How to succeed @ SPDM

International SPDM Conference 2017 | 11-14 June | Stockholm | Sweden

How to succeed @ SPDM

Mark Norris CEng [email protected]

+44 77 12 52 75 12


Agenda

• Industrial Situation and SDM Today• Challenges deploying SDM• A methodology for success with SPDM• Conclusion


The Data Management Overhead is rising inexorably

• Capacity to run simulations is doubling every 18 months, or faster

• Number of files per simulation is increasing due to:– increasing complexity of product– Increasing complexity of simulation– Increased granularity of data for faster set up and simpler iteration

– More system level simulations

• Analysts are spending more time manually managing data and have less time left for engineering

• “We are doing too many experiments and don’t have time to critically evaluate the results”

NAFEMS European SPDM Conference 2015

CELEBRATING 15 YEARS OF SDMIN PRODUCTION IN MUNICH

BMW announced that they had passed the milestone of 2,000,000 simulation runs in SDM


• SPDM 1.0 - customisation and custom extension of SPDM & PLM platforms, SPDM applications– 2000 to today, SPDM lags other PLM domains

• SPDM 2.0 - Configurable SPDM solutions– 2012 onwards, but few vendors, finite list of use cases– Enables Alignment Workshops and Agile POCs

• SPDM 3.0 : the management of virtual experimentation across the virtual enterprise– MOSSEC to be developed as a STEP standard


Key Points for SDM1.0

• It’s a software development program– Significantly more effort than SDM2.0

• Engage with M&S teams, key users to understand requirements

• Develop a value proposition for Analysts and the Enterprise

• Team with a software developer who masters xDM, understands CAE & can architect SDM solution – hard to find


Key Points for SDM2.0

• It’s NOT a software development program– Faster, cheaper, lower risk

• Easier engagement with M&S teams, key users–Alignment Workshops

• Can & should demonstrate feasibility – to Key Users for Endorsement and Engagement– To Management to justify investment

• Virtuous circle starting with low hanging fruit


0

40

80

120

160

200

2000 2002 2004 2006 2008 2010 2012

• SPDM can raise productivity through lean, effective processes and process automation • SPDM can provide a global platform for remote working• SPDM can accelerate Analyst development through access to defined & proven methods• SPDM can enable industrialisation and democratisation of Simulation Methods8

Sim

ulat

ion

C

apac

ity

Simulation Compute Capacity:Moore’s Law

Expert Analyst Capacity: Educate-PhD, Train, Develop Expertise,…

Indu

stria

lise/

Democratise

Raise

Produ

ctivity

Work Re

motely

© 2013 theSDMconsultancy

Develop An

alysts


SPDM deployments today

• SPDM Visionaries:– Successful and productive, increasing engineering throughput– Expanding deployments– Taking SDM to the Enterprise: SDM3.0

• Specialist SDM vendors are deploying faster based on SPDM2.0 Configuration-based approaches

• Early adopters of newer solutions are struggling to achieve:– Deployment– Business Value – User acceptance


CHALLENGES DEPLOYING SDM


Why is SPDM deployment challenging?

• Information system usage & deployment is new for Modelling & Simulation organisations (NAFEMS Survey) Most Analysts still use the file system & logbook

• Transformational IT like PLM is difficult• SPDM is significantly more difficult still• Even PLM Methodologies don’t fully

address Key Success Factors for SPDM• Small SPDM practitioner community


Engineers need better Information Management

There is a clear opportunity to increase time spent on technical work and hence increase throughput

Reference: Mark A Robinson An empirical analysis of engineers’ information behaviours


Stalled or Failed SPDM Project

• Analysts won’t use a an IT solution which costs them 30+ minutes per day of unpaid overtime

• Proposed solution reduces useful output of a department: Engineering Throughput

• Solution Value Proposition was not established• Client tested software A, purchased software B , …….

13


SDM Success Requires the Right Approach

Specific SDM Objective ?

Project FAILURE

No

No

Business Case to Achieve?

Yes

Engage Analysts

No

Yes

Identify the right SDM Approach &

Technology

No

Yes

Right Selection Criteria

Yes

No Fit For Your Purpose & Your IT Landscape

Yes

NoTeam with Right Skills

Yes

No

Bad advice on any of the first SEVEN steps is nearly certain Project Failure

Yes


A METHODOLOGY FOR SUCCESS WITH SPDM


SPDM Arch of Success

Applications,Scripts,Solvers,Queues,

Data ModelClassificationSDM MethodsSIM GeneratorsDashboards

Quality,Traceability,Throughput,Governance

Saving Time,Eliminating Laborious Tasks


Business Value of SDM

• 2 year study reported through NAFEMS Simulation Data Management Working Group based on a decade of production experience

• Provides a framework to justify an SDM project

• Business value is different for each organisation - a specific Value Assessment is needed

17


Project Design

• Designing a project to deliver Value is different to designing a project to deliver Functionality

• Identify achievable sources of value – To the Enterprise– To the Analysts

• Design Project to deliver value to both communities

• Manage solution delivery for Value

18


PLM methodologies

• Focus on Enterprise Value and Solution Delivery• Critical Business processes staff are required to use• Necessary but not sufficient for SDM

19


SPDM-SDMSPDM-Solution Delivery Methodology

Challenges

Engineering

Business

PotentialEngineeringValue

PotentialBusinessValue

SPDM capabilities

1 SPDM Assessment

Workshop/SDM2.0 POC

Project Proposal

2 Solution Evaluation

Business Sponsorship

Engineering Engagement

Business Sponsorship

Engineering Engagement

Project Phase 1

Project Phase n

3 Project Delivery

Value Achievement Diagram

Alignment Workshop/ POC output

Gantt chart for Deliverables

©theSDMconsultancy2018

NA

FEM

S SP

DM

Edu

catio

n fo

r Ass

essm

ent T

eam

NA

FEM

S SP

DM

Edu

catio

n fo

r Pro

ject

Tea

m


Value Achievement Diagram

Shar

ehol

der

Increase available Analyst Capacity by eliminating laborious tasks

Del

iver

Incr

ease

d En

gine

erin

g C

apac

ity (€

)

Cas

h Fl

ow

Value ObjectivesOperational Capabilities

NavigationReduces time finding data

Search SuccessReduces time re-creating lost data

Graph StructureEnables search by navigation

Search Reduces time finding data

Automatic Meta-Data CapturePopulates DatabaseSaves Time

UI ErgonomyDrives AdoptionPopulates DatabaseSaves Time

Migration UtilitiesPopulate Database

Simulation GeneratorEnables automated runs with audit trail creation

DashboardsEnable program management

Automatic Linkage CreationPopulates Database Saves time

Del

iver

Incr

ease

d En

gine

erin

g Ef

fect

iven

ess

(€)

Audit trailsEnsure Quality and Traceability


How to succeed @ SDM• Learn about the SDM domain, Industrial

Experiences of Early Adopters– eLearning, SDM Bibliography 200+ papers

• Identify an Industrial Objective for Project• Identify Solutions adapted to the Objective• Build a deployment plan adapted to the Solution• Build a strong team: Analysts, Vendors,

Management, IT,…Book “How to succeed @ SDM” in preparation


Conclusion:Value and Know-How are Key to SPDM success

Thank you for your attention

Enterprise’s Needs

Engineer’s Needs

Capabilities of SPDM

SuccessfulDeployment

€ Hours

CAE Data Process

[email protected]

www.nafems.org

Thank you for your attention

[email protected]+44 77 12 52 75 12 UK+33 6 42 28 58 76 France

How to succeed @ SPDM

Documents