System Dynamics

Certification and/or Accreditation

Atlanta, 2002

An old idea revisited

A feasibility study

by Gordon Kubanek, P.Eng.

Ottawa, Canada chust@monisys.ca

Feasibility Study - System Dynamics Certification and Accreditation

I keep six honest serving men,

(They taught me all I know)

Their names are What & Why and When

And How and Where and Who. Rudyard Kipling

Table Of Contents

A. Why

B. How

C. What

D. Who

E. When

F. Where

G. Pros & Cons

H. SWOT Analysis

I. Conclusions

J. References

K. Appendix - Survey Data & Previous Discussions, CL, model

Abstract

The purpose of this report is to help act as a catalyst for the System Dynamics Society to look into developing an accepted Body of Knowledge [BOK] and begin the development of a certification and/or accreditation framework. It is the sense of many who see the System Dynamics method as the tool for dealing with the development of policy in situations of dynamic complexity that it is better to develop a certification program before exponential growth comes our way. The Society would be wise to consider ways to reduce the risk of overshoot and collapse of our field [as happened to TQM] due to poor quality modeling and group facilitation skills. It is the author's hope that these thoughts and collection of previous conversations/survey results will help move this issue forward in a constructive and helpful fashion.

A. Why?

A. “ …SD may still be self-actuating but there is a lot of poor modeling going on in the name of learning. If we are ever to make progress as a discipline, there must be a role for a quality control process.” [Ken Carpenter - from SD list serve, June 6, ‘01]

B. It is probable that the “World” will discover SD within the next few years & realize, as we do, the great potential it has in helping managers make better decisions in complex situations. We must be ready for this as there is sure to be little warning but hints like references in Kaplan’s books to SD may be giving us a indication that Managers are finally ready for what SD has to offer them. If we are not ready we will not get a second chance in becoming THE tool for the playing out managers’ mental models of policy, strategy using scenario planning and prototyping. If we are ready we can avoid an overshoot and collapse scenario. [eg. TQM]

C. Although SD as a “field” of knowledge and application is healthy as seen from the “inside” but is relatively unknown outside a small circle of devotees. A subtle corollary to this is the author’s observation that SD seems to attract almost entirely one personality type: those who are process driven & introverted. The challenge for continued sustainable growth is to find ways to change the external perception of SD, while maintaining it’s current healthy internal state, and to attract other kinds of people – specifically the people-driven, extrovert personality. This latter group is, by the way, probably the majority of the potential users of our methods.

B. How?

There are at least two possible way to achieve these goals is to:

1. Certify individuals and/or

2. Accredit Institutions teaching SD

While is it easier to certify Institutions most SD practitioners have never taken more than the occasional course from a University so the certification of individuals would have a wider reach and impact. Similarly, if the perception of quality when SD is applied is a concern individual certification has a lot more to offer.

Delivery Method

A key part of “HOW?” question is the building of an infrastructure that could reach individual spread around the globe. It seems logical that the Internet via eLearning and encrypted answers and marking programs could be used to reach anybody in the World who really wanted to learn SD and be certified.

See Reference #10 for a good example of this from the field of Project Management [PM].

C. What ?

Before we can even begin to set up an infrastructure an accepted Body of Knowledge [BOK] needs to be defined. Clearly in our field there are, at least, two aspects that need to be addressed.

1. Modeling for Insight – this includes Systems Thinking, group mental model building,

facilitation, ST/SD as learning, SD for insight, etc. Tools are CLs, BOTs, Stock & Flow Policy models that run but have only a few stocks, moving from an event to system structure view of policy, including group model building/facilitation skills, the trick is to train the basic MATH concepts [stock & flow] so that everybody knows how & why the math is necessary.

2. System Dynamics – more complex SD computer simulations, ranging from models that are essentially “models for insight’ to complex, thousand variable models that take months to build; here the focus is on developing a Policy “answer” that guides decision making in a more predictive way than “just for learning”.

A committee would have to be struck to define this BOK.

