Generic selectors
Exact matches only
Search in title
Search in content
Search in posts
Search in pages

For Better Estimation

For Better Estimation

Nate Silver teaches us how signal and noise can be confounded in magical ways, and how we are trying our best to get the best signals out, even during the fall of 2016 and 2020. Data, and in particular big data, are always a treasure to help people establish better understanding of our life and the life of others in the world (if we care). Except
when they are not.

We cannot for a single second blame the error-prone yet painstaking efforts of data collectors: they risk their lives to keep account how many nasal swab samples are tested positive, how many second-dose vaccines are being injected, and what is the virus concentration of tap water specimens – they simply are the most respectable warriors. It is not their job to separate signal from noise, and in fact, sometimes
they really should not waste their time to do that – if not many of them are doing the same thing in the same way.

Data analysts and modelers as we are, who sit at the back-end on the field of the battle against, say, against COVID (in fact, any data-driven estimation task has an enemy – the unknown), are obliged to never squander the efforts of first-row warriors. It is us who must bring out the best cuisine with raw data, upon which life-saving policies can be made.

Cooking is not easy; and in many cases, something could be skewed. Unconsciously. In Chinese cuisine, a dish is evaluated in three dimensions: look, smell, and taste – it is very probable that a cook may fail to deliver the best taste, the core of a dish, because he’s focusing too much on look and smell. Such may happen to a data cook: one might claim to yield a very important estimate of the basic reproduction number of COVID by using a very complex while realistic model (even that the results have quite narrow confidence intervals), yet the estimation scheme he adopted might be left unchecked.

We believe that the choice of estimation scheme plays a non-trivial role in parameter estimation. Which mold are you using determines what you produce. And for the estimation of comprtment models, one failure mode makes things wrong: conforming to norms. That means adopting a least squares estimation (summing over the squares of the difference between data series and model series, and then minimizing the sum; that really seems right, isn’t it) is not only easy, but also safe.

As one shouldn’t stay long in the comfort zone, we try to take a step out; we find that, NO, least squares is not a reliable estimation scheme for noisy data. In this study, we test a panel of advanced estimation schemes, and discover that the performance of least squares can be improved by a substantial margin. This means that policy recommendations based on least squares estimation results may be less accurate than we hope they are.

What’s the alternative? Well, we don’t know the optimal scheme for sure, but some solutions might be helpful. The negative binomial likelihood performs well across a range of conditions, so does the technique of Kalman Filtering. If these two methods seem too complicated (as they sound), then even a simple scaling of residual’s variance could largely remedy least squares. These are better molds for an equipped modeler
in simulation studies.

But these techniques do not, still, guarantee the complete removal of bias in parameter estimates. For one thing, as there is no golden mold that is one-size-fits-all, there is no perfect estimation scheme that best suits all data conditions. For another, estimation techniques are always second-order: you’ve got to first have a good model. As all models are wrong, no model estimate will be perfectly right.

This is not a doomsday call, though; this is calling on us to keep going in bringing out the best from data. We should be as painstaking as the front-line workers, and play with our own model specimens. For sure, we are as error-prone as they are – and we may be making even more errors – but just keep trying.

 

Li, Rahmandad, and Sterman are authors of “Improving Parameter Estimation of Epidemic Models: Likelihood Functions and Kalman Filtering”, winner of the Dana Meadows Award at the 39th ISDC. If registered, you can see the presentation of their work until August 31st 2021 here.

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

ISDC 2021 Highlights: A Peek into the Future of System Dynamics Visualization

ISDC 2021 Highlights: A Peek into the Future of System Dynamics Visualization

The International System Dynamics Conference (ISDC) convenes practitioners who demonstrate what’s new and developing in their fields with System Dynamics. This section of the WiSDom Blog, “Conference Highlights,” asks system dynamicists to spotlight key presentations and innovations presented at the conference. Check it out and let us know what you think!

Conference Highlights Editorial Team: Saras Chung, Will Glass-Husain, Jack Homer, Sara Metcalf, and Remco Peters with coordination by Christine Tang

This highlight by Billy Schoenberg gives us a peek into the future of System Dynamics data visualization. 

I’d like to describe a fascinating Plenary Session I attended on Tuesday of this year’s conference. Entitled Visual HPC Workflows for the Analysis of System Dynamics Models (HPC stands for High-Performance Computing), the 30-minute plenary talk was given by Brian Bush and Danny Inman of the US National Renewable Energy Laboratory (NREL). 

Here’s the background. Over the past two decades, NREL has developed a series of large System Dynamics models, many of them open source.   A big part of the task has been to exercise the models over wide ranges of uncertainty.  This means, first, developing technology to run these models on NREL’s HPC supercomputer clusters.  Second, it means finding an effective way to visualize an enormous volume of simulation output.  To do this visualization, they’ve built a “holodeck”-style room where they and their stakeholders can perform model testing. In addition, this room allows them to “walk through” the data in a 3D fully virtual environment.

Brian and Danny put their innovative system in context by saying that it was inspired by the idea of double-loop learning where learners adjust their own mental models of a problem by playing through and taking part in the system.  

One of the biggest challenges they’ve faced is the high dimensional complexity of the models they work with.  Their models might have thousands of input variables and parameters affecting dozens of output variables and metrics.  The big question they’ve been tackling is “How can a visualization system help stakeholders rapidly explore and confidently understand what it is that these models are saying?”

This is where the immersive visualization environments come in.  Using these specially equipped virtual-reality rooms, users can reach out and touch the data.  You can see what this process looks like in the photo above.

NREL uses this environment to enhance collaboration via “hands-on data processing”, where stakeholders can work together to filter and highlight key dimensions of simulation generated data.  This allows them to debug models, design scenarios, identify anomalous results, formulate and test hypotheses, as well as explore the outcomes of a large number of sensitivity runs.      

