23
Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Embed Size (px)

Citation preview

Page 1: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen

Part 1

Sponsored by:

Page 2: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Chapters Chapter 1 Making the Case for and DefiningSustainability, Social Responsibility and Environmental ResponsibilityChapter 2 Conveying and Reporting a Mission and Vision of Financial, Environmental and Social ResponsibilityChapter 3 The Local – Global Three Bottom Lines: ISO

9000, 14000 and 26000Chapter 4 Social and Environmental Measures Chapter 5 Resources, Finance and Return on Responsible InvestmentChapter 6 FESUP - Financial, Environmental and

Social Unity Projects: research, statistics and continuous improvement

Chapter 7 Sustainable Commercial and Industrial Plant Operations Chapter 8 Responsible Lean Logistics Chapter 9 A Sustainable Economy-----------------------------------------------------------------------------Appendix A Basic quantitative analysis Appendix B Environmental and social responsibility survey Appendix C Heat Literacy: what every manager should know

about heat energy

Available at Amazon, Kindle, Nook, B&N and ASQ

Page 3: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Last Session Check Off

Page 4: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

WHY Improved with Big Data ?

• Save Operating Costs • Occupant Satisfaction• Intelligent Operations and Maintenance• Project Justification• Capital Budgeting • Depletion• Pollution• Earth’s Temperature• CO2

• Population • Migration• Sustainability• Environmentally Responsibility

Page 5: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

The Biggest of Big Data Normal Variability

1471013161922252831343740434649525558616467707376798285889194971001031061091121151181211241271301331361391421451481511541571601631661691721751781811841871901931961992022052082112142172202232262292322352382412442472502532562592622652682712742772802832862892922952983013043073103133163193223253283313343373403433463493523553583613643673703733763793823853883913943974004034064094124154184214244274304334364394424454484514544574604634664694724754784814844874904934964995025055085115145175205235265295325355385415445475505535565595625655685715745775805835865895925955986016046076106136166196226256286316346376406436466496526556586616646676706736766796826856886916946977007037067097127157187217247277307337367397427457487517547577607637667697727757787817847877907937967998028058088118148178208238268298328358388418448478508538568598628658688718748778808838868898928958989019049079109139169199229259289319349379409439469499529559589619649679709739769799829859889919949971000100310061009101210151018102110241027103010331036103910421045104810511054105710601063106610691072107510781081108410871090109310960.000

50.000

100.000

150.000

200.000

250.000

300.000

350.000

400.000

CO2 over a million years (blended cores)

Time in thousands of years starting in 1958 and working back - annual observations

CO

2 (p

pmv)

Page 6: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Page 7: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Page 8: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011/13)

Big Data: What is it?No fine line between small, medium and big data

• Large Volume• Sample size

• High Velocity • Streaming data• High Speed Response to Streams

– Specific value turns into action– Proportion in variation across a distribution

• High Variety – Number of variables – Multiple sources - internal/external – Non-uniform time series – Incomplete – Complexity – Nonlinearity

Page 9: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Large Volume: Sample Size

• Population vs. Sample: the craft and mechanics of modeling and differentiating changed!

• Inferential statistics turn into factual statistics – This is difficult for scientists to digest – Requires a new research paradigm– P-values become less important and sometimes useless – Coefficients become facts – R-square becomes more valuable – Sample size becomes population – t-test significance level becomes meaningless – Design Of Experiments becomes more important

Page 10: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

High Velocity

• Streaming data/data in motion• High Speed Response to Streams

– Specific value turns into action– Proportion in variation across a distribution

• The proliferation of real-time, web-based data acquisition systems combined with more sophisticated hand-held devices means you get on board or get out of the way

Page 11: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

High Variety

Number of variables • Combining multiple sources - internal/external • Non-uniform time series

• Cleaning (culling) data is an essential skill

• Incomplete data • Missing data techniques

• Complexity • Nonlinearity

– Dropdown menu items in statistical software are linear– Model building is one of the most sought after skills in multiple

industries

Page 12: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Where do you get it?

N

S

EW

Natural Capital

Organizational Capital

Human Capital

Manufactured Capital

Renewable & Alternative Distributed

Energy

Conservation

Restoration

Communicate

Building

Educate

Train

Equipment

Mission & Vision,

Social System

Product Supply Networks

Current Resource Supplier Networks

Warehouse

Retail

Customer

Big Data Analysis Predictive Statistical Modeling,

Hypothesis Testing, Linear and Nonlinear Programming

Lean, Six Sigma ISO

Benchmarks

Reverse Logistics

Carrying Capacity

Energy Efficiency

Technology

Technology

Optimize

Energy Systems

Disassembly

Page 13: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Intelligent Internal Systems• Btu/product • Btu/person• Btu/DD• Watts/sq/ft• Watts/product• Water/other

