelastic surfactant fluidshave advantages over crosslinked polymersin the way they propagate fractures.

As the viscoelastic surfactant fluid ispumped, much of it leaks off into the forma-tion. The viscosity of the remaining fluid pro-duces drag forces on the rock, initiating thefracture. With a fluid that does not build a fil-ter cake, the rate required to create a fracturecan be calculated from Darcy’s law. Themain factors that control this fracturing rateare the kh product, injected fluid viscosity,fracturing pressure and reservoir pressure.18

Of these, only the injected fluid viscosity can

31

Density porosity, p.u.

Neutron porosity, p.u.

Density correction, g/cm

0

0

60

60

-0.25 0.253

XX100

Perforated interval XX075 to XX097 ft

Gas-water contact XX110 ft

Gas zones

XX200

■Fracturing above a water zone. This gas well in the High Island area in the Gulf of Mex-ico was a perfect candidate for a ClearFRAC stimulation. The gas zone from XX072 toXX110 ft sat directly atop a water zone from XX110 to XX160 ft. A frac pack was carefullypumped through the perforated interval from XX075 to XX097 ft to bypass wellbore dam-age yet stay above the gas-water contact.

be controlled, unless a fluid-loss agent isused—thereby reducing the kh product bydecreasing the permeability. The fluidleakoff limits the length of fracture that canbe created, but it affects the job cost if largevolumes of fluid leak off. The fracture is cre-ated by a volume of fracturing fluid thatdoes not contain proppant. Once the prop-

32

pant-laden slurry reaches the tip of the frac-ture, it bridges off and the fracture can nolonger increase in length. Proppant at thetop and bottom similarly prevent heightgrowth. This situation is called a tip screen-out. Further pumping causes the fracture toballoon or widen. In soft rock, the fracturewidth can increase four to six times withthis technique.

An interesting application of ClearFRACfluids is in the treatment of selective com-pletions. In the spring of 1996, PhillipsPetroleum Company performed two fracpacks in a well in the High Island area inthe Gulf of Mexico. The lower gas sand wascompleted with a crosslinked-guar fluid sys-tem, and the selective zone above was com-pleted with ClearFRAC fluid because thefrac pack was to be left behind pipe untilthe lower zone was depleted. Phillips didnot plan to flow back the fracture immedi-ately for cleanup; hence, a polymer fluidsystem might undergo severe dehydrationand damage the formation over time. TheClearFRAC fluid remains in place andbreaks over time as it contacts the formationgas. When the selective zone is produced atsome point in the future, it should flow backas if it had just been fractured.

Fracturing for Water ControlFrac packing with ClearFRAC fluid hasanother advantage—the capability to frac-ture near but not into water zones. Theleakoff rate of the viscoelastic surfactantfluid helps prevent the fracture from grow-ing up or down, possibly contacting nearbywater zones. In certain applications, a typi-cal guar-fluid frac pack might break downthe formation being treated, allowing thefracture to propagate downward directlyinto an underlying water zone, acceleratingwater production. The predictable leakoffcharacteristics of viscoelastic surfactant flu-ids make them the fluids of choice in theseoperations. Once a tip screenout isachieved, the frac will widen rather thanincrease vertically into the water zone.

Coastal Oil & Gas Corp., like many oper-ators in the Gulf of Mexico, has changed itscompletion practices from gravel packing tofrac packing. Frac packing is the comple-tion of choice for high-permeability wellsthat may have potential sand-control prob-lems, drilling-induced formation damage,fines migration problems, perforation dam-age, or other borehole damage. A typicalfrac pack accesses the formation beyondthese damaged zones. Coastal’s frac-packcompletions have held up over time, requir-ing less remedial work than offset gravel-pack completions.

Coastal has been completing wells withfrac packs at a rate of one or two per monthduring 1997 and has had six wells stimu-lated with ClearFRAC fluid. The reasons forusing ClearFRAC fluid instead of a guar fluidinclude the capability of fracturing withouthitting water zones, a shortened cleanuptime and a cleaner fracture.

Oilfield Review

9000

6000

15

10

5

Pre

ssur

e, p

si

Pum

p ra

te, b

bl/m

in

Time, min

3000

0 600 610 620 630 640 650

Beginning of tip screenout

Treating pressureBottomhole pressurePump rate

■Pumping pressure and rate chart. Following the tip screenout in this frac pack operation,the bottomhole pressure began to increase slightly as the treating pressure decreased,indicating the fracture width was expanding. The final fracture had a half-length of 30 ftand a width of 1 in.

Pad

Fluidvolume, gal

FluidPump rate,

bbl/minStage Proppant,

lbProppant

sizeProppant

concentration, ppa

12 4% surfactant 3500 0 20/40 0

1 12 4% surfactant 500 250 20/40 0.5

2 12 4% surfactant 750 1500 20/40 2

3 12 4% surfactant 1000 4000 20/40 4

4 12 4% surfactant 1000 6000 20/40 6

5 12 4% surfactant 250 2000 20/40 8

Flush 12 2% KCI water 3340 0 20/40 0

■Proppant schedule. This STIMPAC treatment placed 13,750 lbm of 20/40 proppant with7000 gal of 4% ClearFRAC solution followed by 3340 gal of 2% potassium chloride.

In a High Island area well located offshoreLouisiana, Coastal encountered a 36-ft [11-m] gas zone overlying a 50-ft [15-m] waterzone (previous page). Many operatorswould have perforated the upper section ofthe gas zone, performed a gravel-pack com-pletion and flowed the well at reduceddrawdown to stave off water coning.

The well was perforated underbalancedwith tubing-conveyed perforating guns, 13 ft[4 m] above the gas-water contact. Thescreen and packer assembly were thenpicked up and run in the well. The packer

Autumn 1997

was set, and the pipe was pickled. ADataFRAC analysis was performed to deter-mine the optimum pumping scheduledesign based on closure pressure andleakoff coefficients. Pumping time for thetreatment was about half an hour (top).

The pumping operation was successful(above). Based on bottomhole pressuredata and computer simulation results, thefinal frac pack was estimated to have a 30-ft[9-m] half-length and 1.5-in. [4-cm] width.The well was flowed back at moderate rates

that were increased hourly until reachinganticipated production rates about 12 hrlater. The well flowed back normal amountsof sand and load water and appeared toclean up properly as the gas contacted andbroke down the ClearFRAC fluid. The wellwas then shut in as the rig was skidded tothe next slot, and production resumed sev-eral days later. Initial production rates aver-aged 11 MMscf/D [300,000 scm/d] of gasand 300 BOPD [48 m3/d] with only traceamounts of water.

The frac pack successfully avoided thewater zone. A comparison of the productionto offset wells is difficult and potentiallyinaccurate due to the extremely faultednature of the producing reservoir, however.On this well, the frac-pack treatment wasslightly more costly than for a guar fluid. Thetotal cost to drill and complete the well wasmore than $3 million. Coastal felt the costdifference for the fluid was minor, given theability of the ClearFRAC fluid to ensure aclean frac pack above the water zone.

Looking AheadClearFRAC fluid was commercialized onMay 12, 1997. As of November 1997, morethan 400 ClearFRAC jobs had been pumpedin the USA and Canada. At present, use ofClearFRAC fluid is highest in Canada, fol-lowed by the Gulf of Mexico and then theeastern USA.

The two largest hurdles for ClearFRAC fluidto overcome are its upper temperature limi-tation and cost relative to guar fluids. Theupper temperature limit for ClearFRAC fluidsis 200°F, but research is under way toincrease the temperature range.

The higher chemical cost of ClearFRACsurfactant increases the total cost of a frac-ture treatment by some 5 to 20% on smalltreatments, such as frac packs and jobs withless than a few hundred thousand pounds ofproppant placed. Reducing the cost of thesurfactant will increase the population ofwells where the application of ClearFRACtechnology is economically feasible. Toaddress this cost issue, research is also inprogress to increase the efficiency of thecurrent surfactant and to develop lessexpensive second-generation surfactants.