While it’s probably more than just a few years before we all get our own immersive visualization environments, the technology they’ve been developing is gradually trickling down into tools (based on open source software) that can be run on consumer-grade virtual reality or augmented reality headsets.  

The NREL team has found a few specific types of data visualizations to be particularly effective. These include 3D scatter plots of sensitivity testing results (pictured below); “parallel planes” (pictured below) showing input parameters versus outputs across many runs which are a series of scatter plots where the same observations on each plot are joined by a polyline; and “self-organizing maps” in 2D or 3D space that group together results that are similar according to cluster analysis or other algorithms.

All in all, this was a pretty fantastic look at what the future of System Dynamics model outputs may look like.  The immersive visual environments have a retro-futuristic feel, like a Star Trek holodeck brought to life. I doubt that this technology will ever trickle down to ordinary modelers in its full room-based form, though some of the visualization methods (e.g., 3-D scatterplots) are now available and getting some use.  What does the future hold? That’s anybody’s guess, but I enjoyed seeing System Dynamics at the forefront of the discussion about how simulation modelers may someday be able to conveniently visualize and make sense of massive amounts of multi-dimensional data.     

Billy Schoenberg – billy.schoenberg@iseesystems.com

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Don’t Dismiss Goodness-of-Fit

Don’t Dismiss Goodness-of-Fit

Next year, the famous Limits to Growth (LtG) study will turn fifty. The coming anniversary is causing many to look back to discuss its merits, its message, and the impact it had or could have had. Once again, LtG (and its 20- and 30-year follow-up books) is making a stir.

Exhibit A is a recent article by Gaya Herrington, a director at the Big Four accounting firm KPMG, who has done a fit-to-history analysis of the World3-03 model presented in 2004’s Limits to Growth: The 30-Year Update. She computes goodness-of-fit statistics for 10 different output variables, and four different model scenarios, against historical data from 1990 to 2020.  She finds that the model—and especially its overshoot-and-collapse Business-as-Usual scenario—generally fits the data well. Running with the news, Vice magazine has proclaimed: “MIT predicted in 1972 that society will collapse this century. New research shows we’re on schedule.”

Our colleague Tom Fiddaman has taken issue (in an email forwarded to the MIT SD Group) with both the Herrington study and how it was reported by Vice magazine. He feels that the use of goodness-of-fit testing, and the sensationalistic way it is reported, tend to overemphasize point prediction and that they obscure the fact that LtG was primarily meant to communicate broad dynamic insights.  Tom is discouraged that “the message isn’t out there sufficiently”, and he wonders, “Perhaps it’s time for a new approach?”

Let’s think about this. First, is the Vice magazine piece in fact harmful?  Although the headline is admittedly misleading, the article itself does a rather good job on the facts. Far from trashing LtG, the author echoes Herrington’s plea for people to wake up to our precarious global situation and to support efforts for global sustainability. On balance, I would say that the article is a plus for the cause of LtG.

Second, is the goodness-of-fit analysis done by Herrington really off-track and should it be cause for discouragement?  Setting aside the fact that Herrington is an avid supporter of the LtG message and a leading advocate for it within KPMG, did she do something wrong by emphasizing goodness-of-fit as a demonstration of model validity?

I don’t think so. I do not scold Herrington for her analysis—in fact, I applaud her. The reliability of system dynamics models rests on the twin pillars of structural realism and behavioral realism, and one useful way of judging behavioral realism is through goodness-of-fit (also known as behavior reproduction) testing.  It is far from a sufficient proof of validity, of course, but I do believe it is a necessary and important one, and I have found it indispensable in my own modeling going back decades.

Indeed, if there is any fault I can find in the LtG books (especially the later ones, when global data had become more plentiful), it is that they did not do goodness-of-fit testing.  I think such testing might have helped minimize the vicious backlash from economists like William Nordhaus and others concerned about the model’s apparent lack of quantitative support.  It might have smoothed the way for greater acceptance and changing more people’s mental models. I’m grateful that Gaya Herrington and others have done goodness-of-fit testing to demonstrate (even if in a small way) the behavioral solidity of World3.

So, is it time for a new approach in how we present World3 and other System Dynamics models, as Tom suggests? Maybe, but I do not think that our main problem has been poor communication skills. I think the bigger problem is that we have not wanted to play by the same rules as other policy modelers do. We have considered it beneath us to do goodness-of-fit testing because it is (so the story goes) a weak demonstration of behavioral validity. I disagree with this dismissive attitude and rather view such testing as an important tool of persuasion to help open people’s minds to systems insights.

by Jack Homer (VP of Professional Practice)

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

System Dynamics for Climate Change Mitigation

System Dynamics for Climate Change Mitigation

We had an insightful Webinar with the participation of With Juliette Rooney-Varga, Carolyn McCarthy, Sibel Eker, and Steve Arquitt .

Integrated System Dynamics models of economy and environment have long been used for research and decision support for sustainability problems, starting with the seminal work of World Dynamics and Limits to Growth. We discussed how System Dynamics models support decision-making, stakeholder, and public engagement for climate change and sustainability problems. We reflected on existing models and tools, such as Climate Interactive’s En-ROADS and Millennium Institute’s iSDG tool, and their use cases. We also discussed how the Climate Change Initiative at UMass Lowell uses System Dynamics tools to raise awareness on climate change.

If you’re a member, you can watch the webinar recording here.

Below are the answers to questions asked live during the Webinar.

Learn more about the Seminar Series.

Q&A Seminar | System Dynamics for Climate Change

Climate Interactive:

How would you describe the interaction between complex models (GCMs …) and simpler system dynamic models in more detail? (how can they support each other?)

Shortly, large detailed models help for cross-validating the simpler models. In return, simple models support the complex models in stakeholder engagement and scenario co-production.