unit • Innovate

N

S

EW

Natural Capital

Organizational Capital

Human Capital

Manufactured Capital

Renewable & Alternative Distributed

Energy

Conservation

Restoration

Communicate

Building

Educate

Train

Equipment

Mission & Vision,

Social System

Product Supply Networks

Current Resource Supplier Networks

Warehouse

Retail

Customer

Big Data Analysis Predictive Statistical Modeling,

Hypothesis Testing, Linear and Nonlinear Programming

Lean, Six Sigma ISO

Benchmarks

Reverse Logistics

Carrying Capacity

Energy Efficiency

Technology

Technology

Optimize

Energy Systems

Disassembly

Page 14: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Intelligent External Systems• Utility Pricing

Structures• Industry

Standards • Weather • Technological

Developments • Demographics• Politics • Suppliers • Innovate

N

S

EW

Natural Capital

Organizational Capital

Human Capital

Manufactured Capital

Renewable & Alternative Distributed

Energy

Conservation

Restoration

Communicate

Building

Educate

Train

Equipment

Mission & Vision,

Social System

Product Supply Networks

Current Resource Supplier Networks

Warehouse

Retail

Customer

Big Data Analysis Predictive Statistical Modeling,

Hypothesis Testing, Linear and Nonlinear Programming

Lean, Six Sigma ISO

Benchmarks

Reverse Logistics

Carrying Capacity

Energy Efficiency

Technology

Technology

Optimize

Energy Systems

Disassembly

Page 15: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

15

Page 16: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:
Page 17: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Industry Specific Micro/Internal & Macro/Aggregate

Big Data Grows in Volume, Speed and Variety

As time passes, more data is generated by multiple facilities, in multiple locations under multiple conditions

– Access control – Energy management systems– Computerized maintenance

management systems – Asset management Systems – Camera systems – Fire life safety systems– Utility systems – Smart meters – Expert systems

– Elevator escalator systems – Human resource systems– Power & distribution

management– Switch gear– Emergency/standby power– Power Quality– Intrusion Detection – Lighting Systems

Page 18: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

What do you do with it? • Measure• Descriptive distributions – actual behavior • Benchmark – industry standards and comparative facilities • Test for significant differences

– T-tests (paired and single) • Build relational models

– Correlations – Regression

• Displays of quantitative information – – figures– tables

• Make Decisions!!!

Page 19: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Measure, Measure, Measure• Commissioned construction projects• Number of energy and other resource audits • Number of employees with environmental training• Management attention to environmental issues• LEED certified buildings• LEED accredited professionals on staff• Clearly articulated vision of sustainability • Number of green vehicles • Percentage of “green” office space• Sustainability committee• Senior managers with environmental

responsibilities • Number of functions with environmental

responsibilities• Sustainability education opportunities• Matching funds for energy grants and incentives • Investments in cleaner technologies ($)• Number of water efficiency projects• Number of facilities registered as a LEED project

• Percentage of products undergoing life-cycle analysis• Reduce emissions (percentage reduction)• Energy conservation plans• Climate action plans • Safety training programs (hours)• Number of employees hired from high unemployment

(target) neighborhoods• Environmental accounting systems in place • ISO 14001 certification (number of facilities)• Number of suppliers who are considering ISO 26000• Number of employees who contribute to drafting regulation

for the industry• Number of employees who contribute to drafting

international standards• Number of employees who publish in the areas of

environmental responsibility • Number of employees with financial incentives linked to

environmental goals• Number of sustainable sites• The number of innovations in operation and upgrades in

sustainable technologies• Monitoring results of indoor environment• Quantity of materials & resources used in the manufacturing

process

Page 20: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Descriptive(getting to know your internal and external data)

• Totals • Averages (means, medians, modes)• Frequency distributions • Minimums and maximums • Run charts • Histograms • Range • Scatter plots • Variance • Standard Deviation • Other non-inferential statistics and displays of quantitative

information

Page 21: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Source: Joseph J Jacobsen (2011)

Page 22: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Smart Systems

• Smart grid technology has the potential to reduce energy use by up to 50 percent. For example, a distributed generation (DG) microturbine with combined heat and power (CHP) can achieve an 88% efficiency rating when optimized (Swedish, et.al, 2004). Compare this to a 38% efficient coal fired power plant and we have a gain of 50% in efficiency.

Page 23: Data Rich–Analysis Poor: Big Data Webinar with Dr. Joe Jacobsen Part 1 Sponsored by:

Big Data or Big Science: Part 2

Join Eagle Technology and Dr. Joe Jacobsen for the second webinar in a series of three “Big Data or Big Science” webinars

Tuesday, May 12, 20151:00 PM CDT

Register at: https://attendee.gotowebinar.com/register/

6434301729939622657