Incose Sweden Model Management01292011 V8

Model based Engineering inModel-based Engineering in Medical Device Development –pIssues & SolutionsINCOSE SwedenLinkoepingApril 4 2011April 4, 2011

© Siemens AG 2011. All rights reserved.

Contents

Goals Goals

Brief look on Siemens Healthcare

Business challenges

Model-based Engineering: Issues & SolutionsModel based Engineering: Issues & Solutions

Recommendations

Further Information

Page 2 Arnold Rudorfer, Siemens Healthcare Imaging & Therapy SYNGO© Siemens AG 2011. All rights reserved.

Contents

Goals Goals


Business challenges

Model-based Engineering: Issues & SolutionsModel based Engineering: Issues & Solutions

Results and Summary

Further Information


G l f thi T lkGoals of this Talk

Disc ss the e periences Discuss the experiences using engineering models (in different phases of product development of a next generation imaging platform)

Identify and share needs and Identify and share needs and requirements with colleagues from diverse industries to d d tadvance product

development tooling


Contents

Goals Goals


Business challenges

Model-based Engineering: Issues &Model based Engineering: Issues & Solutions

Results and Summaryy

Further Information


Siemens HealthcareTHE I t t d H lth CTHE Integrated Healthcare Company

in-vivo diagnostics (imaging)

X-Ray ComputedTomography

MagneticResonance

MolecularImaging

Ultrasound Oncology

in vitro diagnostics (laboratory systems)

Tomography Resonance Imaging

syngo.via

in-vitro diagnostics (laboratory systems)

Immunodiagnostics Clinical ChemistryNucleid AcidTesting

Hematology Lab AutomationUrinAnalysis

Near PatientTesting


Example: syngo.mCT Oncology:O l W kfl T D t ti St iOncology Workflow: Tumor Detection, Staging

Automated Case Preparation Multi-Modality AccessDisease-Oriented ReadingAutomated Case Preparation Multi Modality AccessDisease Oriented Reading

Automated bone and bloodpoolremoval

Automated sorting of images, etc.

Image fusion for CT, PET, and MR images, etc.

Preferred layout applied Automated lesion segmentation Automatic loading of prior exams, etc.

Page 7 Arnold Rudorfer, Siemens Healthcare Imaging & Therapy SYNGO© Siemens AG 2011. All rights reserved.Courtesy of University Hospital of Munich-Grosshadern/ Munich, Germany © Siemens AG 2011. All rights reserved.

syngo.mCT Oncology is cleared in the U.S. with syngo.PET&CT Oncology.

Contents

Goals

Contents

Goals Goals

Brief look on Siemens and Vector

Goals


Project syngo.via

Business challenges

Business Challenges

Model Management: Issues & SolutionsBusiness challenges

Lean Requirements Engineering

Model Management: Issues & Solutions

Results and Summary

Results and Summary

Further Information

Further Information


B i Ch ll i M di l D i I d tBusiness Challenges in Medical Device Industry

Industry regulations

Cycle time down

Industry regulations

Cycle time down

ComplexityComplexity

Need to industrialize

product

Costs down Complexity growth

Need to industrialize

product

Costs down Complexity growth

product development

product development

Focus of this talk

Application Life-cycleManagement Tools

Model-basedEngineering

PlatformsApplication Life-cycle

Management ToolsModel-basedEngineering

Globally Integrated

Platforms


Globally IntegratedClinical Workflows

Disclaimer:Disclaimer:

The content discussed in this presentation needs

to be considered as work in progressto be considered as work in progress.


Contents

Goals

Contents

Goals Goals


Goals


Project syngo.via

Business challenges

SYNGO products

Business challengesBusiness challenges


Business challenges

Model-based Engineering: Issues & Solutions

Results and Summary

Further Information

Results and Summary

Further Information


i P d tsyngo.via Product

Project Summarysyngo.via: Next generation imaging software covering the g g gentire reading process

Project data: > 5,000 single product requirements Several million lines of code C++/C#Several hundred developers in many locationsCli i l li ti fClinical applications for

Radiology, PACS, X-Ray, CT, PET/SPECT, Oncology, Particle Therapy and MR


Therapy and MR.