What are the similarities between the EN-Roads Model & the iSDG Model? What are the main differences?

Both EN-Roads and iSDG are based on the System Dynamics method. Both emphasize transparancy, user friendliness, and shared learning. Both place great emphasis on facilitation and support in shared learning. The differences: EN-Roads is a global model, where iSDG is customized to support planning in a particular country or geographic region. EN-Roads is focused on strategies to keep global temperature below a specified level, where iSDG’s focus is more diffuse taking on all the SDGs.

What are some of the major impacts that the Climate Interactive team see on the application side of the models?

This question was not clear to me during the webinar. If “application” means the use of Climate Interactive’s models, we see quite a substantial impact. Only En-ROADS has been used in 73 different countries so far, engaging almost 63000 people. We have a wide audience, from policymakers and philantropists to higher education students and community members. One of the most striking and uplifting recent examples of En-ROADS outreach is the events organized by one of our ambassadors with smallholder farmers in Tanzania. 

Why don’t you use python as the intermediate language? Thanks

Answered during the webinar. Python is a user-friendly language but not as fast as C. We need speed in interactive simulators, so the Vensim model is converted to C.

Are system dynamics models being used in conjunction with Big data and AI?Can system dynamics models learn with machine learning?

There are initiatives about this as far as I know, and ML is very useful for quantifying empirical relationships, but outsourcing the model building completely to AI is not possible, neither desirable in my opinion. System Dynamics’s main strength lies in its descriptive nature, accessibility and understandability. While a hard coupling of SD and machine learning can provide many benefits, it might override the main strengths.

Since there seems to be many questions/comments regarding implementation/compliance, might it be helpful to start focusing on modeling the topic of governance itself, in order to identify and understand the influencing dynamics and loops on the gaps between the ideal solutions actually implemented?

In general, especially regarding specific sustainability governance problems, I agree that this should be the approach, because problem delineation and understanding the system strructure is key to developing any solution. In En-ROADS, though, the primary purpose is public engagement around the topic of “solutions”, hence the underlying dynamics are not co-modelled but shared with the users through various indicators and graphs.

Does the Climate Interactive climate-economy feedback have anything to do with Nordhaus’ “damage” function?

Since En-ROADS is an interactive simulator, it includes a damage function that allows the users to experiment with various “damage functions” found in the literature, including Nordhaus, or make their own assumptions. You can read more about it here 

When will the nature-based/land-based parts in En-ROADS be accessible online?

In the next few months. Please check either the En-ROADS simulator or this page

Very interesting presentation Sibel. Can SDeverywhere and the implementation into a website be done by somebody completely unfamiliar with C or Java or any programming? Many thanks.

I must say that it would be a bit challenging for someone who has no programming experience. There are guidelines, though, which might be helpful to get started.

Questions to Millennium Institute

What are the similarities between the EN-Roads Model & the iSDG Model? What are the main differences?

Both EN-Roads and iSDG are based on the System Dynamics method. Both emphasize transparancy, user friendliness, and shared learning. Both place great emphasis on facilitation and support in shared learning. The differences: EN-Roads is a global model, where iSDG is customized to support planning in a particular country or geographic region. EN-Roads is focused on strategies to keep global temperature below a specified level, where iSDG’s focus is more diffuse taking on all the SDGs.

Are parts of iSDG Model publically available?

Yes, go to www.millennium-institute.org/isdg . There you can access a demonstration model, videos on the iSDG, and the model documentation.

To what extent is the Millennium Institute SDG model open source? It would be so nice to use it rather than starting modelling from scratch in every research project.

At this time the iSDG is not open source. iSDG models are developed within the frame of a specific project. However, much can be learned about the model and its structure by visiting www.millennium-institute.org/isdg.

@Steve, how do you integrate “indicators” of SDG’s to report a single metric?

The iSDG reports the status of each of the 17 SGDs separately. The level of performance of the targets falling under each SDG are averaged to calculate the SDG performance. Targets can be thought of as desired levels of indicators. For a complete explanation see https://www.pnas.org/content/pnas/suppl/2019/10/29/1817276116.DCSupplemental/pnas.1817276116.sapp.pdf

SDG and how it is implemented in real world is highly context-dependence – how iSDG address this?

The iSDG is customized for the country or regional setting. Workshops are held with local experts, decision-makers, and stakeholders who review the model and identify key issues and policies to include inthe iSDG model.

Steve’s question regarding connecting real action to the plan is important. How do we make the interactive modeling tools an integral part of follow up and feedback on action?

With climate change and the SDGs both this really hits the crux of the matter. With the iSDG it is important to involve a broad spectrum of stakeholders on the modeling team and in the workshops who are motivated to see that the selected scenarios are being transformed into policies and then funded. This will require a well-trained team that can run scenarios, derive policies and work with the relevant government people to assure implementation. Monitoring is essential, and needs to be built into the projects. I fully agree with Juliette that the models need to engage with citizens who will then push leaders to make the necessary changes. I would love to hear others’ ideas and experiences on this.

Why choose poverty as a key #1 SDG?

“No poverty” as SDG1 was defined and designated by agreement of the 193 Agenda 2030 signatory countries. There is debate about which SDG is the most important. The iSDG takes no position on which SDG is the most important However, in the iSDG poverty is linked to almost every SDG.

Is this model (MI iSDG Tool) built in STELLA?

If you mean the iSDG, yes the model I showed was built in STELLA however we also have a version in Vensim.

What are some of the active projects that MI is doing today?

Currently we are working on iSDG projects in Afghanistan, Bhutan, China, Uganda, Namibia, Djibouti, Kenya, Democratic Republic of the Congo.