—KR

33

Jacques Bourque Frances TuedorLee TurnerMontrouge, France

Samuel GomersallAberdeen, Scotland

Paul Hughes, Jr.Quito, Ecuador

Robert KleinBP Exploration Operating Co. Ltd.Aberdeen, Scotland

Business Solutions for E&P ThroughIntegrated Project Management

Today, more and more operators are turning to oilfield service

providers to help them plan and execute exploration and

production (E&P) activities. As a result, the business focus for

larger oilfield service companies has shifted from supplying a

traditional package of individual products and services to providing

customized, value-added solutions with a guarantee of technical

integrity and quality of work. Accomplishing this task requires

the merging of project management skills and proper technology.

34 Oilfield Review

George NilsenSanta Fe Energy Resources, Inc. Houston, Texas, USA

David TaylorChauvco Resources InternationalCalgary, Alberta, Canada

DESIGN-EXECUTE-EVALUATE and FIV (Formation Isola-tion Valve) are marks of Schlumberger. MS Project is amark of Microsoft Corporation.

A revolution in the relationships between oiland gas companies and service companies iswell under way, as evidenced by the recentexplosive growth in alliances and partner-ships. Following on the heels of E&P industryrestructuring, current oilfield economics andthis new business relationship environmentare motivating operators to outsource moreto the service sector (see “Changing BusinessRelationships,” page 37).

Obviously, much has changed over thepast decade. Some oil and gas companiesused to own drilling rigs and operate seis-mic vessels—practices that have all but dis-appeared. Former models for conductingbusiness have been scrutinized andreplaced by proactive approaches that arebased on teamwork.

Many oil and gas companies have reorga-nized for greater efficiency and profitabilitywhile, at the same time, redefining corecompetencies they wish to retain—evaluat-ing prospects, selecting the most attractivecandidates, negotiating the most favorable

Autumn 1997

lease terms and managing the long-terminvestment risk associated with field devel-opment. Noncore activities are being con-tracted to the service sector, includinggreater responsibility for field operations andan expanded role in the development andapplication of new technology.

In response, some major service compa-nies have also restructured to meet thisexpanded role and the growing clientdemand for project management expertise.This move has led to the formation of dedi-cated organizations that can manage theintegrated services and technologiesrequired for complex field developments.

As demand for oil and gas continues toincrease, there is a greater focus on newapproaches to field development andimproved management of producing fieldswith new technologies. Extending the life ofexisting reservoirs and increasing ultimateoil and gas recovery are part of the quest forhigher efficiency. Under current economicconditions, the development of evenmarginal fields is a growing requirement.

Today, operators are turning to engineeredintegrated solutions to drill wells and pro-duce hydrocarbon resources more cost-effectively. The scope of work, organizationand location of integrated projects varywidely. At one end of the spectrum, an inde-pendent operator may require a single pro-ject manager and wellsite supervisor from aservice company, for a few months, to coor-dinate the services required to drill shallowonshore wells.

At the other extreme, there are large, fullyintegrated, multiyear project alliances todrill, complete and tie-back multipleextended-reach offshore wells, and deliverthe oil to a tanker for a consortium of majoroperators. Recently awarded contracts go farbeyond well construction, with scopes ofwork that include production and life-of-field operations. The bottom line is that ser-

35

Integratedteam

Servicecompany

Servicecompany

Servicecompany

Servicecompany

Servicecompany

Oilcompany

Traditional

Oilcompany

Servicecompany

Servicecompany

Servicecompany

Oilcompany

Integratedprojectteam

AllianceIntegratedservices

Integratedalliance

■Evolution of the business environment. In a traditional relationship, a service companysupplies products or services to the oil and gas company (left). An alliance is a long-termrelationship between an operator and a service company and there is close cooperationbetween the two to develop shared objectives (left center). An integrated services contractcombines expertise from several service companies to work as a team on one project(right center). In an integrated project management alliance, oil and gas companies allywith service companies to remove barriers that inhibit full cooperation, sharing objectivesand appropriate risk-reward incentives (right). These alignments give the integrated teamaccess to best-in-class technology and the ability to apply solutions early in a project.

1. Chafcouloff S, Michel G, Trice M, Clark G, Cosad Cand Forbes K: “Integrated Services,” Oilfield Review 7,no. 2 (Summer 1995): 11-25.

vice companies must be able to provide thesolutions requested by operators, and mustalso be structured to pull together the inte-grated team necessary for project imple-mentation (right).

The consequences for service companiesare numerous. The major players have anincreasing need for project design and engi-neering resources, a wider scope of exper-tise, and a reliable and efficient integratedquality, health, safety and environmental(QHSE) management system. Oil and gascompanies must have confidence in theability of service companies to deliver cost-effective solutions that are environmentallysound, with a guarantee of technicalintegrity and personal security and safety.

There are several key factors to successfulalliances and partnerships. First, there is aspirit of teamwork and mutual trust estab-lished through alignment, shared benefitsand adoption of risk and reward incentives.Second, there are clearly defined, mutuallyagreed-upon objectives that lead toimproved operational efficiency and pro-ductivity. Finally, to achieve these goals, theoptimal technological solutions for eachproject must be selected and implemented.This new way of doing business has broughtmany players into the arena, including sev-eral of the larger service companies, such asHalliburton Energy Services, Baker-HughesINTEQ, Dresser DDPS and Schlumberger.

The Schlumberger Integrated Project Man-agement (IPM) organization was establishedin 1995.1 This new model of business inte-gration originally began with a focus onwell construction, mainly drilling servicespackages. Today, projects are oftenextended to include well management—completion, production monitoring andmaintenance—finally leading to field devel-opment—reservoir study, developmentplanning, production facilities, drilling andwell completion (next page).

The IPM charter is to provide clients withproject management and well engineeringexpertise and the optimum technology forintegrated projects to jointly create greatervalue in an environmentally responsiblemanner. Defining the added value and bene-fits of an integrated approach is a key issue.Overhead and headcount reduction throughoutsourcing, in many cases, has been theprimary efficiency driver and still may be forsome operators. However, many clientshave discovered that integration has gonebeyond this, bringing closer communicationand cooperation with service companies.The positive outcomes are better access totechnology, teamwork and an alliance spirit,with mutually-agreed objectives.

36

IPM is a business, but it is a business builton processes. These processes are the struc-tural elements that glue the complex organi-zational and technical factors necessary forintegrated services together. At every level,management commitment is the corner-stone of each process. The five key compo-nents that make up the fundamental pro-cesses of this business structure are:

• commitment at all levels• QHSE systems• training• information technology• management tools.

Success requires the integration of a num-ber of functions and tools—an experi-enced, well-trained staff of specialists bothfrom within and outside the service sector,

forward-looking recruiting and trainingpractices, as well as supporting softwareand management systems for QHSE. Merg-ing all these functions from scratch is a difficult task that requires investment andcommitment from the top levels of com-pany management.

IPM teams are a diverse mix of peoplefrom many cultural and oilfield professionalbackgrounds, some with years of industryexperience from oil and gas companies oroilfield service companies. Many are recent

Oilfield Review

Changing Business Relationships

Integrated reservoiroptimization

Integratedwell services

Coiled tubingStimulationConformance fluidsCompletion fluidsProduction servicesSlickline

Dataacquisition

LoggingData services

Integrateddrilling services

CementingDrilling FluidsMWDLWDDirectional drillingMud loggingTestingBits

Rigs

Logistics

Completion

Production optimization

Well intervention services

Data management

Fielddevelopment

services

Wellconstruction

Productionsystem

Productionprojects

Evaluation Exec

utio

n

Design and planning

■The scope of IPM activities. Oil and gas companies may have different strategies andvarying requirements for each project. IPM has developed a flexible approach thataccommodates integrated projects from the simplest to the most complex and covers a variety of domains, including well construction, production projects and field develop-ment services.

engineering graduates from leading universi-ties worldwide, who have successfully com-pleted a rigorous 15 to 18 month company-sponsored postgraduate training program.