I d t I d P i P i tIndustry Issues and Pain Points

Pain Points Business ImpactPain Points Business Impact Product structure

intransparent, domain model partially incomplete

Technology-driven product platform, no link to businessdriversOpaque relationship between problem & solution spacemodel partially incomplete Opaque relationship between problem & solution spaceRe-scoping sessions w/ customers on basis of many specs.

Ambiguity and lack of accuracy of specifications

Textual-based specifications are subject to interpretationTextual use case descriptions work only for smaller projectsy p p y p jNatural language subject to interpretation, inconsistent,

incomplete Controlling architectural Redundancy of architecture components due to lack of

complexity understanding of problem- and solution spaceBusiness needs not consistently linked to featuresToo much variability in software architecture

Lack of V&V efficiency Test specifications in natural language are mostlyexecuted in a manual way onlyTest cases manually createdThousands of pages of requirements


Thousands of pages of requirements

Pain point 1: Product Structure Intransparent, D i M d l P ti ll I l tDomain Model Partially Incomplete

S l tiSolutions:

A. Requirements Engineering Meta-model

Selected issues: Opaque relations between

problem- & solution space

B. Feature Model

problem & solution space Re-scoping sessions w/

customer on basis of many engineering specs.


g g p

Solution A: Requirements Engineering Meta Model B fit & T l R i t (1)– Benefits & Tool Requirements (1)

Characteristics:Problem Space Characteristics: Provides needed artifacts, their

attributes and relationships to each other

DB

Business PlanMarket Analysis,

Stakeholder Request

Problem Space

other. Using meta-model, it is also possible to

prescribe the way how / which data are captured .

MRS

SW Platform Roadmap, Business Case, etc.

input for

input for

Benefits: Lean artefact infrastructure, no

MRS

Market Requirement

SW Feature AnalysisUse Case Specifications,

Load Profiles,Concept Papers

analyzedby

redundancy Guidance for engineering tasks

with structured inputSWRSSW Req.

(Problem)

Concept Papers

Established link between business drivers, requirements, design and test

SW Req.(Solution)

Specification/Structural Element Content Mapping or


Document Structural Element Content pp gInput

Solution A: Requirements Engineering Meta ModelC t (2)– Concepts (2)

Feature A Feature represents a characteristic of a product which provides a business value and supports purchase decisions […]. A Feature structure requirements in a meaningful way and is no specification in itself (only “container”).

Stakeholder Request

Feature Feature A Feature Model is a hierarchical "tree" to describe the structure, dependencies, commonalities and variabilities of Features within a product or product line (e.g. SW Platform, Finished Medical Device).

SW Feature SW Feature

Market

Finished Medical Device).

Requirements specify the Feature. They represent the functionality already implemented within the product and functionality that is planned for future versions. Depending on the project phase requirements are less or more detailed (e gRequirement

SW Requirement

project phase requirements are less or more detailed (e.g. Market Req., SW Req.).

Stakeholder Requests are wishes towards an existing or future product. Stakeholder Requests (SR) are gathered from various sources (e.g. end customer, business units) and may have an impact on different levels and phases of the product lifecycle .


Solution A: Requirements Engineering Meta ModelC t (3)– Concepts (3)

Problem Space Sol tion SpaceDB


Roadmap Business Case

Stakeholder Request

Problem Space Solution Space

Roadmap, Business Case, Version Goal

input for input for

MRS

Market R i t

SW FeatureAnalysis

Use Case Specifications, analyzedby

SWRS

Requirement

SW Req. CDSArchitectural

Load Profiles,Concept Papers, etc.

Architecture

by

(Problem)

SW Req.(Solution)

DesignDescription

Build. Blockrealizes


Solution A: Requirements Engineering Meta Model C t (4)– Concepts (4)




Stakeholder Request



input for input for

Example: Feature - „Demo of Images“

Market Requirement: As a Clinician, I want to mark images to support

MRS

Market R i t

SW FeatureAnalysis


want to mark images to support presentations for a medical conference.