This question–or 2 questions–are for Steve. First, are worldviews and values included in the iSDG models? If so, how? A second question relates to how the highest-level decision-makers regard the models. I’m new to SDS but spent a number of years working with a roughly analogous set of high-level

After some relection, worldviews and values are pervasive in the iSDG model by virtue of the SDGs themselves. iSDG is intended to help policy-makers design strategies and allocate resources for attaining the SDGs. This includes the “leave no one behind” principle, gender equity in education and economic opportunities, equitable income distribution, preserving biodiversity for future generations, rule of law and many others.

Questions to CCI

@CCI any tips on how to engage kids with these tools?

Find resources: Comprehensive Facilitator Resources  Online World Climate Resources

Steve’s question regarding connecting real action to the plan is important. How do we make the interactive modeling tools an integral part of follow up and feedback on action?

Watch the recording for a full answer

@Juliette, could you say a bit about hope? There is a political divide especially in the US, but I read recently that % of the US population who feels anxious about climate change is ~68%. Could role-play games help deal with this anxiety?

Watch the recording for a full answer

“Research shows that showing people research doesn’t work”. What are your thoughts on this @Juliette?

We agree with John Sterman! But if you want to read more about this research, you can here

Why do you think that higher levels of “hope” begin and end higher with the i-H group?

Watch the recording for a full answer

Was the ethnic cultural diversity in your simulation group meetings more diverse than the photos would suggest? If not it’s concerning that you have a rather restricted sample?

Thank you for this question. The breakdown of participants’ racial and ethnic diversity for Building Consensus for Ambitious Climate Action through the World Climate Simulation can be found on page seven. Limitations relevant to our sample can be found on page 24 and reads, “Our sample was not randomly drawn from the general population and is therefore not expected to be representative of the American public. In addition, because the youngest participants in our study were drawn from programs serving low-income, first-generation-to-college students, age likely correlates with other demographic traits in our sample. We therefore do not claim that the observed effects of the simulation or demographics extend to the general American population.”

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

ISDC 2021 Highlights: Can Trust be Modeled?

ISDC 2021 Highlights: Can Trust be Modeled? The International System Dynamics Conference (ISDC) convenes practitioners who demonstrate what’s new and developing in their fields with System Dynamics. This section of the WiSDom Blog, “Conference Highlights,” asks system...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Watch the recording below

Whoops, this recording is available for members only. If you have a membership, please log in. If not, you can definitely get access! Purchase a membership here. If you're not a member but have purchased a ticket to this webinar, please contact us at office@systemdynamics.org

Practitioner Profile: Hyunjung Kim, California State University, Chico

Practitioner Profile: Hyunjung Kim, California State University, Chico

Welcome to Practitioner Profiles, a series of up-close blog-length interviews with experienced System Dynamics practitioners.  We have a standard set of 10 questions and let practitioners take the responses in any direction they choose.  They tell us about who they are, how they got involved with the field, how they work with clients, and in what new directions they may be heading.  A new profile will be posted every few weeks during 2021. 

For any questions or comments, please contact the editors of these interviews, Dr. Jack Homer (jack@homerconsulting.com) and Dr. Saras Chung (saras@skipdesigned.com). 

For today’s spotlight, we talked with Hyunjung Kim from California State University, Chico.

What kinds of SD project applications do you do at CSU Chico?

I teach and conduct system dynamics research in the areas of service delivery, environmental policy, and resource management. With a team of colleagues at the College of Business, we have incorporated SD into our undergraduate and graduate business curriculum. Between our capstone course in strategic decision making and the introductory system dynamics course, over 800 students get exposed to SD every year. My most recent SD research has been with the US Army Corps of Engineers, examining their flood risk management programs and identifying areas where Systems Thinking and SD modeling could be applied. We are also working on a game interface for them to accompany the modeling.

What is distinctive in your approach to SD projects? 

My research involves developing and applying formal methods for using qualitative data in system dynamics. It is important to understand perspectives of diverse stakeholders and systematically generate insights from their input. When it comes to teaching, it is important that my students have a positive first experience with SD so that they are motivated to explore it beyond the course.

In what way is your situation for SD modeling at CSU Chico perhaps different from that of academics elsewhere?

Our institution focuses on undergraduate education, and most of my students will go into the workforce soon after graduating. It is important for me to find SD topics that are relevant to my students and teach those topics in a way that can be easily understood and retained over time.

What other SD activities have you been involved in lately?

Currently, I am developing system dynamics learning materials for an exciting project called the Diaries During and After the Lockdown. The project was started by a group of SD modelers and an epidemiologist who created a blog targeted for a non-modeling audience. Using storytelling, it communicates insights from a system dynamics model of the COVID-19 pandemic.

How did you originally get interested in SD, and when was that?

It was during my graduate school orientation at the University at Albany. I was sitting next to a professor with a big smile who asked me why I wanted to study public administration. I told him I wanted to understand policy outcomes before actually implementing a policy, and he said, “Oh, then you should study system dynamics!” That was my first encounter with my mentor George Richardson.

What individuals and organizations are inspirations to you?

The Thursday Group members! The Thursday Group is a system dynamics peer mentoring group, and we have been holding weekly online meetings since 2011. This supportive group inspires me with research ideas and provides collaboration opportunities and feedback on my work.

What accomplishments are you proud of?

I find it rewarding when my former students tell me how much impact the systems perspective has had on them in their profession, and how much they appreciate what they got out of the SD courses.

What challenges have you experienced?

Communicating technical aspects of SD to people with little to no background can be a barrier to reaching a broader audience. Part of the hope of our current COVID Diaries project is to communicate these types of system insights.

What kinds of SD work would you like to be doing over the next 5 years?

I would like to focus on communicating system dynamics to a broader audience with no modeling background. I want to promote general public understanding and to use SD, with rigor and quality, on a daily basis.

Have other questions or comments? Leave a comment below or reach out to Hyunjung Kim.