In this article, we discuss the roles playedby two of the key components of this busi-ness structure—QHSE and training. How-ever, the other components are stronglylinked. For example, information technology(IT) provides not only the communicationsystems used for project and QHSE manage-

Autumn 1997

ment, but also those for engineer trainingmodules (see “What’s in IT for Us,” page 2).Three integrated project examples illustratehow operators and service companies havesuccessfully combined their experience,expertise and fit-for-purpose technical solu-tions to reduce cost, achieve superior resultsand jointly benefit both parties.

Traditional models for doing business in the oil

and gas industry are no longer acceptable in

today’s streamlined business environment. Take

the drilling sector, for example. Historical rela-

tionships between oil and gas companies and ser-

vice companies—day-rates and price list on one

hand and turnkey or lump-sum payments on the

other—contain an imbalance of financial risk

between the two parties.

In a day-rate relationship, the service provider,

paid strictly on a time basis, may not give ade-

quate attention to efficiency and overall perfor-

mance. Conversely, the operator is tied down with

close supervision, providing step-by-step instruc-

tions to all service contractors and managing a

large number of contracts through cumbersome

paperwork, beginning with the bidding process.

Here, the operator assumes maximum financial

risk, while service company financial exposure

and technical input to the project are minimal.

At the other extreme, a turnkey relationship

relies on paying the service supplier a fixed lump

sum for a given scope of work. Here, the operator

has little or no involvement in day-to-day project

execution. The contractor has an almost unlimited

financial exposure and little—in any case always

capped—potential for increased profits. As a

result, quality may suffer, as can health, safety

and environmental standards, since the contractor

is mainly concerned about minimizing cost. The

service company assumes maximum financial

risk, while there is little operator financial expo-

sure and opportunity for technical input.

Both approaches lead to conflicts of interest and

few incentives for synergy, and as a result, there is

a high probability that quality and the outcome of

the project—the well—will suffer. Neither approach

creates a real motivation to identify and implement

optimal technology for the project.

Today, the industry is moving towards a busi-

ness culture built on communications, quality ser-

vice and organizational structures that benefit both

parties. People are the key resource—the critical

element—for business success. Going beyond tra-

ditional working relationships between operators

and service companies, the highest technical stan-

dards can be maintained while working in an inte-

grated team approach with common objectives.

37

38 Oilfield Review

■Management systems. The Well Engineering Management System (WEMS)strategy follows a three-phase, auditable DESIGN-EXECUTE-EVALUATE process (right). Each phase is implemented in accordance with Quality Management System (QMS) policies and procedures, and each has a formalreview, endorsement and approval process. WEMS is an MS Project softwaretool that is used to plan well proposals, the field basis of design and individ-ual drilling programs, and to provide data archiving and tracking (below).This gives IPM teams progress reports, audit trails for performance, a learning tool for design optimization and finally, an end-of-well report.

Contractors

QHSE Responsibilities in Integrated Projects

Office

Wellsite

QHSE QHSE

OperationsRig

toolpusherDrilling services

wellsite supervisor

Drilling servicesproject manager

Operatorwellsite

supervisor

Rigmanager

Operatordrilling

supervisor

■Typical integrated project lines of responsibility. MultipleQHSE interfaces must be managed by the operator, rig con-tractor, project management company and subcontractors.The operator’s wellsite representative often takes on a qualityassurance and audit role, rather than the traditional role ofday-to-day operations supervision. Ultimate responsibility forkey decisions regarding HSE or technical integrity of the wellbore remains with the operator’s representative or the rig contractor’s offshore manager, depending on the location.The project manager’s wellsite supervisor, usually workingclosely with the rig contractor offshore manager or tool-pusher, is charged with day-to-day operations, includingoverall QHSE management.

25March

3 17 24 31 7 14 21 28 5Start Duration % Done

April10

Ready

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

Project Manager [0.5]

ORGANISE PROJECT

REVIEW OBJECTIVES AND FINALISE W

Well Engineer

Project Manager

Project Manager

Project Manager

Project Manager

WELL PROPOSAL APPROVED BY OPER



Project Manager

AUTHORISE BRIDGING DOCU

DEVELOP BASIS OF

OPERATOR AWARDS CONTRACT

Prepare Project /Well Filling System

ORGANISE PROJECT

Review Contract

Review Scope of Work

Set up Project Organization and Resources

Develop Preliminary Project Schedule, Resources and Responsibilities

Define Project QHSE System /Procedures

Define Project Administration /Procedures

REVIEW OBJECTIVES AND FINALISE WELL PROPOSAL

Enter the Well Proposal into the WEMS

Determine Permitting/SMS Requirements

Determine Site Survey Requirements

(Meeting) Review Object. and Well Proposal with Operator

Update Preliminary Project Schedule, Resources and Responsibilities

WELL PROPOSAL APPROVED BY OPERATOR

DETERMINE STATUS OF AFE APPROVAL

MEETING TO AGREE ON POLICIES W/OPERATOR & DRILL. CONTR.

(BRIDGE.MPD) - Develop Bridging Docs

AUTHORISE BRIDGING DOCUMENT

DEVELOP BASIS OF DESIGN

Perform Offset Data Analysis

Identify Rig Requirements or Specifications

Casing / Hole size Basis of Design

Directional Basis of Design

Mud Basis of Design

Cementing Basis of Design

Bit Program Basis of Design

Riser OR Conductor Basis of Design

Completion Program Basis of Design - if required

1 Mar '96

1 Mar '96

4 Mar '96

6 Mar '96

6 Mar '96

6 Mar '96

6 Mar '96

6 Mar '96

7 Mar '96

8 Mar '96

11 Mar '96

11 Mar '96

11 Mar '96

11 Mar '96

19 Mar '96

19 Mar '96

20 Mar '96

20 Mar '96

20 Mar '96

21 Mar '96

2 Apr '96

20 Mar '96

20 Mar '96

25 Mar '96

25 Mar '96

25 Mar '96

28 Mar '96

28 Mar '96

28 Mar '96

25 Mar '96

1 Apr '96

119d

1d

1d

3.38d

3h

3h

1h

0.5d

1d

1d

6.75d

6d

0.5d

1d

0.5d

2h

0d

0.5d

1d

5d

0.5d

18d

3d

2d

3d

3d

2d

2d

2d

5d

1d

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

0%

Dril

ling

Well evaluation Casing

Well trajectoryC

ementing

timing, specifications, policiesand

procedures

DESIGN: Process maps, plan

templates,project

programs, diagrams, responsibilities, procedure

library,

devi

atio

nfro

mpl

ans

and

pred

ictions

safety reports and process improvement suggestionsEXECUTE: Daily reports, log books, policies

PerformanceFindingsActionsResults

Project objectives

andprocedures, performance curves, auditable

trails

,

EVA

LUA

TE:

End-

of-w

ell r

epor

ts,perf

ormance databases,

audi

ts,p

roce

ssim

prov

emen

t sug

gestions and

WEMS Display

educodntcpaotv

jQaa

tifooamleha

affWimat

mfgwutttgds

QHSE Management System

Managementcommitment

QHSE policies and procedures

WEMS andcompetence

assurance

Project specific bridging documents and QHSE plans

ContractorQMS/SMS P

roje

ct le

vel

Wor

ldw

ide

ContractorQMS/SMS

ContractorQMS/SMS

ContractorQMS/SMS

ContractorQMS/SMS

Drilling contractorQMS/SMS

QMS

SHIELD

QMS

Local legislation

andregulation

Client QMS and SMS

scope ofwork

Subcontractors QMS/SMS

■QHSE management structure. QHSE management in IPM projects uses a two-tieredstructure. The upper tier is generic and applies to all projects worldwide; it consists of fourmajor components—management commitment, QHSE policies and procedures, a qualitymanagement system (QMS) and SHIELD, the IPM safety management system (SMS). Thelower tier contains elements that provide support for the components on the top tier. Eachelement is tailor-made to fit individual project requirements and consists of project-spe-cific bridging documents to link components from the top tier to individual QHSE man-agement systems, policies and procedures of the operator, drilling contractor, and IPMcontractors and subcontractors.