SWRS

Requirement



Architecture

by

(Problem)

SW Req.(Solution)

DesignDescription



Solution A: Requirements Engineering Meta Model C t (5)– Concepts (5)




Stakeholder Request



input for input for Use Case: „Demo of Images“ with activity diagram to derive software

i tMRS

Market R i t

SW FeatureAnalysis


requirements.

SWRS

Requirement



Architecture

by

(Problem)

SW Req.(Solution)

DesignDescription



S l ti B F t M d lSolution B: Feature Model

Highest LevelHighest Level Characteristics: Hierarchical tree to describe the

structure, dependencies and liticommonalities

Lays out the basics for variant management and impact analysis

Graphical View

Benefits: Higher level abstraction of grouping

of requirements into sellable units: From 5,000 product requirements to 800+ features (factor ~ 6)

Visual domain model for healthcare

Hi hi l Vi

workflows (tree & graphical) Reduction in time to understand

aspects of the system


Hierarchical View

S l ti B F t M d l & It R l ti (1)Solution B: Feature Model & Its Relations (1)

Application Use CasesStakeholder RequestsH IStakeholder RequestsHow can I manage huge amounts ofRequirements?

How do I scope a product version efficiently?

Architecture Model


S l ti B F t M d l & It R l ti (2)Solution B: Feature Model & Its Relations (2)

Application Use CasesStakeholder RequestsStakeholder Requests

Architecture ModelHow do I manage stakeholder requests from different businesses mostbusinesses most effectively?


Pain point 2: Ambiguity and Lack of Accuracy of S ifi tiSpecifications

Level Requts. Manually Manual test

Issues: Textual use case descriptions work only

for smaller projects < ~ 100

Level Requts.Object

Manually Embedded graphs case creation

Document Requt. 1(text)

Picture / diagram

requirements Natural language subject to

interpretation, usually inconsistent, incomplete with inccorrect version (and

Features

(text)

Requt. 2(text)

Picture / diagram

p (conflicting)

Root causes:T t l i t i i d t

Paragraph Requt. 3(text)

Picture / diagram

Textual requirements engineering do not scale for platform projects

Missing versioning No direct access to single requirements

Solutions:

C Application Use Cases Lack of product structure Inconsistently executed change

management process

C. Application Use Cases


Solution C: Graphical Modeling of Clinical W kfl (1)

Characteristics:

Workflows (1)

Characteristics: Used to describe clinical workflows that

consist of a collection of steps in a defined sequence together with accompanyingsequence together with accompanying specification of pre-/post-conditions, business rules, performance aspects, etc.

B fitBenefits: Increase expressiveness of clinical

workflows to describe dynamic behaviors f li i l kflof clinical workflows

Early analysis of stakeholder requests from customers Improved impact analysis of change Improved impact analysis of change

requests Joint modeling sessions to describe the

needs from the customer‘s point of view


needs from the customer s point of view

Solution C: Graphical Modeling of Clinical W kfl (2)Workflows (2)


Architecture ModelHow can I analyze a feature or a stakeholder request?


P i i t 3 C t lli A hit t l C l itPain point 3: Controlling Architectural Complexity

Solutions:

D Architecture Model Mapping

Selected issues: Business needs not consistently

linked to features/ requirements; D. Architecture Model Mappinglinked to features/ requirements; dependencies between features not easily visible

Too much variability in software


architecture

S l ti D A hit t M d l M iSolution D: Architecture Model Mapping

Characteristics:F Feature Model

Identifies links between features and their implementation

Explicit modeling of variability in the architecture

F F F

SWF SWF SWF SWF SWF SWF SWF

Benefits:

SWF SWF SWF SWF SWF SWF SWF

S Architecture Model

Architectural decisions motivated by features and product-line variability

Enabling reduction of architectural complexity

SS SS SS

C C C C C C

Enabling reduction of architectural complexity

Support impact analysis for (de-) scoping sessions

Early identification of architectural risks Early identification of architectural risks

Improved accuracy of early effort estimates

Reduction of number of scoping sessions


Reduction of number of scoping sessions

S l ti D A hit t M d l M iSolution D: Architecture Model Mapping


Architecture Model

Which component pimplements which feature?