A simple yet inclusive defining statement accessible to non practitioners is needed – for example:

“Project management is the application of knowledge, skills, tools and techniques to a broad range of activities in order to meet the requirements of the particular project. A project is a temporary endeavor undertaken to achieve a particular aim. Every project has a definite beginning and a definite end. “

Note: While the PMI [Project Management Institute, www.pmi.org ] BOK can be seen as so rigid and simplistic [ some say that it is often not able to be directly applied by Project Managers without modification ] it nonetheless had given PMI an excellent reputation. Due to its highly visible successes many organizations [the Canadian Federal Government] mandate that any Consultant hired to implement a Project MUST be PMI certified. Certification has made all the difference in this field.

Software: Awareness and Use of SD by non-technical managers

Part of our limits to growth is, that in spite of the superb developments in SD software packages, the building of models seems beyond many Managers who have no mathematical background. To be accessible we need to have a version of the software that can serve this audience. A good example from the PM field is Microsoft Project or the software by Primavera which uses “our” catch phrase “WHAT IF ?” .

D. Who?

The Policy committee [David Packer] of System Dynamics Society with the Education sub-committee headed by Jack Pugh need to spearhead and co-ordinate all efforts. The trick will be to get the right people at the table so the full spectrum of SD practitioners is represented: Academia, Government, Consultants, Software Developers, K-12 & good representation from around the world. We must avoid the problems that the Operational Research Society [OR] has had with the “Fellowship” developed by a sub-group of the OR Society:. We must act as “All for one and one for all.”

E. When?

Within 5 years we should be ready with a complete certification/accreditation program that has course material based upon an accepted BOK [body of knowledge] and an infrastructure for testing on-line; anywhere, anytime.

F. Where?

This initiative needs to be spearheaded by the Policy Council and Education committee of the SDS. However because of the global nature of the field and wide variations in approach to it around the globe, it is vital that players from outside North America are key participants in this process.

G. Pros & Cons

Most of the comments below are edited versions of emails from the list serve the author found since 1998.

PROS

> Quality control process needed to reduce the risk of poor modeling

Ø Certification supports self-directed learners[most SD practitioners] better than University

accreditation

Ø accreditation may encourage Universities to “bundle” their courses into a SD approved program

Ø PMI [Project Management Institute}] has been hugely successful in using external/on-line/individual certification to give their field/methods credibility and wide-spread use

Ø The SD Society should identify skills sets desired of a SD practitioner so an individual can recognize their strengths/weaknesses and not go beyond their depth as they apply SD to their problem

Ø Official certification and an accepted BOK would make it easier for new SD programs to start as

Administrators would see the field as more credible and thus more likely support such an

Initiative

> due to the push from Globalization and the ISO-9000 movement International standards are being

sought after by many bodies i.e. Accountants

Ø Practitioners would like to know their relative degree of knowledge & skill

Ø ASQ [American Society for Quality] and PMI certification programs have helped their quality of work done by demonstrating competence and with a defined BOK new certifications are added as

new areas of focus are found

Ø There are so called SD practitioners who present “SD” flight simulators which are in fact nothing more than rigged time series inputs – self taught presenters do not know what they do not know

Ø Referring to the Tipping Point concept: if we certify now before we’re discovered we can avoid overshoot and collapse experienced by the TQM movement which collapsed not because it is not a valid methodology [the companies using it properly do make more money] but because most implementing it did so very badly as they had no real skill and understanding of the subtleties

Ø “I’d be thrilled to accept a slower rate of growth of the field if it meant overshoot and collapse was avoided.”

CONS

Ø OR Society Fellowship concept was a failure

Ø Legal issues – could we make ourselves liable? Does certification mean and SD certified professional is accountable for the model results? What is the model results are used in a predictive fashion and are incorrect – can the professional be sued in the same way an Engineer can be sued if the bridge falls down?