2. Duncan E, Gervais I, Le Moign Y, Pangarkar S, StibbsB, McMorran P, Nordquist E, Pittman T, Schindler Hand Scott P: “Quality in Drilling Operations,” Oilfield


Building an Integrated Project Management OrganizationPlanning and organization are critical tobuilding project management structure.Early involvement of all parties assurescommitment to the process. IPM has theability to draw on experts from around theworld to create the optimal organizationfor assisting the operator in designing andimplementing a project while, at the sametime, bringing in other best-in-class servicesuppliers and third-party vendors (see“Andrew and Cyrus: Teamwork and Tech-nology in Action,” page 40).

While IPM is an independent, stand-alonecompany within Schlumberger Oilfield Ser-vices, it serves the interest of clients first andforemost. Independence of action and thequality of solutions provided are uppermostin this new culture.2 For example, whilethere are more than 50 rigs involved inactive IPM projects, only 10 of them arefrom Sedco Forex.

Autumn 1997

Review 8, no. 1 (Spring 1996): 20-35.

Quality, Health, Safety and EnvironmentRegardless of the scope, oil and gas compa-nies expect an integrated project manage-ment organization to administer the QHSEaspects of a project with the same commit-ment, resources and long-term vision thatthey themselves would as the operator.Management systems, such as the WellEngineering Management System (WEMS)—a tool used to ensure operations integrity—contain modules that can be applied toquality and HSE risk assessment and man-agement (previous page, top).

To understand the challenge of managingQHSE, particularly for large projects, it isnecessary to look at the changing nature ofthe organization and management interfacesthat are involved, especially those at well-sites. Integrated projects vary widely, and notwo have exactly the same requirements.However, there is at least one common fac-tor—operators have fewer contractors tomanage directly, allowing them to focusattention on their core competencies (see“Remboué, Chauvco’s First Gabon Permit,Onstream,” page 44).

This contrasts dramatically with pastxperience when operators contractedirectly with as many as 20 to 30 individ-al service companies during a drillingampaign. With an integrated team, this isften reduced to a maximum of four—rilling services, the drilling rig, civil engi-eering and a logistics and transportationeam (previous page, bottom). This majorhange drives the project management com-any to develop an integrated QHSE man-gement system under the umbrella of theperator’s systems, and to link this directly tohe systems of the drilling contractor andarious subcontractors.Given these complex interfaces, the pro-

ect management company cannot organizeHSE on an informal, ad hoc basis. Rather, formal, well-defined and integratedpproach is required (left). Uppermost in the QHSE management sys-

em is management commitment andnvolvement. The most common cause ofatalities and major lost-time incidents in theil and gas industry is not a lack of policiesr procedures, but a lack of commitmentnd leadership throughout all levels of theanagement chain. Unless line managers

ead by example and everyone followsstablished policies and procedures, theyave no value. QHSE must be as importants any other business objective. Balance is crucial and can be achieved by variety of management tools, includingrequent verbal and written communicationsrom executive levels to field managers.

ithin Schlumberger, QHSE is the first topicn all general management and informationeetings. Yearly QHSE objectives are part of

n overall management-by-objectives sys-em for all managers.For wellsite supervisors, an effective QHSEanagement system should offer two main

eatures. The first is prescriptive rules anduidelines that tell them what to do andhat is expected of them in most critical sit-ations. The second is freedom to exercise

heir expertise and judgment when a situa-ion arises that is not specifically covered byhe rules. Again, balance is the key. Thereater the volume of policy and proce-ures, the less frequently they are read,tudied and adhered to at the wellsite.

39

Andrew and Cyrus: Teamwork and Technology in Action

Forties field

Andrew field platform Cyrus field satellite wells9.7 km [6 miles] from Andrew

Brae field

Oil exportline

Gas export line

2 subseawellheads

UK

Nor

way

■Andrew and Cyrus fields.

Courtesy of BP

In the North Sea, the Andrew field—a 1974 discov-

ery—and the Cyrus field are being brought on

stream.1 Over the past 20 years, numerous devel-

opment plans failed to meet required economic

hurdles. However, recent technology and organiza-

tional advances have provided the spark to jump

start these and other fields previously considered

marginal. Success of the Andrew and Cyrus

endeavor required an alliance of companies with

jointly developed objectives, and a new working

culture. The Andrew alliance consists of opera-

tor—BP Exploration; Schlumberger IPM for well

management and data acquisition; Baker Hughes

INTEQ for integrated drilling services; Transocean

mobile rigs and Santa Fe platform rigs.

The Andrew field, northeast of the Forties field,

produces from a Paleocene sandstone, saturated-

oil reservoir in the UK sector (right). Estimated

Andrew field reserves are 112 million bbl

[17.8 million m3]. Cyrus, a smaller subsea

field—fewer than 24 million bbl [3.8 million m3]

of undersaturated oil—is tied back to Andrew and

located just to the northeast. The advent of proven

horizontal drilling and well completion technology

significantly improved economics allowing the

number of Andrew wells to be reduced from 18 to

10 while increasing output from 45 to 58 thousand

BOPD [7.1 to 9.2 thousand m3/d]. Two subsea hori-

zontal producers replaced two earlier Cyrus hori-

zontal appraisal wells close to the oil-water con-

tact that rapidly produced high water cuts.

Improvements in platform and facility construc-

tion, including onshore commissioning, increased

lift capacity, and reduced offshore manpower

requirements also helped reduce project cost.

The Andrew venture was conceived as a best-

in-class project to take a previously marginal

field from uneconomic to commercial success.

Achieving this goal required exceptional perfor-

mance from an integrated alliance team,

addressing organizational culture and precon-

ceived behaviors in every aspect of the project—

from corporate alignment to individual responsi-

bilities. The well engineering approach was to

maximize productivity, accelerate first oil, maxi-

mize reserves through optimum well design and

minimize costs. A risk-reward, gain-share struc-

ture—designed for well operations—was imple-

mented. Aligned business objectives focused

40

alliance members on cost and well productivity.

Contractor alliance members were selected by an

interactive process that emphasized technical

ability, performance and cost.

A two-part contractual framework was estab-

lished for the Andrews Well Engineering

Alliance—Works Contract and an Alliance Agree-

ment. The Works Contract between BP and each

alliance member covers standard terms and

divides finances into direct cost, profit and over-

head. This arrangement is profit-, not revenue-

driven and acts as an incentive to improve perfor-

mance without losing income.

The Alliance Agreement, common to all mem-

ber companies, defines the alliance, including

risk-reward payment and minimum performance

standards—a structure for potential incentive pay-

ments. Savings, measured against established

budgets, are shared by the alliance members

(next page, top). Rewards are proportional to each

party’s risk.

Technological Solutions

The Andrew and Cyrus fields are among the first

all-horizontal developments by BP. Using horizon-

tal well technology to reduce the total number of

Andrew wells meant that economic success

depended on long-term performance over the life

of the field. To meet this requirement, the Andrew

Well Engineering Alliance defined four minimum

performance standards. Wellbore trajectories

must be within the specified horizontal target; the

gas cap must be effectively isolated; data acquisi-

Oilfield Review

Contractorreward

BP andpartnersaving

BP andpartnerrisk

Contractorrisk cap

Alliancetarget cost

Alliancecost estimate

Alliance budget

Finalcost

Gain

Loss

■Alliance gain-shar-ing mechanism. Costsavings are shared bythe alliance in a risk-reward paymentarrangement. Rewardsare proportional toeach party’s risk andoverruns are alsoshared up to a maxi-mum cap.

