example of diagram influence of the shifting of the burdle p

[LAUJJL]

Hi

Problem exposed in the Sd mailing list.
On the influence diagram, the dotted lines represent informations link, full lines represent flow, levels are in a blox and rates are in italic.
Regards.

[LAUJJL]

Hi

The prop.mdl diagram is of course simplistic and never conforms to any reality.
It has the advantage of simplicity and can help to take decisions even at the qualitative level.
But lots of things are wrong.
The model is closed and in the reality often the problem is driven by exogenous factors and
endogenous one’s.
The crisis if it happens may have disastrous effects if it leads to bankruptcy or positive effects
if it influences the percentage of long term action and decreases the threshold to take action.
The amount of resources is critical too.
If the resources are limited and the consequences of the crisis are strong and can lead to bankruptcy
it may be better to take short term actions. It is better to stay alive with a hypothetical future than be
dead with a bright future.
The problem with a more complicated model, is that it becomes impossible to draw any conclusion
from it, if it stays qualitative. And there are so many parameters, especially delays, that without knowing
the parameters, it is impossible to find a policy. Not to mention that the bigger the model, the more difficult
it is to analyze, and the more chance it has to be bugged conceptually especially or to miss something important.
I do not believe that it is possible to use practically such sort of generic models, unless the model is built on a
very specific case.
An example of more complicated qualitative model is the prop2.mdl. I prefer not to lose my time trying to make
it quantitative.
Regards.
Jean-Jacques Laublé Eurli Allocar
Strasbourg France.

[crbnblu]

JJ,

Thanks for the models. And I agree with you that real world situaitons are far more complex than the simple models oftern used to present learning points. Good food for thought.

be well,
Gene

[Ralf]

Hi JJ,

Just by sheere luck I stumbled across the board and saw your posting. Downloading the model, kicked-in the SD learning and inquiry mode in me:-))

Thanks for providing such great work - now it is time to discover our collective learning out of that. How to make the conversation fluid and valuable?

At first glance I sensed that there is lots to learn from it (especially about mental models in the current crisis and how to get out of it for the better in the long-term).

Cheers,

Ralf

[LAUJJL]

Hi Ralf

I posted a reply to the query of Gene Bellinger in the SD mailing list with the reference to the Vensim UK Forum where you can post models.
So you did not need normally any chance to get informed by my posting.
About the great work, it took me between one and two hours to build both qualitative models, and quantifying them would certainly show severe
conceptual bugs. My idea is that such very general model can be modelled in very different ways and can have any sort of behaviour depending
on its structure and the parameters you put on it, not to mention that having no reality to compare it to and no precise question, your model is
bound to represent just nothing but a vague representation of how things in general work and behave. I am interested by concrete problems and
mine in particular.
So to solve problems one must be much more specific and ask precise questions about a real world problem.

Gene Bellinger sent me a direct e-mail and a post in the Vensim forum. Anyhow it did not interest many people, but in the SD mailing list, many
gave their opinion about that problem, just with words.
It has been one of the repetitive characteristic of the SD mailing list to be model adverse since I subscribed to it more than 6 years ago.
People can talk weeks about a subject but taking the time to build even a simple qualitative model is too much of an effort.
I posted two models in the Sdsbizsig, and had never any return or question about it, one being an interesting bubble.
I have since unsubscribed from the Sdsbizsig where there is never any post exchanged, nor any model proposed.
It would of course be interesting to study some interesting cases (making models) and preferably with a short time horizon to be able to
see the effect of the modelling effort, but who would participate?
I think that experienced modellers do not care and mind their own businesses, as once told me Aldo Zagonel, one participant of the SD mailing list
certainly an experienced modeller. So if you have a problem, it is better not to hope for an outside help, only technical as on this forum.
So having tried for years to give something without getting anything in return, I have learned the lesson.
A member of the Sd mailing list, having written books on SD and having a prominent academic position, told me last year that I had nothing
to do on the mailing list where my ideas could mislead newcomers to the field!
So I think he is right about my participation on the mailing list.
I post rather frequently on this forum, if I feel that I can help or encourage somebody, because I know that if I ask a technical question I will get some
help too. As for the rest, I even do not know if I will renew my membership to the SDS this year.
I know that I am not very encouraging but it is better to face reality whatever it is.
Best regards and happy New Near to you and all people on this forum and good progress in the art of modelling useful models.
JJ

[Ralf]

Hi JJ,

I can clearly hear the disappointment about the dynamics in the interpersonal actions concerning the questions on modeling and bringing the field (a garden never only exists of the blossoms one can see, but also the fertile underground, the seeds, water, light, warmth one would not appreciate as such - if looked just at the beauty of the garden).