P i i t 4 N d t I V&V Effi iPain point 4: Need to Increase V&V Efficiency

Solutions:

E M d l b d S T

Issues: Test cases partially manually

generated from prose use casesC l ti f t t t t l E. Model-based System Test Cycle times for system test too long

Compliance requirements QSR 21 CFR 820, ISO 13485:2003, EU MDD 93/42 and other medical device


93/ a d ot e ed ca de cesoftware regulations.

S l ti E M d l b d S t T t (2)Solution E: Model-based System Test (2)

B fitBenefits: Early identification of requirements

defects through validation by testers Effort reduction for test (cycle time,

cost ~ -30%) and increase of test coverage (*) Decrease number of defects

Model-based testing is highly structured, reproducible and efficientp Increase reuse of development

artifacts Quicker impact analysis by parsing

model for late requirements changes


(*) Based on experience from other projects

S l ti E M d l b d S t T tSolution E: Model-based System Test


Architecture Model

Which application use case is covereduse case is covered by which test case?


F R i t E i iFocus o Requirements Engineering …

Sync between toolsSlM tProblem Space

(Features & Requirements)Solution Space

(Architectural Building Blocks)SlaveMaster

MRSSW Feature

Market

EnterpriseArchitect

Architectural Build. Block Enterprise

Architectrealizes

Market Requirement

SWRS

SW Req

Caliber RM SW Feature CDSArchitectural Build. Block

SW Req.(Problem)

SW Req.(Solution) Caliber RM

DesignDescription

realizes*

Together


Contents

Goals

Contents

Goals Goals


Goals


Project syngo.via

Business challenges

SYNGO products



Business challenges


Results and Summary

Further Information

Results and Summary

Further Information


Major Changes to Development ApproachM d l b d E i iModel-based Engineering

Requirements Engineering Meta Requirements Engineering Meta-Model

Feature Model

Graphical Modeling of Clinical Workflows

Architecture Model Mapping

Model-based System Test


Note: syngo.via

B fit f M d l b d E i iBenefits from Model-based Engineering


ResultsB i C C t ib ti f LBusiness Case – Contribution of Levers

Description of Benefit Distribution ofDescription of Benefit Distribution ofBenefits

uct on

Tracing with less effort (based on feature model‘s nature no verticals)

25%

Prod

uD

efni

ti feature model s nature, no verticals)

More effective stakeholder management (less scopingsessions, reduction in review times and effort)

Proj

ect

Plan

ning Transparency and easy overview on product functionality 23%

Des

ign Reduction of product complexity (transparent modeling of

product lines and product variability)7%

DTe

st

More effective Testing 45%

Easier bug-fixing


Source: OREA Business Case, February 2011 (to be expected)

ResultsB i C NPV P j tiBusiness Case – NPV Projection

B fit li d Benefits are realized across engineering workflows: Product Definition, Project Planning, Design and Testand Test

Break-even reached in 2nd year

Hints for other RE practitioners: Hints for other RE practitioners: Business case can only be carried out

with a complete technical concept

Sensitivity analysis done on risksshows that the effort is worthwhileeven when everything goes wrong.

Business case will vary depending on lccal circumstances


Source: OREA Business Case, February 2011 (to be expcetd)

K T k A f M d l b d E i iKey Take-Aways for Model-based Engineering

Modeling competence of prod ct managers needs to be b ilt p stepModeling competence of product managers needs to be built up step-by-step thereby facilitating acceptance of new approach

Seamless model-driven engineering is only partially tool-supported;Seamless model driven engineering is only partially tool supported; the biggest gap remains in requirements engineering.