Ø Accreditation of academic programs may discourage the spontaneous popping up of SD courses

Ø Other fields have multiple pre-requisites that filter people in the field before an certification step i.e. Doctors, Engineers

Ø We need an adaptive “bootstrapping” process to move us forward from our current ground zero

Ø Competency based testing and/or just in time learning has been a major source of disasters in Defense Dep. Disaster in Australia – a BOK that only tests and defines important element in parts without a sense of the whole is a dangerous thing

Ø The field has grown because it has not been formalized and most practitioners are self taught

Ø Most SD practitioners are not near a SD University program so certification of individuals with eLearning over the internet is a better choice than accreditation

Ø If there ever was a field of human endeavor less suited to certification using multiple choice exams than SD I have not heard of it

Ø We should be asking ourselves good questions, eg.: What drove the success of the Balanced Score Card [BSC] which has been around for 90 years? [author: there was finally a need for such a tool!]

Ø In Engineering many do not seek licensing [P.Eng.] precisely because of the cumbersome and expensive and time consuming nature of the process and if they are a P.Eng. they then have a legal duty of care whenever they sign a document [as a P.Eng. myself if would like to state that this is a partial truth – the bigger issue has to do with who has the right to call themselves an “Engineer” – the traditional definition only included Mechanical, Electrical, Chemical, etc. – with Software “Engineers” and others the control on the use of the word Engineer was lost…]

Ø We do not have the Resources for this kind of under taking

Ø There is no clamor or demand for this initiative

Ø There are not enough users [full time] to justify the effort [but do we know the total # of SD practitioners?]

Ø certification may give SD & modelers a false sense of legitimacy

I. SWOT Analysis

Internal	External
Strengths	Opportunities
45 years of development continued stable growth in the number of users global & broad range of users excellent quality of choice of software Body of Knowledge broad, deep and evolving Used in K-12 education to billion dollar projects Centers of Excellence eg. MIT, London, Bergen Many Consultants using SD	Rapid prototyping/simulations are now accepted tools Inclusion of SD in Kaplan’ recent books [BSC] has given us a wide audience [The Strategy Focused Organization] PMI is looking for partners to illustrate best practices for advanced project management in strategy, scope control and integration
Weaknesses	Threats
Inward looking and lack of willingness to incorporate with methods No accepted BOK [body of knowledge] Rarely taught as a separate discipline Little awareness in Business/Government Lack of formal linkages with Societies that are logical partners; PMI, ASQ, OR used as a black box by many consultants	Intellectual snobbery by academics that SD lack philosophical depth/credibility & is just another Engineering tool/software package Other groups steal our goals/values/ideas eg. What If? By Primavera We are not organized enough to meet the demands when we are discovered Perception that we not a defined field of inquiry

“Certification is like Tiger Woods: it changes the whole game.”

J. Conclusions

Like any good idea this proposal will only be as good as its execution when the right people commit to action. The following points are highlights of what many of you have said.

A. We must prepare for exponential growth in the field by developing an infrastructure for

quality control before we are “discovered’ to avoid overshoot and collapse.

B. At least 2 levels of certification are required;

1. which is syntax/recognition based where multiple choice questions would suffice

2. conceptual/semantic knowledge requires case studies in a problem based learning framework to show real modeling expertise

To make the advanced level of certification meaningful the modeler should demonstrate real competence with SD. Thus the “test’ must create a simulation where real problems with real clients so genuine double loop learning occur – so learners must be placed into a situation where they must integrate basic competencies into a problem solving context [Keith Linard]

C. This framework for an accepted BOK suggest in 1998 by David Packer is a good starting point.

1. Demonstrate knowledge of core SD literature ands models i.e. Industrial Dynamics, Principals of Systems

2. Demonstrate knowledge of basic SD archetype structures, applications & behaviors

3. Demonstrate knowledge/understanding by explaining structure & behavior of a referent SD model, including a validation test

D. A timeline of 5 years is realistic to get the whole infrastructure running but in the meanwhile we could be working with the ASQ and/or PMI to have SD included in their advanced certification programs to both raise the level of awareness and build knowledge from how they carry out this activity. There should be a 1–2 year pilot stage period to ensure the content and deliver method meet the needs of the learners.