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Practitioner Profile: Matteo Pedercini, Millennium Institute

Practitioner Profile: Matteo Pedercini, Millennium Institute

Welcome to Practitioner Profiles, a series of up-close blog-length interviews with experienced System Dynamics practitioners.  We have a standard set of 10 questions and let practitioners take the responses in any direction they choose.  They tell us about who they are, how they got involved with the field, how they work with clients, and in what new directions they may be heading.  A new profile will be posted every few weeks during 2021. 

For any questions or comments, please contact the editors of these interviews, Dr. Jack Homer (jack@homerconsulting.com) and Dr. Saras Chung (saras@skipdesigned.com). 

For today’s spotlight, we talked with Matteo Pedercini from the Millennium Institute.

lWhat kinds of SD project applications does Millennium Institute do?

We develop integrated simulation models for national planning, with the purpose of providing national governments with the tools and capacity to create sustainable development strategies.

What is distinctive in your approach to SD projects?  

I can think of two ways our approach is distinctive.  First, our primary goal is to provide tools and capacity, so that our client-partners can eventually perform the required analysis themselves.  Second, most of our modeling starts from a “template” that accumulates the learning from each country project into an ever-enriching framework.

In what way is MI perhaps different from other organizations doing SD project work?

I think we are an especially open organization, with a flat network structure, that allows all of us to express our ideas and learn from one another.

What is your role at MI?

I supervise our project work and facilitate knowledge exchange within the organization.

How did you originally get interested in SD, and when was that?

My first exposure to SD goes back to the mid-nineties, when I first played the Beer Game and did some very preliminary modeling at undergraduate level.  I got so passionate about it that I started looking for self-instruction materials—for example, Road Maps—reading avidly anything I could find.  The Fifth Discipline was also influential for me.  At the time, I was working for an NGO in Central Africa, and the systems perspective that I was slowly building was proving to be a tremendously effective framework for understanding the issues I was trying to address.

What individuals or organizations are inspirations to you?

I was, and continue to be, inspired by the work of the SD group at the University of Bergen, my alma mater. Besides their research and teaching, what really struck me is how Pål Davidsen and Erling Moxnes embedded that systems perspective and principles in their everyday work, leading the group by example.

What accomplishments are you proud of?

I am happy to see that SD modeling is gradually coming out of the shadows and is increasingly used for sustainable development planning.  It is still not a mainstream method, but systems perspectives seem to be gaining broader recognition and acceptance.  This acceptance is necessary for the world to adopt a deep understanding of the sustainability issues we face and the changes that need to be made.  I believe MI has contributed to such positive change, and I am proud of that.

What challenges have you experienced?

The single most difficult challenge that we have faced in many projects is the high turnover of key personnel in our client-partner institutions. Training is a central component of our approach and is critical for proper institutionalization.  High turnover can undermine our attempts to establish a solid core of well-trained personnel.

What kinds of SD work would you like to be doing over the next 5 years?

I’d like to further explore hybrid modeling methods, in particular blending SD with dynamic stochastic general equilibrium modeling, to strengthen our economic formulations, and agent-based modeling, to bring granularity and network dynamics where needed.  I have seen such hybrid modeling attempted, but there is plenty of room for improvement in its coherence and transparency.

Are there any specific changes you would like to make in your approach to SD projects?

It is sometimes hard to balance project efficiency and client-partner involvement.  Our approach emphasizes such involvement, but we often experience delays trying to adjust the project timeline to the client-partner’s availability.  When coupled with the problem I mentioned of client-partner personnel turnover, the delays can lead to a lot of rework.  I’d like to be able to increase efficiency without giving up on client involvement, which in the long run is critical for institutionalization.

Have other questions or comments? Leave a comment below or learn more about Matteo Pedercini’s work here.

 

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Top 10 Tips to Engage People with a System Dynamics Model

Top 10 Tips to Engage People with a System Dynamics Model

Andrew P. Jones is Co-Founder and Co-Director of Climate Interactive and a Research Affiliate at MIT Sloan. Andrew is an expert on climate change and energy issues, a prominent System Dynamics modeler, and a keynote speaker.

In his webinar with System Dynamics Society, Jones presented his 10+1 tips to engage people with a System Dynamics model. Here’s a summary of the tips, but you can watch the recording below for all details!

All Seminar Series are free for members of the Society. Join us today and unlock all benefits!

1. Make it a challenging adventure on their terms in your virtual world
Make your model playful to entice curiosity about what it tests. This is a virtual world where people will experience something new. The challenging adventure or game is what we do mostly with the interface of the model. If you open the En-ROADS interface, you will see two main graphs, the main output, and several sliders. When playing with the sliders, your changes will be reflected in the graphs. Avoid telling people: “I’m going to show you how to do this”, instead, try saying: “On your terms, you’re going to have to figure out how to get that graph down to your target line”

“We do NOT model systems. We model problems.

2. Focus on mental models you want to improve
We do NOT model systems. We model problems. In System Dynamics, we define a problem as a behavior over time, a reference mode that we’re curious about why does it behave like that. We’re so curious about the problem that we develop a simulation to understand it. We build interfaces to improve the mental models of the stakeholders or clients. System Dynamics modelers know that mostly the mental models of decision-makers don’t match with the real world and how the actual system behaves. We don’t build interfaces to give access to policy levers in the system, but we are trying to leverage the necessary mental model improvements.

3. Make participants mentally simulate before running a scenario
If people don’t have a theory, they don’t notice the mental model they use to forecast what’s going to happen. Then they don’t get to learn and improve their understanding. Therefore, you must encourage people to anticipate, based on their own theories, what’s going to happen when running a scenario. No theory, no learning.” (Edward Deming)

“Tell people what to look at, let them guess what they are going to see, and then make it exciting like a movie”

4. Create a dynamic visual experience
Humans like to see change and it is important to employ this fact to get people to run and look at your graphs without losing their focus. Make people think and anticipate and then direct their eyes to what you want them to notice. If you do this well, you don’t even need to draw a Causal Loop Diagram because they are drawing it in their heads. Tell people what to look at, let them guess what they are going to see, and then make it exciting like a movie.