Completiondelivery

Raw hydrocarbondelivery

Hole deliveryInitiating

hydrocarbondelivery

Maintaininghydrocarbon

delivery

Well constructionservices

Rig and associatedservices

ImproveproductivityCompletions

Start-upMaintain

well output

■From line manage-ment to a process-based organization. It was apparent earlyon that a differentorganizationalapproach would berequired to realize the full benefits from an integratedAndrews Well Engi-neering Alliance. The end result is anorganization in whichindividuals have proprietorship over aprocess—projecttask—with completeresponsibility fordelivering results.

1. Gomersall S, Klein B, Clark G, Sneddon I and Simpson M:“Andrew Well Engineering Alliance: A New Industry Model,”paper SPE 36872, presented at the SPE EuropeanPetroleum Conference, Milan, Italy, October 22-24, 1996.

2. Mason NE and Gomersall SD: “Andrew/Cyrus HorizontalWell Completions,” paper SPE 38183, presented at the SPE Formation Damage Conference, The Hague, TheNetherlands, June 2-3, 1997.

3. Kusaka K, Patel D, Gomersall S, Mason J and Doughty P:“Underbalance Perforation in Long Horizontal Wells in the Andrew Field,” paper OTC 8532, presented at the Off-shore Technology Conference, Houston, Texas, USA, May5-8, 1997.

4. Patel D, Kusaka K, Mason J and Gomersall: “The Develop-ment and Application of the Formation Isolation Valve,”paper OMC 6697, presented at the Offshore MediterraneanConference and Exhibition, Ravenna, Italy, March 19-21,1997. For a review of FIV tool and other technology developmentthrough partnerships:Edmonds P: “Linking Solutions to Problems,” OilfieldReview 8, no. 4 (Winter 1996): 4-17.

tion must be acceptable for reservoir manage-

ment; and at least 75% of the productive horizontal

intervals must contribute to flow.

Maximizing production from horizontal wellbore

sections in the Andrew field required several

design decisions.2 Horizontal well cementing,

although difficult and often problematic, has

improved. Cemented liner completions were,

therefore, selected to provide flexibility in future

well diagnosis, intervention and recompletion, to

allow selective, oriented perforating for minimiz-

ing sand production and ensuring uniform inflow,

and to prevent formation damage during comple-

tion and well suspension.3 But to avoid impaired

productivity, it is best to perforate underbalanced.

This advantage, however, is lost if wells must be

killed to remove perforating guns.

Perforating gun deployment with coiled tubing

units allows underbalanced perforating and gun

removal, but requires multiple runs, increasing

job time and cost and resulting in subsequent bal-

anced perforating operations. These limitations

led to development of the FIV Formation Isolation

Valve tool at the Schlumberger Perforating and

Testing Center, Rosharon, Texas, USA. The FIV

valve is a 4.562-in., fullbore mechanical ball valve

which allows long horizontal sections to be perfo-

rated underbalanced in a single trip using a

hydraulic workover unit to recover the guns without

killing the well.4

Team Building

From the beginning, significant time was invested

to establish a team of individuals from each

alliance company. Technical information and

reservoir knowledge were shared, leading to an

increased awareness of project issues and uncer-

tainties. The result was a closer working relation-

ship between geoscience and operations. At every

level, the team was encouraged to be flexible and

innovative—organizationally and technically—

breaking down barriers and roadblocks to progress.

Only through technology, combined with cultural

and behavioral changes, could exceptional perfor-

mance be achieved.

An advanced process-based organization with

clear accountability was developed and imple-

mented (above). In this organization, individuals

are responsible for—and own—defined pro-

cesses. Teamwork and results are emphasized

with authority resting with individual process own-

ers who are empowered to make decisions in the

best interests of the field. Day-to-day decisions

Autumn 1997

are made by the offshore staff with onshore man-

agement support. This approach requires better

communications than in traditional organizations,

but quality decisions—based on the best available

input and a more complete understanding—are

worth the extra effort.

Well planning involved development of a Well

Basis of Design and Well Operations Manual

located on a computer server so experience could

be quickly captured and used for the future. Expert

peers from outside the asset well engineering

team review well design to identify opportunities

and risks. The net result is the potential to deliver

improved performance.

Performance during template drilling for Andrew

and Cyrus wells was compared to similar horizontal

wells drilled by other operators. The first two tem-

plate wells ranked in the top quarter and the third

Andrew template well was the best in class among

30 other wells in the North Sea that were evaluated.

—MET

41

IPM Management Strategy

• Leadership with commitment

• Organizations, roles and responsibilities

• Polices, procedures and information

• Evaluation of scope of work and risk management

• Design, planning and engineering process

• Execution and monitoring procedure

• Project performance and management and continuous improvement

■IPM quality management system. TheIPM quality management system strategyis based on seven key elements derivedfrom the Exploration and ProductionForum Guidelines for the Development andApplication of HSE Management Systems.

Recruiting

Training

"SMART"competenceassessment

Placement andperformancemanagement

Project

Staffingrequisition

Oilfield service

companies

Universities

Oil and gascompanies

■Meeting the needs for a competent staff. The competence assurance system starts withstaffing needs and requisitions from the project, which leads to the recruiting process.New recruits are trained in a company-sponsored and structured postgraduate engineertraining program. The competence assessment process assures that engineer traineeslearn all the basic skills and tools needed to work effectively in project teams. Through-out an engineer’s career, reassessment and continual training are integral to placementand performance management.

IPM has prioritized all QHSE-critical oper-ations from a risk perspective to developand implement policies and procedures. Forthe remainder, professional expertise andpersonal judgment are necessary, foundedon a competence-assurance program that isdiscussed later.

A quality management system—Quality asdefined by IPM is a process, not just a busi-ness. The processes developed by IPM for itsQuality Management System (QMS) arebased on the The Exploration and Produc-tion Forum report, “Guidelines for theDevelopment and Application of HSE Man-agement Systems.”3 Quality managementlinks all aspects of planning, control andimprovement into a continuous system forensuring conformance to established stan-dards (above left). Operators demand thatthe oilfield project management companyguarantee operations integrity, which meansdemonstrating that potentially significanttechnical as well as HSE risks that may arisehave been identified, and that a qualitymanagement system is in place to reducerisks as much as possible.

For a quality management system to besuccessful, it must be simple and welldefined so that it can be effectively commu-nicated to all participants through a com-prehensive awareness program and, as aresult, become all-pervasive within an orga-

42

nization. To maintain state-of-the-art techni-cal leadership, the entire international com-munity of IPM technical and managementemployees have access, through globalcommunications and information technol-ogy, to a special Well Engineering Manage-ment System (WEMS) bulletin board, allow-ing open electronic discussion of problemsand sharing of lessons that are learned (nextpage, top). Experts within the Schlumbergerorganization regularly monitor these bul-letin boards and become valued contribu-tors, analyzing and resolving problems, andlending assistance when needed.

An essential part of quality management isthe Competence Assurance System, whichenables project management personnel tomeet expanding operator staffing demands(above). The process includes five steps—staff requisition, recruiting, training, compe-tence assessment and placement. Projectrequirements lead to staff requisitions and

recruiting. Some recruits come from thepool of experienced engineers in the oilindustry and other Schlumberger OilfieldServices companies, but most engineers arerecent university graduates.

After recruitment, the next phase of thecompetence assurance process is training,followed by competence assessment. Thefinal phase in this process is placement andperformance management. For positionsthat are critical to HSE, a competenceassessment is performed before anemployee is placed on the job. Thereafter,performance on the job is managedthroughout the year, starting with a jobdescription that defines the function andresponsibilities of a position, reporting rela-tionships and standards of competence. Per-formance is measured against establishedobjectives known by the acronym SMART—specific, measurable, attainable, realisticand time-bounded. Performance reviews aredone quarterly; formal written appraisals aredone annually.

Oilfield Review

Autumn 1997

■Global communications. A key component of continual improvement in qual-ity management is the ability to communicate information up and down themanagement chain. This is a common challenge facing all companies, espe-cially those working in multinational environments. All IPM project offices andmost wellsites are connected through a global intranet. This network providesfull e-mail and Internet capabilities with 24-hour support. The QHSE functionmaintains a home page on the WorldWide Web where the HSE managementsystem, policies and procedures, HSE alerts and information bulletins areposted and continually updated.