I feel sometimes the same way (in several fields;-)) and I always sense that asking provoking questions (even as a newbee) shakes the current stable situation. People are seeing that the past stability in the system (whether it SD, LEAN, OL, OD, etc.) is shaking (as the financial crisis has clearly done).

Pushing into that stability is always threatening to people and they will react (either in putting you outside their inner circle, not talking, not giving advice). And yet there will some amongst them (as you clearly show) who come more from the practical side of life and have been drawn into SD (by sheere necessity!).

Small steps and yet the cracks in the ice, from where you can open up the whole lake (you just have to know how to find the leverage!).

If however system dynamics is merely for academics (which I would clearly doubt, even though some personal talks with people -highly educated PhD students and students in their SD studies) showed something different;-(() then we should leave the field and found:

Real World Dynamics

...hm sure, whether that would be a good idea, as this would weaken the forces even further and silo thinking.

I have the feeling that the underlying "fluid oil" of TRUST between people in the field opening up their mental models and assumptions openly is strongly missing. Even in a closed group (to about 20 people on similar issues is a rather challenging journey - finding the leverage point to kick it really off the hook is not easy!).

As I am quite positive not just us two are reading, writing or thinking about the issue I would like to ask the question:

"What would you as a practicing systemdynamicist have to give up before to take part in a more collectively orientated approach (which surely should feed all our needs - earning money!)?"!

Best regards

Ralf

[LAUJJL]

Hi Ralf

There are it is true some difficult relations between people in the SD community. But I think that it comes from the different jobs of its members.
There are great differences between the problems that encounter a businessman, an academic, a researcher or a consultant. The big majority
Of SDS people are academic or researchers. I am even a special type of businessman as I own my business and make my model by myself.
To resume I do not sell SD to anyone, I use it. These characteristics will generate incomprehension between members. But the SDS cannot satisfy
everybody and I just recognize the fact that being nor academic nor research and nor consultant oriented
I just have no reason to be a member of the SDS, being not able to look otherwise than I am and having different needs.

System dynamics could be much more used in practical problems. The reason that it is not is not easy to understand. But it is a fact.
I use it now for my business with profit, but it has taken me years to become proficient enough and I have still a lot to learn.
The fact that it took so much time is really a problem.

I do not really understand your question about giving something up to take part in a more collectively orientated approach.
You may mean that if I was not interested by money I could eventually best participate to a collective approach?

I am first not personally much interested by money. I am interested in succeeding in my endeavours.
A real collective approach needs personal efforts from every one. For instance look at this forum. I think that to help the field, some supposed experts would be better helping novices to build models than discuss about intellectual questions and lament about the poor range of practical applications and discussions should be model oriented. SD is about making preferably useful models.
Best regards.
JJ

[Lee Jones]

Hi JJ,

Very interesting comments and you are right, the SD Society has been unable to satisfy differing and often conflicting needs of the community. But if business people like yourself see no value in being a member of the society, then there is a huge problem.

As you know, I am a consultant and have worked in simulation and System Dynamics for the past 20 years. System Dynamics is valued by certain businesses who, because of the competitive advantage they gain from using it (perceived or real), they prefer not to publicise its use.

I see value in membership of the society and, indeed, we are sponsors of the International organisation and UK chapter but I fully understand where you are coming from.

I hope you continue to be a valued contributor to this forum, there are many new users of Vensim across the globe who have benefit from your support.

Cheers

Lee

[LAUJJL]

Hi Lee

I do not think that being not interested in the SDS is a problem, because there are very few people like me as members or they
eventually never participate to any forum or something else.
I have been 7 years member of the SDS and honestly I must look hard to find anything that has helped me.
The only thing that I can recognize is having heard of Coyle’s book when he was still participating to the forum.
I owe to his book to have persevered trying to use SD, otherwise I would have given up. And I had to insist that his book be
mentioned in the short bibliography list of the SD web site. It was not even mentioned!
I would have attended the international conference if the modelling workshop was not ridiculously short. Twice one hour before
lunch. It is the only thing that could interest me and I would prefer that there was a continuous modelling workshop during the 4
days, 8 hours a day and have a complete report of the problems to be discussed before the beginning of the conference!
This is what I call practical SD. But I think that if there are only two hours of modelling workshop the reason is that there are
few people interested by practical problems at least at the conference. I think that the potential interest is very short.
I was mainly staying member to contribute modestly to the field, but I am only contributing to more academic and
research work that if it is not paired with practical concerns has no value to me.
I attended the last English conference at the university of South Banks in London. It was ok and not too long.
I did not see anybody from the Ventana England there.
I might go again this year if the practical cases that are supposed to be exposed interest me.