5. Set up participants to talk with each other
People learn socially and with hot-button issues like climate change, they need to know that their friends are not going to hate them if they change their minds. You need people to be processing the information collectively and socially with others. When presenting your model, stop frequently and advise: “Turn to the person next to you and discuss what you think of that conclusion”.

“Keep the discussion on improving system performance not on the tool you’re using”

6. Know along the way that you are playing at least seven roles

Try to present your model with the help of, at least, another person. Ideally, you need someone to facilitate the discussion while another runs the model. As a facilitator, you will be playing four roles including:

Coach – Helping participants to extract insights out of the model.
Professor – Teaching the audience theories and sharing factual information
Playwright – Creating an emotional journey of ups and downs.
Fellow Traveler – Being authentic, vulnerable, and just another person trying to solve a complex problem. If it is worth making a model about, it’s very important. If it is very important, you must deeply care about it.

You want the conversation to be around how to solve a complex problem together and what your audience is going to do about it. Keep the discussion on improving system performance not on “oh that’s a cool model how did you make that”. That’s all side information.

You want to minimize your roles in:
Tech – you must keep the conversation on improving the system performance, not the way or tool you have used to develop the model or create the graphs
Advocate – when people disagree with you, your job is not to fight them or disagree with them back, but to set up others to find the voice of your work.
Defender – Don’t get involved in the fight of “YOU HAVE A BAD MODEL!”. Avoid this fight as much as you can.

Read the article “Teamwork in Group Model Building” on the System Dynamics Review for more insights on strategies for efficient and effective model building in groups.

7. Build confidence and share testing as needed

You can share your tests and comparisons to other models and/or predictions to build stronger confidence. For instance, Climate Interactive and MIT Sloan built the En-ROADS with the best science available, using the data sources such as the Intergovernmental Panel on Climate Change (IPCC) and International Energy Agency. All the assumptions are available open-source in the 400-page reference guide. Many of the assumptions that someone might not believe in are changeable within the model. Several models do not have good literature of other scenarios to compare against, while En-ROADS can be compared with six integrated assessment models to build confidence.

8. Use loops and stock/flow diagrams only to illuminate
As you help people improve their intuition, sometimes you need to use loops and stock/flow diagrams, especially if you are presenting to a more technical audience but always connect it to the simulator. Avoid showing this to policymakers, they are usually not interested in the loops, instead, tell a story about reinforcing or balancing feedback.

9. Make space for feelings and processing

Deliver your presentation with excitement and intensity, but you need to slam on the brakes and let participants compromise with their feelings and do the necessary processing. You may create a scenario of success, and they get to create their vision, something that they would love to see. And it’s time to slam on the brakes and may invite them to 60 seconds of silence. Yes! It is weird but imagine 60 seconds of silence of people sitting with a scenario of success.

“Create the conditions where people are open to changing their minds”

10. Pay attention to three legs of the learning stool
Reflective Conversation – Create the conditions where people are open to changing their minds, surfacing and testing assumptions, and talking to their peers about improving their assumptions. You’ve got to make the space where people are open to being wrong and thinking differently.

Vision – Help people see a future that they fall in love with they just want it so badly that they see the gap between the vision they want and the reality that they feel and experience that tension in between. Orient towards what one genuinely cares about.

Systems Thinking – Explaining how a complex system works where time, cause, and effect are distant in time and space can be difficult, especially when you include stocks, flows, feedback loops, and other inner relationships. You can learn how to facilitate a training that uses systems thinking and System Dynamics for free with Climate Interactive learning resources

10 + 1. Give them the simulator

People need your help as a facilitator, but you must give them something to play with that naturally gives them the mental model improvements that you want and guides them towards committing themselves to action to improve system performance. As Buckminster Fuller states “If you want to teach people a new way of thinking, don’t bother trying to teach them. Instead, give them a tool, the use of which will lead to new ways of thinking.”

Watch the recording below

Whoops, this recording is available for members only. If you have a membership, please log in. If not, you can definitely get access! Purchase a membership here. If you're not a member but have purchased a ticket to this webinar, please contact us at office@systemdynamics.org

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

ISDC 2021 Highlights: Can Trust be Modeled?

ISDC 2021 Highlights: Can Trust be Modeled? The International System Dynamics Conference (ISDC) convenes practitioners who demonstrate what’s new and developing in their fields with System Dynamics. This section of the WiSDom Blog, “Conference Highlights,” asks system...

Upcoming Events

En-Roads: an online Climate Action Simulation

En-Roads: an online Climate Action Simulation

SYDIC is proud to announce the “En-Roads: an online Climate Action Simulation” workshop, that will be held on 21-22 September 2021 on the occasion of the “#All4Climate – Italy 2021” manifestation, promoted by the Italian Ministry for Ecological Transition (MiTE), and...

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

System Dynamics Biomedical Modeling

System Dynamics Biomedical Modeling

In this Seminar, authors present their work published on the special issue of the System Dynamics Review on Biomedical Modeling. The special issue brings together new research that explores a fascinating array of biomedical problems, from the modeling of pharmacokinetics, hematologic disorders, and blood cell production to chronic disease progression at an aggregate level. System Dynamics health modelers engage with interdisciplinary teams of medical and policy experts in order to explore exciting new biomedical research opportunities.

Learn more about the Seminar Series.

Watch the recording below

Whoops, this recording is available for members only. If you have a membership, please log in. If not, you can definitely get access! Purchase a membership here. If you're not a member but have purchased a ticket to this webinar, please contact us at office@systemdynamics.org

Here some answered questions from the Q&A Session!