Assessment and c

Operations

Engineering, d

Risk assessme

Policies, procedures

Organization, rol

Contractor and

Communication and

Personnel and co

Management comm

4

3

2

1

10

9

8

7

6

5

SH

■Ten elements of SHIELD—the IPM HSEmanagement system. Like many othercompanies, IPM considers health, safetyand the environment (HSE) to be as impor-tant as other business objectives. It is anIPM objective to be completely proactiveso that injury, ill health, and property orenvironmental damage arising from pro-ject activities can be prevented. Undereach of the ten elements of the HSE man-agement system, specific responsibilitiesfor each job category are defined toensure success in meeting this objective.The value of any HSE management systemis not the document itself, but in its appli-cation on a project. Regular audits areconducted to ensure compliance.

An HSE management system—All IPM pro-jects use a safety management system(SMS), called SHIELD, with a frameworkbased on ten major elements found in mostcommonly used management and auditprotocols (below).4 The goal of this system isto provide project managers and wellsitesupervisors with a set of decision-makingtools that make HSE an integral part of day-to-day work and reduce risk as much as rea-sonably practical, while simultaneouslyallowing them to make decisions based ontheir own judgment and experience. Thesetools include an emergency plan and a risk-management system.

43

ontinuous improvement

and monitoring

esign and planning

nt and management

and information control

es and responsiblities

vendor management

management of change

mpetence assurance

itment and client focus

IELD

3. The Exploration and Production Forum “Guidelines for the Development and Application of HSE Manage-ment Systems,” Report No. 6.36/210, July 1994.

4. The management systems and audit protocols forquality and HSE surveyed include: the UK Health andSafety Executive’s Guidelines for Health and SafetyManagement Systems, the Exploration and Pro-duction Forum’s Guidelines for the development andapplication of HSE Management Systems, the HSEmanagement systems from several major operators,the integrated QSE Management System of the Hiber-nia Alliance, DNV’s International Rating System forSafety, Environmental and Quality, ISO 9002, theIMO’s International Safety Management Code, theMalcolm Baldrige Quality Award and the EuropeanQuality Award.


Remboué, Chauvco’s First Gabon Permit, Onstream

Gabon

Cameroon

EquatorialGuinea

Republic of Congo

Libreville

ATLANTIC

OCEAN

RambouéPermit

Port Gentil

0 100 miles

0 161 km

■Remboué location inGabon. Chauvco’s Rem-boué Permit straddlesthe equator along theRemboué River.

Chauvco Resources chose Schlumberger IPM as

project manager for development of the Remboué

Permit, a 224,000-acre [907 km2] tract on the

equator in the heart of the West African Republic

of Gabon’s tropical rainforest (right). The IPM

alliance was charged with handling logistics,

well engineering and field development drilling.

As a result, a single-source solution was devised

to bring Remboué onstream in an efficient, timely

and cost-effective manner that combined thor-

ough planning with advanced drilling and com-

pletion technology and a strong QHSE manage-

ment system.

Remboué was first explored in the early 1990s

by British Gas, which drilled eight wells and shot

600 km [372 miles] of 2D seismic surveys before

discovering oil in 1991. Indicated reserves were

10 to 15 million bbl [1.6 to 2.4 million m3]. How-

ever, despite the discovery and identification of 23

additional prospects, for business purposes, BG

relinquished the undeveloped permit in 1993.

In July 1996, however, because of new technol-

ogy and a government willingness to renegotiate

contract terms, Chauvco (now Chauvco Resources

International) signed a production-sharing contract

with the Gabonese government to re-explore the

block, develop Remboué field and begin produc-

tion of the estimated 10 million bbl of recoverable

oil reserves.

At the time, Schlumberger IPM was providing

integrated drilling services for Shell in neighbor-

ing Rabi Permit, drilling horizontal wells in the

same early Cretaceous stratigraphy as Chauvco’s

Remboué reservoir. Drawing on Schlumberger IPM

experience in the area, Chauvco chose IPM as the

project manager, and a team immediately began

to assemble an alliance of companies best suited

for this integrated project.

44

Schlumberger’s Well Repair Center, already in

the country with an SK15 light drilling rig, was

selected as the drilling contractor and has handled

the logistics of bringing equipment up the Rem-

boué River by barge from Port Gentil since the

beginning of the project. It has also repaired old

roads and wellsites, built new ones, and pur-

chased most of the material and equipment

required, including wellheads, casing and down-

hole pumps for project startup.

Of the other alliance members: Anadrill pro-

vides directional drilling, MWD expertise and the

fishing tools; Security DBS provides the bits and

does the coring; Dowell provides drilling and com-

pletion and cementing fluids; Weatherford runs the

casing; mud logging is done by Geoservices; and

Schlumberger Wireline & Testing performs wire-

line logging and testing.

IPM planned the drilling program with Chauvco

to spiral outward from the discovery wells Abre 1

and Abre 2. A vertical well, ReRe 1, was first

drilled to a total depth of 440 m [1444 ft] into

unconsolidated reservoir sands, and the remaining

11 wells were drilled horizontally to the Gamba

target zone beneath salt strata.

Oilfield Review

Autumn 1997

Exposure cataloging

Risk probabilityanalysis

Risk Management Process

•Avoid

•Transfer •Manage

•Reduce

Risk monitoring

Projectrisk

register

Risk evaluation review

Risk impactanalysis

Elements that make something a risk: • Uncertainty • Undesirable effect

Risk identification

Mitigation strategy

■The risk management process. Two elements that make something a risk are uncer-tainty and undesirable effects. During all operations, qualitative risk management is acontinuous monitoring and adjustment process. Quantitative risk assessment generallyrequires the assistance of outside specialists, and is set forth in the operator’s scope of work.

Emergency response plan—Every projecthas an integrated suite of emergencyresponse plans, which build upward fromthe project to regional offices and headquar-ters. Plans are updated continually, anddrills are held regularly in close coordina-tion with operator personnel.

Risk management system—Risk affects bothHSE and the operational integrity of the pro-ject, again demonstrating the tight linkbetween quality and HSE (above). Everyproject also uses the IPM HSE risk manage-ment process. The WEMS tool contains arisk-management module that can beadapted to each project.

Corporate-level QHSE management sys-tems cannot possibly take into accountworldwide differences in legislation and reg-ulations, the scope of the work of variousoperators in different countries and on mul-tiple projects, the huge number of differentquality and HSE management systems ofoperators, drilling contractors and integratedproject management company contractorsand subcontractors, plus countless regionalpolicies and procedures. The answer to thischallenge is a specific bridging document atthe project level. This document addressesall HSE-critical interfaces, and requires sig-nificant upfront time and attention from all

David Taylor, Chauvco’s Manager of Business

Development for West Africa, says that the com-

pany contracted with Schlumberger IPM for the

project because it wanted to hit the ground running

in an area where Chauvco had never worked before

and needed the expertise that Schlumberger

brought to the alliance to speed development of

the Remboué Permit.

“Although we had some logistical problems

with our first well, and we didn’t have all the

equipment we needed on site, the Chauvco/IPM

team worked together and fixed that situation.

Then, everything went well after that and we could

really see the benefit of IPM on subsequent wells.

Cost efficiency came into play once we were under

way, and we had a team that worked smoothly,

with good interface between Chauvco management

in Calgary, Alberta, Canada and operations in

Gabon. We’re also very proud of the fact that we

had no lost-time accidents.”

Phase one of the Remboué development pro-

gram has successfully concluded, and phase two is

already under way, with the IPM team planning the

next series of wellsites, designing the wells and

plotting out the overall campaign. Drilling was

expected to commence in mid-November 1997 on

five to eight new wells.