Tool vendors need to stronger leverage the experience of leading development organizations and uptake it into technology roadmaps.

Acceptance of model-based engineering is a huge organizational h t d l 10% f i ti hchange management endeavor, only 10% of organizations have

already gained practical experience.

Continuous assessment and verification of business benefits for Continuous assessment and verification of business benefits for model-based engineering is a must to maintain sponsorship from management.


Th k f tt ti !Thank you for your attention!

Page 39 Arnold Rudorfer, Siemens Healthcare Imaging & Therapy SYNGO© Siemens AG 2011. All rights reserved.© Siemens AG 2011. All rights reserved.

Contents

Goals

Contents

Goals Goals


Goals


Project syngo.via

Business challenges

SYNGO products



Business challenges


Results and Summary

Further Information

Results and Summary

Further Information


R fReferences

US Food & Drug Administration Design Control Guidance for Medical Device Manufacturers; March 11 1997US Food & Drug Administration, Design Control Guidance for Medical Device Manufacturers; March 11, 1997

US Food & Drug Administration, Quality System Regulation,; January 1, 1997, http://www.fda.org/cdrh/qsr/01qsreg.html

Brian Berenbach, Daniel Paulish, Arnold Rudorfer, Juergen Kazmeier, Software Systems Requirements Engineering; Mc-Graw Hill 2009; http://www.mhprofessional.com/product.php?isbn=0071605479

Renate Loeffler: Formal Scenario-based Requirements Specification and Test Case Generation in Healthcare Applications, Diplomarbeit, Univ. Paderborn, 10/2009

Siemens Healthcare Imaging & Therapy SYNGO: Optimized Requirements Engineering Approach, Siemens Healthcare, 2010

Siemens Healthcare Imaging & Therapy SYNGO: Microsoft Team Foundation Server Business Case May 2010Siemens Healthcare Imaging & Therapy SYNGO: Microsoft Team Foundation Server Business Case, May 2010

Medical Device & Diagnostic Industry 101, Carey Smoak, Roche Molecular Systems Inc., Pleasonton, CA, Pharma SUG 2010, Paper 1B01

Arnold Rudorfer, Christof Ebert: Lean Requirements Engineering in Medical Systems, MedConf 2010, Munich, Germany, October 14 2010; http://2010 medconf de/downloads/abstracts2010/T2 T3 V1 vector siemens pdf14, 2010; http://2010.medconf.de/downloads/abstracts2010/T2_T3_V1_vector_siemens.pdf

Arnold Rudorfer, Christof Ebert: Quality Requirements Engineering in in Medical Systems, OOP11, Munich, Germany, January 27, 2011 http://www.sigs.de/download/oop_2011/downloads/files/Do2-3_Ebert_Rudorfer_Update_QualityRE_OOP2011.pdf

Arnold Rudorfer: Model-based Engineering in Medical Device Development – Issues & Solutions, INCOSE Workshop, Phoenix, AZ, USA, January 31, 2011, http://www.omgwiki.org/MBSE/lib/exe/fetch.php?media=mbse:2011iw_incose_modelmanagement_rudorfer.pdf

Siemens Healthcare Imaging & Therapy SYNGO: Optimized Requirements Engineering Approach Business Case, February 2011


syngoTh C t l I H b I H lth W kflThe Central Image Hub In Healthcare Workflows

More patient exams in less time*

Sound diagnosesin less time*

Effi i t i S th d f tEfficient image creation, usage, archiving, and sharing

Smooth and fast collaboration

syngo. It’s all about you.


syngo. It s all about you.*Results may vary. Data on file.

Arnold RudorferDirector Software Initiative and Process ImprovementProcess ImprovementSiemens AG Healthcare Sector

Hartmannstrasse 16D-91052 Erlangen

Phone: +49 9131 – 84 2299Fax: +49 9131 – 84 8691Mobile: +49 174 1537825

Email:[email protected]


[email protected]

Documents

Incose Sweden Model Management01292011 V8