E. On-line delivery of the course, learning and testing is recommended as SD practitioners are spread around the globe.

F. A simple, inclusive definition of the field that could be meaningful to most policy makers must be termed. Eg.“ System Dynamics is the application of System Thinking with Computer Simulations to the creation of Policy in Complex Dynamic Systems”

K. References

1. The Strategy Focused Organization R.Kaplan & D.Norton [2001] Harvard Business School Press

2. The Balanced Scorecard R.Kaplan & D.Norton [1996] Harvard Business School Press

3. MS certification: www.microsoft.com/mcpl.mktg.cert.htm

4. The Tipping Point by Malcom Gladwell, [2000] Little Brown @& Company

5. After the Goldrush by Steve McConnell – Certification & Licensing of Software Engineers

6. Things that Make us Smart by Donald Norma

7. http://www.cs.umd.edu/~ben -Ben Schneiderman -Interface Design for Syntax/Semantics Debate

8. Problem Based Learning – http://www.mcli.dist.maricopa.edu/pbl/problem.html
Problem-based learning (PBL) is a total approach to education. As defined by Dr. Howard Barrows and Ann Kelson of Southern Illinois University School of Medicine, PBL is both a curriculum and a process. The curriculum consists of carefully selected and designed problems that demand from the learner acquisition of critical knowledge, problem solving proficiency, self-directed learning strategies, and team participation skills. The process replicates the commonly used systemic approach to resolving problems or meeting challenges that are encountered in life and career.

9. http://www.primavera.com/

Since 1983, Primavera has become widely known as the leading provider of comprehensive project management, control and execution software. As the Internet has become an effective platform for project management and as new markets have increasingly turned to project management for competitive advantages. Whether bringing innovative new products to market, developing mission-critical information systems or managing billion-dollar construction programs, companies must choose projects with the highest market potential, maximizing the return on investment to the corporation. That's why Primavera solutions bring together the disciplines of strategic planning and project management, and provide a framework for effectively managing both the resources and the tactical plans for completing projects.

10. http://cess.wcu.edu/cobmpm/overview.html

Western Carolina University's On-line Master of Project Management Degree Program is designed to bring the benefits of modern technology and high quality, graduate-level instruction to project Management professionals in a form that is relevant to their fast-paced working world. Fully accredited at the national level, the goal of the master's degree program is to fit into the project manager's busy working schedule by providing flexible, direct access to work-related graduate education from any location that has a communication link, and at any time, by making the best use of the latest advances in educational technology. The degree program can be completed in two years.

11. Maani & Cavana, Systems Thinking and Modeling, 2000, Pearson Education

12. J.Sterman, Business Dynamics, 2000, McGraw-Hill

13. http://www.albany.edu/cpr/sds/SDS%20PC%20Web/accred.htm

14. www.optimlator.com

L. Appendix

A. Previous Discussions and Survey Data

This question has a long history. Since at least 1998 [Quebec City conference in Canada] accreditation and certification have been a point of debate. I have gone through many of the conversations stored in the SD List-serve database and the Survey results collated in Feb.2000 at the Optimlator web-site done by J.Mukerjee. Here are some select highlights of these several years of conversation that serve as a useful backdrop to this presentation in 2002.