You have of course a different point of view as you are in the SD business and it is the only place
where you can have the international contacts you certainly need. I had certainly some interesting discussions on the SD mailing list,
but frankly it was most of the time intellectual speculations with no practical applications.

About the value of SD for some businesses, I have a doubt. How could an effective method be kept secret for such a long time?
In my case, I prefer not to show my models to anybody, although I am a founding member of an association of French renters
who buy in common cars, vans and trucks. My models are happily more simple compared to the current models I can see and
I still prefer not to show them to my associates because I do not want to be considered as a scientific out of reality.
The only thing I can do is try to explain in words the reasons of the policies. They will understand that.

I will certainly continue to contribute to this forum because it is useful to me and to other people.
You have the chance to do a very interesting job.
Best regards.
JJ

[Ralf]

Hi JJ,
hi Lee,
hi to all readers,

I really like this forum as it offers totally different perspectives and frames on SD issues (that are actually all around us;-)).

It really depends, on what we define as SD?

Is is just the modeling or perhaps more the process on getting along the path with business partners, colleagues, students, pupils or even friends (totally out of the SD issue;-))?

For me and my business SD offers a way and an "easy" language to clear up the foggy reality!

Challenging the mental models through first using CLD, reference modes and further on into practical modeling that leads to further LEARNING is essential to me.

I am really thankful to John Sterman, who answered and inquiry some three years ago on where to get the Beergame. He mentioned the Conference in Nijmegen to me and this has been the STARTING POINT to get deeper into the issue.

@JJ, of course colleagues will frown at first at the CLD and your models - if presented to them at a ready to go state. Working with them in slow movement so they can see the benefit by themselves is time consuming and yet the ONLY WAY to bring the field forward and more widely spread (especially in understanding!)..

The whole economy is an action research field and we as practioners in our fields have all the power to connect the strings (business with SD:-)).

Looking forward to seeing Jay Forrester's dream in reality in some time:-))

All the best

Ralf

[RandyPark]

Hi all,
I am relatively new to SD and Vensim, but from a physics and engineering background, so it seems quite familiar.

As far as the question of sharing models, it seems to me that the business opportunities for those of us who want to also make a living from SD work lie in the customization of the models to clients, and the revelations produced by the model performance.

If correct, it means that the core models could in many cases be shared and/or jointly developed, with the customization being confidential.

As an example, I am developing a model for the rise and fall of membership in an organization such as an association or a club. In and out flows depend on things like retirement rates, pool of potential members, member satisfaction, visibility of the association/club, limits of facilities if any, and other issues. Each organization will be different; working with them will help them discover the leverage points, and reveal the assumptions they have been making in the past (which is a large part of my work.)

Best regards,

Randy Park

[LAUJJL]

Hi Randy
There is maybe a market for such a model as this problematic is shared by many associations.
In fact the problem is not only for association, but for service businesses too.
Are generic models reusable with ad hoc parameterization?
Bob Eberlein mentioned once that Ventana had tried in the past to provide customizable models but found out that due to the high level of peculiarities of the customers, it was better to build a new model from scratch, eventually using the knowledge developed by some generic models.
The large scope of the applications explains too this approach.
But it is always possible to try providing a generic model.
One another reason generic models are not very efficient, is that to my opinion a model to be useful must be perfectly understood by the customer and that can only be accomplished by his close participation to the modeling effort. Providing a customizable model will not be convincing enough to my opinion for the end user to make the considerable efforts necessary to apply the policies model because he will not be totally persuaded of the potential outcome.
A model is unfortunately complex to understand, even for those who have built it and much more for somebody who did not.
Another thing about modeling is that the outcome of modeling does not necessarily come from the model itself, but from the work made to build it, gather the data, thinking about the problematic etc.., and there is a risk that this kind of work may be avoided with a customizable model.
Another question about your model, is that before trying to commercialize it, you must learn to use it once finished and to prove that it can truly deliver a sufficient added value to justify the effort of the client to pay for it in money and time. There are to my opinion in the SD community plenty of theoretical models that ‘should’ be useful, but the only way to demonstrate the usability of a model is to use it on a concrete problem and to verify later on, the effective benefit of it and that the client was satisfied.
Regards.
JJ

[Ralf]

Hi Jean-Jaques,
hi Randy,

A generic model could be -in my view- perfectly developed: by US.