As a result of the benchmark success IPM has

achieved at Remboué, Chauvco, a newcomer to

Gabon, was able to establish itself as an operator

within 13 months of signing its production-sharing

contract. With one vertical well and 11 horizontal

wells, it is now producing more than 4000 B/D [600

m3/d] of 34° API oil from the Remboué Permit and

plans to increase production to over 7000 BOPD

[1100 m3/d] by year-end. The income this new-

found oil is generating for the country is a welcome

addition, and because of this achievement, Chau-

vco has been awarded three additional petroleum

concessions by the Gabonese government.

—DG

45

• Roles and responsibilities - operator representative- project management company- wellsite supervisor- toolpusher

• A common interface for variousemergency response plans

• Well control measures - shallow gas- H2S- shut-in method

• Permit to work system

• Management of change

• Pollution incident response and other environmental arrangements

• Fitness of purpose of equipment and itsmodification, maintenance and spare parts

• Health arrangements- vaccinations- medical examination requirements- medical evacuation procedures

• Security arrangements and policies- drugs- alcohol- firearms

Project-Bridging Document Elements

■Critical elements of a project-specificbridging document. Many project andregional factors have a large impact onthe project management company’sefforts to manage QHSE at the projectand wellsite level. To address these complex needs, the project managementcompany must create and supervise aproject-specific QHSE bridging document,usually in close cooperation with theoperator and drilling contractor. The elements shown here are common to allprojects. United Kingdom Offshore Operators’ Association (UKOOA) guidelines,for example, provide a good template tocustomize for individual projects.

■Active involvement. IPM trainee Salil Pande (left) conducts a drilling safety meeting in

groups involved (above). Constant effort isrequired over the course of a project toensure compliance. This is an essential stepto ensure risk reduction, but one that paysbig dividends in the final QHSE perfor-mance of the project (see “ReducingDrilling Costs and HSE Risks in Ecuador,”page 48).

TrainingTo meet the need for project managementresources, a graduate engineer trainingprogram provides comprehensive class-room, on-the-job and interactive training.In-house courses currently cover well engi-neering and production engineering. In thefuture, a program on reservoir optimizationwill be added.

46

Most new IPM project engineers arerecent university graduates and require thecomplete graduate engineer training pro-gram that lasts 15 to 18 months (next page,left). Others are experienced industry engi-neers who are familiar with the technologyof well engineering, production or reservoiroptimization, but may need to learn thecultural and business aspects of the servicecompany role in project management. Stillothers, with service company experience,may understand the service company cul-ture, but need to learn many new tools todevelop the broad range of skills requiredfor well engineering and production.

Engineers recruited directly from universi-ties undergo a structured training programdesigned to ensure that they acquire theskills, knowledge and attitudes necessary toperform as well or production engineers andwellsite supervisors (below).

IPM uses a variety of training sources andmedia, ranging from technical courses atSchlumberger training centers to specialcourses run by universities and consultants.The full graduate engineer training programis divided into two major phases.

The first phase is an introduction to thepetroleum industry. Trainees gain knowledgeabout industry problems and objectives, andlearn the basic tools needed for the job.Graduates without a petroleum engineeringbackground attend a four-week supplemen-tary petroleum engineering course taught ata recognized university—such as the Col-orado School of Mines in the USA, ImperialCollege in London or Ecole NationaleSupérieure du Pétrole et des Moteurs àCombustion Interne (ENSPM) in France.Those with a petroleum engineering back-ground go directly to one of the training

Oman. Trainees participate in daily activities a

centers where they are introduced to com-pany culture, the service company role inintegrated projects and the essential ele-ments of IPM—the charter, organization andhow IPM differs from other service suppliers(next page, right).

Trainees then enter a six-month, on-the-job training program for well constructionand rig-site operations. They acquire expe-rience with critical wellsite operations andspend time working directly with func-tional operators or crew members at thewellsite, including everyone from drillingsupervisors to roughnecks handlingdrillpipe. Engineer trainees are alsoexposed to workover operations.

The major objective is not only to learnwhat is done, but why it is done, in order toacquire a basic understanding of the entireprocess of well construction. During thisperiod, trainees also assist service compa-nies at the rig site. For example, they maywork with wireline engineers rigging up log-ging tools, calibrating measurements andacquiring and correlating data in the loggingunit. They also work with mud engineersmixing chemicals and mud additives andlearning about the impact of mud systemson drilling and logging measurements.

Before entering the second phase of train-ing, trainees return to the IPM training cen-ter for a well construction project review,and quality and HSE management systemtraining. They concentrate on teamworkand interpersonal skills, as well as synthe-sizing what has been learned through prob-lem, analysis and solution exercises.

Phase two is also an introduction to theIPM business environment. In this phase—half-way through the training schedule—trainees make career decisions. Depending

Oilfield Review

s a team with the rig crew.

Schlumberger week 2

IPM Graduate Engineer Training Program

Rig induction course

Schlumberger week 1

Petroleum engineeringsupplemental course

Introduction to drilling and workover operations

Introduction to the drilling rig

Introduction to Schlumberger

Introduction to the oil and gas industry

Phase 2

Introduction to cultural,communication andleadership skills

Project management

Wellsite supervision

Production engineeringWell engineering

Phase 1

Production engineeringcareer path

Well engineeringcareer path

OFS product line

■IPM graduate engineer training program. The first half of the training program exposesnew engineers to the petroleum industry and helps develop the basic skills needed for aproject. In the second half of the program, engineers participate in projects and developthe skills they learned in phase one.

IPM Training Strategy

• Focus on orientation and QHSE

• Accelerate development of competent people

• Tailor training to personnel andskills inventory

• Utilize training programs from various Schlumberger product lines

■IPM training strategy. The purpose of thetraining program is to teach project engi-neers basic technical, business and com-munication skills, along with a set of corevalues necessary to meet all project objec-tives with appropriate engineering andtechnological solutions.


on educational background, personal pref-erences and available projects, each traineechooses one of two major programs—wellengineering or production engineering.

In these programs, trainees continue todevelop the skills that were learned in theintroductory phase. These courses focus oninteractive case studies using IT tools, such

Autumn 1997

as computer-based simulators and CD-ROM self-study training modules. Theemphasis is on adding value to projects.For example, trainees work on specific pro-ject activities like well design, casingdesign or hydraulics.

Finally, the trainee is assigned to a specificproject team. During this phase, the engi-

neer works much like a physician doing aresidency at a hospital—the doctor helpspatients while learning and sharpening hisor her skills. Similarly, an engineer-in-train-ing adds value to the client’s project, per-forming studies that address problems,implementing the results and contributing tooverall project performance. The engineerbenefits from personal development andincreased confidence and gains job satisfac-tion from contributing to a project.

Every engineer is assigned an IPM mentor.Senior staff members are recruited tobecome mentors and trained to appreciatethe importance of this role in developingnew engineers. Mentors are evaluated onhow well their trainees perform.

At the end of the graduate engineer train-ing program, trainees have acquired thebasic knowledge inventory, skills and atti-tudes to work effectively in project teams. Atthis point, each trainee is reassessed and, ifdeemed competent and qualified, certifiedto perform to required standards.

To meet the demand for trained projectmanagement personnel, the program isexpected to grow dramatically over the next2 years (page 49).

The next four to five years of an engineer’scareer, spent working on projects, are stillconsidered development years. Here, theengineer continues professional develop-ment through courses designed to improvetechnical and on-the-job skills. After thisperiod, some project engineers become pro-ject managers or specialize in areas such asreservoir characterization, intelligent com-pletions or directional drilling.

47

Reducing Drilling Costs and HSE Risks in Ecuador

Ecuador

Peru

0 100 miles

0 161 km

San Lorenzo Colombia

Esmeraldas

PACIFIC

OCEANIbarra

QuitoNueva Loja

Manta

Block-IIDrillsite

Guayaquil

■Ecuador-Colombiaborder region and Block11 location. PetroleraSanta Fe’s concessionlies on the Ecuadorian-Colombian border eastof the Andes mountains.