Survey results showed that:

Ø almost 60% thought the greatest potential of SD was to use simple models to create insight and thus use it as a communication tool – this supports the view that the “soft” skills of group facilitation are every bit as important to the growth and sustainability of the field as expert, complex system, modeling skills are

Ø 60% said teaching SD on line could work

Ø only 24% said NO to accreditation of SD programs

Ø there was an even split on the issue of licensing of individuals – but IF so a Master’s Degree and 3 years of SD experience were considered to be minimum thresholds for such certification

Other comments:

Ø SD without math is not SD –focus must be on stock & flow thinking/skills

Ø SD is powerful when used as a method integrated into a problem/field i.e. Prototyping strategy

Ø Our challenge is to “cross the border” into the “Spreadsheet” mentality world where most Managers dwell – this is a large paradigm shift like our other goal of having Managers see themselves as Policy makers not primarily Decision makers

Ø Linking with PMI so that they recommend and teach as part of their advanced, specialized certification program for strategy, integration and/or scope control SD is possible – they are looking for “best practices”

Ø SD is referred to in a very positive light by Kaplan in his latest book The Strategy Focused Organization.

Ø A PM expert, Kerzner, in his text Strategic Planning for Project Management uses a PM Maturity Model which is very SD oriented

Ø Mentoring is how people are learning SD now – how can we build on this unofficial process?

Ø use external examiners to certify individuals

Ø rather than certify audit or review skills so an individual knows where they stand in a relative sense

Ø build on current strengths like the distance learning program at MIT http://command.mit.edu & roadmaps

Ø to certify we need:

Ø A. standard, accepted BOK & course[s]

Ø B. multiple stages of skills until final “professional modeler” certification

Ø C. on the job evaluation guided by an expert

Ø it’s easy to communicate basic SD competence but the learner can still have no real understanding of SD

Ø certification should include a demonstrated capacity to construct an SD model from a description of a dynamic problem and its causal structures

Ø the hardest part of using SD is getting your audience/customer to understand what SD is & why/how it will help

Ø part of SD Society’s mission should be the change Managers’ paradigm from that of decision maker to policy maker as today’s decisions [fire fighting] are predicated upon yesterday’s [bad] policies that created today’s difficulties [Jay]

Ø Definition of the field for others need development: i.e.“ System Dynamics allows rapid prototyping for Strategic Planning and Policy Design.” or “The SD method and tools are used to change Managers’ paradigm from that of decision maker to policy maker.”

Ø We need to know when SD modeling is the best method to use to intervene in a problem situation

Ø Effective SD modeling is the synchronicity of the right model, delivered brilliantly at the right time to the right audience [i.e. World dynamics was built over 2 weekends but had a huge impact]

Ø And on line course with CD ROM of text, work books, samples, software, using mostly the excellent material already available [with permission] would not be that hard to assemble

Ø SD has little utility when pure – it must be used in the service of other efforts/disciplines to demonstrate utility [note: exactly the same could be said about project management of quality control but these Societies certify – this is thus not a reason to not certify]

Ø There are, in effect, 2 skills of often 2 people needed, to make a modeling effort effective: an expert modeler and expert communicator/ group model building facilitator – expecting one person to be both is often not reasonable – as well these SD professionals must work with expert in the field under analysis to have the model’s relevance understand

define SD: “SD modeling uses symbols to represent high-order, non-linear, time delayed sets of interconnected integral equations to design policy for complex business & socioeconomic systems”

B. Causal Loop & Model

In System Dynamics method and content and intertwined. The above thoughts started with a “chicken scratch” causal loop diagram and policy structure stock & flow diagram to get clear in my head why certification could help the field grow sustainably. The following are meant to communicate and give insight into the facts presented in this feasibility study and are in no way meant to be conclusive or inclusive of all the factors involved in this question.

The above model is “under construction”. I hope to have it ready for the Atlanta conference.

This is a “model for insight” and the values are meant to be accurate in the relative, not absolute, sense. Thus while the values entered are not based upon available data they are meant to reflect the dynamics that the issue of certification has brought to the forefront: quality control once the use of SD has spread from a beyond a small, devoted group into the mainstream. The positive loop is about people “talking up SD” & the negative loop is about “SD models successfully solving problems” – in other words quality of modeling and group facilitation. The mental model built into the model is the use of two lookup functions that reflect the balancing and reinforcing loops shown in the causal loop diagram.

Once the final model is built you are all invited to vary these lookup graphs to see how your mental model effects the growth in the use of SD by managers.