Using a specific client problem always leads to confidential problems. And the knowledge will be lost for others.

When we start building around a model for a problem, we are all part up, there is a much greater chance that the knowledge will spread (is that wished, by the way?), we all learn from building it, we can transform parts of it into our very own models.

Right now there is no web-based version that could be jointly used by several supporters. As far as I know Bob and his team are working on something like that.

Best regards, Ralf

PS.: Sharing is GAINING:-)

[LAUJJL]

Hi Ralf

I do not know at what level of learning you are.

I am not interested by theoretical learning, which for me is building models that are not related to any observable reality. Eveything else is for me speculations without interest.
As Einstein says, one learns through experience, and building a model not related to something real, is not for me an interesting experience, at least at the level of learning I am in.
You are right when you say that client problems lead to conidential problems that are difficult to share.
But this is not a reason to focus on models that 'should' be good'. For me a good model is a model that has delivered a useful outcome, and to do that it must be related to a real problem.
Maybe it is difficult to find a real problem, but I prefer not to do anything than to do something not based on it.
By the way, I posted some years ago a model of the Society Akzenta, that was growing on a bubble at the SD business group that predicted an unavoidable bankruptcy. I never had any comment from it, and it was a real problem.
Akzenta (a german firma based in Munchen) has been forced to bankruptcy this year.
Who is in fact interested by real problems in the SD community mainly composed by academics and researchers?
Best regards.
JJ

[chester]

Hi,

For some reason i can't download the attachment. I'm quite curious about this one as well because this happens to me a lot. I'd like to view the attachment if it is indeed the same issue as mine.


Regards,
Chester


[Edited on 11-20-2009 by chester]

[LAUJJL]

Which attachment?

JJ

[Ralf]

Hi JJ,

I only deal with real data and event not superficial theories. Probably my recent postings sounded somewhat way ahead of what today is feasible to do with system dynamics.

I see however a constraint, that hinders us to make the use of system dynamics modelling be useful for us (in the learning phase) and for potential clients (in the consulting phase):

there is yet no tool for collaboratively working together in realtime on CLD, model!

[LAUJJL]

Hi Ralf

Collaborative tools can help if well designed. But I do not think that the tools are the problem with SD.
And I do not believe in interactive tools at least for SD.
The tools have been improving a lot since the last 20 years, but SD did not progress with it. The problem is more fondamental and has to do with a misperception of what SD can be useful to from the client that generates dreamed needs that cannot be handled by SDers and that SDers fail to demonstrate their impossibility to satisfy.
In fact the SDer fears to lose the customer if he explains that SD does not make miracle, that it is a tool to think, that it does not predict the future and that if one wants to model the reality as it is, one gets a 'model' that is more complex than the reality to handle. Result two complex problems instead of one.
I think that the client wants the modeller to model the reality
instead of asking for a model that is useful, even if it does not reproduce the reality. The Sder should be able to resist to this client pressure, which from my experience with consultant is very improbable.
SD needs too a lot of work from the client and to my opinion a good understanding of the theory of SD and SD practice too.
Regards.
JJ

[Ralf]

Hi JJ,

Thanks for clarifying on perceived reality. Keeping authentic as a consultant when working with a client where the client has 50% of the work to do is a challenge.

The issue I opened is that SD people (either practitioners working as employees within organizations, students or consultants using SD) work -at present- individually across the globe. There is not much collaboration amongst them and across borders.

What would we like to see SD to bring to the world?

[LAUJJL]

Hi Ralf

I do not know if there is 50% of work for the client, but there is certainly a lot of work, especially when it comes to understand how the model works and trying to find implementable and useful policies and implement them.