Petrolera Santa Fe (Ecuador) Ltd., a subsidiary of

Santa Fe Energy Resources, Inc., was confronted

with the problem of drilling two exploratory wells

in the rugged Sucumbios Province of the Ecuado-

rian Oriente. The drillsite is east of the Andes

mountains in a region of dense jungle and diffi-

cult terrain some 30 km [18.6 miles] west of

Lago Agrio (Sour Lake) on the Rio San Miguel

border with Colombia (right).The principal target zones for the two wells,

Cristal 1 and Betano 1, were in the Hollin and

NAPO Cretaceous formations, with target depths in

hard Chapiza precreataceous formations at

approximately 10,100 ft [3080 m] and 10,700 ft

[3260 m], respectively. To meet the dual chal-

lenges of drilling this difficult prospect and doing

so with a minimum of security and environmental

exposure required an integrated team approach

with close communication and cooperation

between the operator and project management

company. For this reason, Santa Fe awarded the

project management contract to Schlumberger

IPM, which had proposed a single-source solution

to the drilling project that brought with it applica-

tion of the latest technology and extensive experi-

ence not only in Ecuador, but in this specific area.

Three main problems had to be assessed and

overcome during the course of the project. First

was reduction of operational nonproductive time

(NPT), second was assurance of drillsite security

and rig personnel safety, and third was confor-

mance to local environmental regulations. Opera-

tional NPT was shown to be related to running cas-

ing, bits and hydraulics used for each section of

the hole, cementing and general rig downtime.

An alliance of Petrolera Santa Fe, Schlumberger

companies and Intairdril Pool Ltd. formed the inte-

grated project team, with IPM as project leader

responsible for equipment logistics, carrying out

civil works and preparing the wellsite. Intairdril

Pool provided Rig 226 for the project, with drilling

fluids and cementing from Dowell. Anadrill pro-

vided drilling services; mud logging was done by

Geoservices; and wireline logging and testing were

performed by Schlumberger Wireline & Testing.

The drilling program was planned by IPM in col-

laboration with Santa Fe to commence with the

48

Cristal 1 well followed by the Betano 1 well. A

third well is also scheduled.

The integrated alliance began work in October

1996 by first establishing a Safety Management

Team (SMT) and several Quality Management

Teams (QMT). Members of each team were

selected from employees of the alliance compa-

nies based on their individual knowledge and

experience in handling specific problems encoun-

tered in the Oriente region.

Careful planning and risk management tech-

niques were employed by the SMT. Both urban

and rural security methods specific to the area

were applied, and there was close coordination

between the systems IPM put into place and those

established by Santa Fe. By the end of the project,

not a single minute was lost due to security-

related disruptions.

A second team was established to address envi-

ronmental issues. Communications were of partic-

ular importance. Communication was maintained

with the Ecuadorian environmental and regulatory

organizations, and issues related to potential dis-

charges and any local sensitivities were discussed

and resolved prior to project initiation. As a conse-

quence, no incidents occurred.

Additional technical QMTs were formed to

address the main causes of nonproductive time

revealed during the drilling of the discovery well,

Cristal 1. Although NPT was kept to just 4.8%

while drilling this first well, the team felt improve-

ments could still be made by addressing four

specific activities with dedicated QMTs: a casing

running team charged with finding more efficient

ways to run casing strings; a bit selection and

hydraulics team with responsibility to review bit

types, verify changes in hydraulics, and propose

ways to reduce per-foot-drilling cost; a cementing

team to recommend ways of maintaining high

cementing standards and to create a better cash

Oilfield Review

Autumn 1997

Another development program trains pro-fessional project managers. This accreditedprogram leads to certification by the Interna-tional Project Management Association inZurich, Switzerland. The needs of each pro-ject manager trainee are assessed andinstruction on project management systemsis provided using commercial business appli-cations and software products.

The End ResultThe impact of modern technology on reduc-ing exploration and production costs isundisputed. Major challenges for today’soften complex field developments are select-ing and applying the proper technology.Building on a new business culture allowsoil and gas companies and service compa-nies to combine expertise and identify opti-mum technologies for project delivery—effi-cient planning, drilling, completion and wellmanagement. IPM conducts business in amanner that ensures all operations are per-formed safely with high, consistent stan-dards, while protecting the environment.

Independence of action assures that theoptimal fit-for-purpose technology is usedfor each project. Access to Schlumbergertechnology and worldwide experience plusthe technology of other service companiesand third parties, provides the flexibility torespond to individual project needs withbest-in-class solutions.

A safety management system bridges theHSE policies of project participants to pro-vide a single HSE management and report-ing system at the project level. A qualitymanagement system covers all technicalphases of planning and executing a project.

1996 1997

IPM Engine

200

150

100

50

0

Num

ber

of e

ngin

eers

Trainees completingphase 1Trainees completingphase 2

Total number oftrainees in program

■Growth of the graduate engineer

As projects become more demanding, theyrequire a team of people with clear rolesand responsibilities. Each project’s needs arematched with the right people trained forthe right job. Teamwork is a natural out-come in an open, trusting relationshipbetween operators and service suppliers.This teamwork is aided by rapid and effec-tive communications that allow efficientsharing of data, expertise and ownership ofproblems and solutions. These operationalbenefits enhance the return on investmentsmade in a field over its productive life.

The Outlook Former adversarial interactions betweenoperators and service companies are beingcast aside and replaced by a true spirit ofcooperation, alignment and communica-tion. This is evidenced by the three exam-ples in this article and the growing numberof alliances and partnership projects. Today,in most oil-producing regions, operatorsand service companies are workingtogether cooperatively as never before.Communication bridges are being built toconnect the two sectors in ways that, if con-tinued, will provide end-to-end solutionsfor the industry.

Process improvements have helped opti-mize service delivery and resource require-ments, with a corresponding drop in welldesign, construction and intervention costs.By aligning with operator goals, the processis being streamlined, reducing cycle timeand increasing overall efficiency, while atthe same time, improving health, safety andenvironment protection at the wellsite. Thefuture growth of the energy industrydepends directly on the successful evolutionof these relationships. —RCH

ers in Training

value added for Santa Fe while maintaining oper-

ational integrity; and a rig downtime team to

ensure minimum NPT for the drilling rig and

associated equipment.

Individual IPM teams conferred constantly, find-

ing ways to improve services by coming together

every two weeks in free-forum meetings with the

operator. These discussions were focused on con-

tinual improvement. As a result of the team effort,

Santa Fe’s second well, Betano 1, reached its tar-

get depth in 31 days—four days earlier than sched-

uled—with a record NPT of just 1.8%.

“The team approach worked very well,” said

George Nilsen, Santa Fe’s Operations Manager.

“Whether security, operations or environmental

issues, it worked well. Security was a definite

worry, but we integrated our security team with the

IPM team and the drilling operations, and didn’t

have a problem. Likewise, IPM set up environmen-

tal procedures to be sure everything was done in

an environmentally correct way, then followed up

to be sure there was no drainage or any kind of dis-

charges, and we had none. Essentially, the IPM

people were the drilling operations staff in our

office; it was a relationship that was really a team

working together to do the best job on the wells.

We saved a lot of our costs because we didn’t

have to build a staff and construct infrastructure

in the field to handle all the logistics of supporting

the drilling operation. The bottom line was a

significant cost savings.”

As a result of this team approach, Petrolera

Santa Fe’s drilling costs on the two wells were

reduced by more than $1 million (US).

Schlumberger IPM earned an incentive bonus on

the Betano well, and because of the positive

results achieved with this project, was awarded a

contract for the third well scheduled for spring

1998, and others in neighboring Colombia and

Venezuela. —DG

49

1998 1999

training program.

Documents

What's in IT for Us? - Schlumberger/media/Files/resources/oilfield_review/ors97/... · Schlumberger Wireline & Testing, Aberdeen, Scotland. ClientLink, InterACT, MAXIS (Multitask