Which critical data elements should one collect to gather further insight into model behavior? Were you able to collect medical data to calibrate your model? If not, what are the main drawbacks with respect to your conclusions?

We had to use existing data in the literature for model calibration. But of course, we had some problems with that. First, the data we found was not exact numerical values, rather they were presented in graphs. Second, individual data of patients were not included, giving all the statistics as sample averages. Therefore, we could not delve more into the personalization aspect. Additionally, the literature lacked sufficient blood count data under chemotherapy, making it difficult to calibrate for different chemotherapy regimens. What we actually needed was continuous, or at least a high frequency, time-dependent neutrophil counts for individuals with different characteristics, for example, neutrophil response to a single G-CSF shot for 40 hours and blood cell response in a single chemotherapy cycle. [by Orkun İrsoy & Şanser Güz]

How did you come up with alternative treatments? Was it through experimentation or optimization? 

Since our main focus was not concentrated on finding an ‘optimal’ treatment, we didn’t use any dose & timing optimization procedure. Effectively, our strategy landscape for treatment protocols is defined by two factors: G-CSF Starting Time and G-CSF Injection Count. The standard protocol starts G-CSF supplementation by the end of the fourth day and continues until the next cycle of chemotherapy. Thus, increasing the injections from the end was not an option as it interferes with the next cycle. In our analysis, we reduced the number of injections to create our alternative protocols, to observe the effects of over-and under-stimulation dynamics. We also tested protocols with different starting times (results included in electronic supplement), but the inferences were no different than the one we derived from our current array of protocols. [by Orkun İrsoy & Şanser Güz]

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

ISDC 2021 Highlights: Can Trust be Modeled?

ISDC 2021 Highlights: Can Trust be Modeled? The International System Dynamics Conference (ISDC) convenes practitioners who demonstrate what’s new and developing in their fields with System Dynamics. This section of the WiSDom Blog, “Conference Highlights,” asks system...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Behavioral Dynamics of COVID-19

Behavioral Dynamics of COVID-19

Over a year since the World Health Organization declared COVID-19 a pandemic, its true global magnitude remains unknown.  Official counts of cases and deaths are known to underestimate the true magnitude of the pandemic, but how much remains uncertain.  Differences in testing and treatment capacity, unknowns like asymptomatic transmission, and most importantly, variable responses to the risk of the virus create wide variation in incidence, prevalence, and mortality across countries and over time. This isn’t just about getting the stats right.  If official counts of cases and deaths are too low, people may not take adequate precautions and governments may (and have) re-opened their economies too soon, leading to more deaths.  Effective responses to the pandemic require an understanding of how these factors interact to shape its trajectory.

To address this need, we developed a model that accounts for behavioral factors such as risk-driven contact reduction, improved treatment, and adherence fatigue, as well as asymptomatic transmission, impacts of weather, disease acuity, and testing and hospital capacity. Estimating this model across all nations that publish sufficient data—a total of 92, spanning about 5 billion people —we found the actual magnitude of the epidemic to be much larger than reported. Specifically, as of the end of 2020, we find cumulative cases were more than seven times larger than official reports, and COVID-19 deaths were about 44% higher than reported. We also found huge variation in death rates, with some countries having over 100 times the per capita deaths of others.

The greatest driver of this wide variation is not demographics, weather, or health care capacity, but rather how responsive are the people and governments of each nation to the threat. Do countries act proactively and aggressively to quash any nascent outbreaks, or do they wait until the situation is severe and hospitals overwhelmed before responding?

Some countries, like Australia, China, New Zealand, and Singapore, have been proactive, responding aggressively to even small spikes in cases, and have largely succeeded in bringing their epidemics under control at death rates below 0.1 per million people per day. Even after caseloads and deaths fell, they remained vigilant, kept masks on and gatherings restricted, and continued testing and tracing until the virus was almost completely stamped out. Despite occasional disruptions from new outbreaks, life in these nations has largely stabilized at a new normal, with few cases and deaths.

Others, including the U.S., many European and South American countries, and India, have been more reactive. Seeking to minimize economic disruption, they delayed action until climbing deaths forced their hands. As each wave subsided, under pressure to reopen their economies, they eased restrictions, only to see cases climbing again. Eventually, faced with rapidly growing outbreaks, they have been forced to lock down again anyway.

Ultimately, countries actually have little choice in how much they must reduce contact levels to control the epidemic. Few communities are willing to tolerate unchecked outbreaks and the horrendous number of deaths that result. By choice or the force of intolerable death rates, all countries have to cut back high-risk interactions – keeping people at home, restricting gatherings, avoiding restaurants and travel, and so on – to control the virus. How much these interactions have to be reduced to keep an outbreak from growing depends largely on the contagiousness of the virus itself, and thus remains similar across nations. What varies across countries is how many cases and deaths it takes to induce strong enough actions to reduce contacts to the required threshold, bend the curve, and stop the growing outbreak. In short, until vaccination is widespread all communities pay a similar price to sufficiently bring down their contacts, yet the more responsive ones save many more lives.

To help policymakers and the public better understand these dynamics, we created a free online simulator that allows you to experiment with your own scenarios for vaccination and other policies, and to explore how to change the course of the epidemic over the coming months.

The world is working hard to roll out mass vaccination, but it will still be many months before most people are vaccinated worldwide. Understanding the central role of responsiveness in shaping the dynamics of outbreaks remains critical – it is still not too late for a swift response to both minimize economic disruptions and save lives.

Rahmandad, Lim, and Sterman are coauthors of “Behavioral dynamics of COVID-19: estimating under-reporting, multiple waves, and adherence fatigue across 19 nations,” which is forthcoming in System Dynamic Review. Rahmandad and Lim are coauthors of “Risk-Driven Responses to COVID-19 Eliminate the Tradeoff between Lives and Livelihoods.