There are two ideas in your post.
The first Is that people do not help each others. I do not think that SDers are very different from other people. The problem is that a modeler can only model what he knows from a problem and to help somebody else he should have a sufficient knowledge of his problem which is very unlikely. This is why on the SDS forum people discuss about very general ideas and very rarely about a concrete problem. This situation has generated a disinterest about other’s problems being not able to help them. Even on this forum, the only help that can be given is technical and if one tries to deep more into the problematic of anybody one gets very quickly to the knowledge limitation. And helping people only for technical language problems is probably not very exciting for most SDers.
I thought that by following courses in the past I would get somebody cast an eye on my problems, but even with this, I never got any little concrete advice on my models. If you want to share your problems with somebody you will have to pay a consultant, there is to my opinion no other solution. I am not even sure that if you followed a course at Boston at the MIT, it would change something. Maybe doing a PHD and preferably with a teacher with a lot of practical experience? Consultancy is quicker and goes right at the heart of the problem. You do not have to follow courses during years to finally realize that you are not able to do anything concrete.
Did you find any collaboration in the business special interest group? The last post not concerning administrative information like the business meeting at the next conference, is from June 2008 and it is from me.
There has been mainly three contributors, you, Kim Warren and me in the past. This is not much and is really depressing. The best thing not to be depressed is to forget it and quit.
I really tried to collaborate in the past, but I will probably not renew my membership at the SDS the next year. I do not feel comfortable with SDS people who are mainly academics and researchers.
About what I would like to see SD to bring to the world?
You mean SDers, because SD cannot bring anything by itself.
I would like that its expectation and pretention to bring anything to the world to decrease considerably and what it actually brings to the world to increase just a little, which is not so difficult, considering the level of it.
Regards.
JJ

[Edited on 20-11-2009 by LAUJJL]

[Ralf]

Hi JJ,

seems to be that after a long time of stillness here in the forum our minds are somehow connected and generating conversations bringing out our thinking (mental models so to say).

You mentioned "Consultancy is quicker and goes right at the heart of the problem. You do not have to follow courses during years to finally realize that you are not able to do anything concrete. "

It depends what one understands under the word "Consultancy". If it is the expert or doctor as described in Edgar Schein's "Process Consultation" than surely the fixes seem quick and relevant (only to last for the moment). On the other hand we can help clients to understand by themselves what is going on in their operations and organizations. The Standard Method by Jim Hines gives a neat and easy to do approach (just one link http://ocw.mit.edu/NR/rdonlyres/Sloan-S ... method.pdf).

The client feels even uneasy in that really early stage of finding the reference modes and building up the causal loop diagrams. That is -in my eyes- the leverage point to connect with the problems the client have.

...and here comes in the constraint: the technical tools to collaboratively work on such issues not only physically in the client's office but via the internet in a remote and sustainable fashion.

I have done that with a friend in New Zealand via Skype over the period of two years. He worked right in the organization and I was never in New Zealand. I can tell probably more about the dynamics within the company than their consultant who is improving processes, cost cutting and other advices.

It will be that client, systemdynamicist and practioner networks that will bring the field forward to more broader use. One small step for example was my giving away a bounded Road Maps to the State Library, SLUB, here in Dresden. People will have a chance to read about system dynamics and eventually they connect it with their own experience.

Would you rather do nothing? What could be another feasable approach?

Regards, Ralf

[VerAye]

If you do nothing what will happen? There will be no result if you do not do nothing. If you will wait the time that do this I think it will not show the thing that you are expecting.


_________________
Tool & Die

[Ralf]

Hi VerAye,

Very truely said - so to say a vicious cyle

It has to be broken first to lead to a better direction. And that is definitely pure system dynamics;-)

Aren't we acting too often blinded in our own way of thinking without opening our eyes for what our mind thinks is not possible?

[LAUJJL]

here is an easier to understand qualitative model of the shifting of the burden problem. It has only 4 loops easy to understand and much metter to my opinion than the prop1 and prop2 models.
To be quantified it probably needs realistic examples like the debt of a state or anything else.
The proper model is prop3_diag and not prop3. Sorry for the error.
Regards.
JJ

[Edited on 19-9-2010 by LAUJJL]

[Edited on 19-9-2010 by LAUJJL]

[Edited on 20-9-2010 by LAUJJL]

[LAUJJL]

A second qualitative model that adds a loop restricting the Santa Claus effect by limiting its working horizon.
JJ

[Edited on 20-9-2010 by LAUJJL]

[Edited on 20-9-2010 by LAUJJL]