 

 

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us

Practitioner Profile: Etiënne Rouwette, Radboud University

Practitioner Profile: Etiënne Rouwette, Radboud University

Welcome to Practitioner Profiles, a series of up-close blog-length interviews with experienced System Dynamics practitioners.  We have a standard set of 10 questions and let practitioners take the responses in any direction they choose.  They tell us about who they are, how they got involved with the field, how they work with clients, and in what new directions they may be heading.  A new profile will be posted every few weeks during 2021. 

For any questions or comments, please contact the editors of these interviews, Dr. Jack Homer (jack@homerconsulting.com) and Dr. Saras Chung (saras@skipdesigned.com). 

For today’s spotlight, we talked with Etiënne Rouwette with Radboud University.

What kinds of SD project applications does your university department do?

I’m the chair of the Methodology department of the Nijmegen School of Management, Radboud University, the Netherlands. We have about 20 research and teaching staff, and about 10 PhDs and postdocs.  SD projects for organizations are done as part of courses and theses, and as consultancy assignments. In the past year, our client projects addressed cybersecurity and circular economics; and we also did internal projects for Radboud on school reputation and the demand for master’s programs.  We have also done projects on sustainability, energy transition, gender and diversity, and health care and mobility. 

What is distinctive in your approach to SD projects? 

Almost all of our projects use the facilitated group model building approach, following the lead of former Radboud professor, Jac Vennix.  We combine empirical research with stakeholder intervention methods, holding meetings with small groups of five to fifteen people.  We employ a combination of brainstorming, Electronic Meeting Systems, participative scenario development, multicriteria decision analysis, and adaptive planning and gaming.

In what way is your organization perhaps different from others doing SD project work?

We have nearly 30 years of experience using facilitated methods and assessing their effectiveness.  We tailor the methods to a specific issue and client circumstances. We are developing our intervention methods further, for example, through analysis of video recordings of the sessions to see how clients react to various elements.  We would like to refine our methods and determine exactly what the effective ingredients are.

What is your role in your organization, especially with regard to the SD project work?

As department chair, my role is to ensure we have the necessary staff for teaching and research and to cover our application areas.  When client prospects come in, I first consider who is interested and has time available to work on the project.  Next, I lead an intake process to build an understanding of the client’s problem and choose an appropriate approach for them.  Sometimes, potential clients do not seek out our SD expertise initially but are rather part of industry consortia coming out of PhD research. We then have to determine the right match between the potential client’s problem and the methods we have to offer. 

How did you originally get interested in SD, and when was that?

I became interested around 1990, when as an undergraduate student in psychology at Utrecht University, I met Jac Vennix.  He made systems modeling fun and showed us the logic and rigor of it. When a group provides the information for a model, and you show them the simulated results, you are in effect saying, “This is the logical consequence of what you have told me.”  How that changes the understanding and decisions of the group is still my major interest in research.

Besides Jac Vennix, what other individuals have been inspirations to you?

Jay Forrester is definitely an inspiration, as the founding father of system dynamics, for his work on grand societal challenges and his emphasis on education.  In 2019, my son was born, and he is an inspiration in an unexpected way.  Suddenly, simulations to the year 2100—for instance, in climate models—have become a lot more relevant.

With regard to SD project work, what accomplishments are you proud of?

Two client engagements come to mind: one on safety in a city district, and the second on criminal justice. I met the leaders for the first project again 15 years after the engagement, and they related the circuitous and unexpected ways in which our causal-loop diagram ended up influencing the city’s policies and life in that neighborhood.  The project on criminal justice was built with multiple stakeholder groups at the national level and resulted in a large simulation model. The project led to over a dozen follow-up projects with national ministries.

What challenges have you experienced with respect to SD project work?

While most clients have been receptive to the systems work, sometimes the findings lead to no change, or the process is terminated midway.  These negative results seem to occur when the findings go against the interest of one or more important stakeholders.  I would like to learn more about how our facilitated approaches might accommodate emotion and politics.

What kinds of SD project work would you like to be doing over the next 5 years?

I recently had the opportunity to work with scientists on the subject of cognitive decline due to dementia.  The participants were specialized in different domains – geriatrics, radiology, sleep, nutrition – resembling the “silos” in private or government organizations.  I found that the integrative approach of group model building worked very well even in this realm of clinical science.  I hope to be doing more such “team science” projects in the coming years—a concept gaining importance in the Netherlands and elsewhere.

 

Have questions/comments? Reach out to Etiënne Rouwette or leave a note below.

Recent Posts

Little’s Law and Stock-Flow Performance

Little’s Law and Stock-Flow Performance Understanding the mechanisms underlying stock-flows is fundamental in managing dynamic systems. Keeping stock-flow systems under control is a challenge faced in every aspect of life. The accumulation of greenhouse gases in...

Upcoming Events

Recent Business cases

Diabetes

Diabetes Client Centers for Disease Control and Prevention (CDC) Authors/Consultants Jones AP, Homer JB, Murphy DL, Essien JDK, Milstein B, Seville DA Diabetes mellitus is a growing health problem worldwide. In the United States, the number of people with diabetes has...

Project management at Fluor

FLUOR Name Managing the Dynamics of Projects and Changes at Fluor Modelers Kenneth Cooper, Gregory Lee Client/Participant FLUOR Client Type Corporation The Issue You Tackled Fluor is one of the world’s largest engineering and construction firms, with 2008 revenues...

Process Innovation at Du Pont

Process Innovation at Du Pont Client Du Pont Authors/Consultants Repenning NP, Sterman JD Improvement programs such as Total Quality Management are embraced by many organizations but are often discontinued before full benefits can be reaped. With ever-increasing...

Join us