System Dynamics is a popular method of complexity reduction.
It was invented back in the 1970s by a team of scientists at Massachusetts Institute of Technology (MIT) in Cambrige, led by Jay Wright Forrester (also popular for inventing an early concept of RAM, computer’s Random Access Memory, and the Whirlwind project).
Software in System Dynamics (SD) is able to simulate a Balanced Scorecard over time.
That’s a number of Input Values, Targets and Formulas on a sheet of paper, describing any technical or natural dynamic system (for example, a production circle in a company), changing over time. Therefore, SD is used for behaviour prediction, amortisation, and all kinds of cost-usage calculations. You can see a Balanced Scorecard being just a simple overview of the situation on a sheet of paper.
At Dynamic Applications, we are developing a free Business Dynamics Platform.
We feel that the general approach of System Dynamics is outstanding, and way too good to be forgotten. When talking about Business Dynamics, this basically means SD for business applications. So we invent all kinds of 21st century micro-business models, and write calculations in Desire, our very own System Dynamics-based formula language.
A Balanced Scorecard is basically a sheet of paper where you put all main topics of interest of your discussion into a circle. It could be a cost, a benefit, a profit, a target, a desire. Then, you interconnect these circles with any number of arrows. So you get a good picture of the situation, and are able to see the interdependencies, even if the system or topic of discussion is complex. You get a pretty good overview pretty fast. As Wikipedia describes, Balanced Scorecards are forming the qualitative approach of System Dynamics. It is a Top-Down approach. More thinking gives larger diagrams, more precise, but after a while the overview gets lost. World’s details are endless. So a good Balanced Scorecard is in balance, itself. Not too complex, not too detailed. No arrows for rare cases of interconnection.
However, there is one Key Benefit: Understanding.
Basic understanding of the interconnections any kind of complex problem, like developing a new product, a service, or a specific, fine-tuned solution in detail. World’s details are endless, so there’s always the problem of missing the big picture. This is the kind of problem that Balanced Scorecards can solve. Understanding of all Interactions. Getting an overview of your current situation.
Oh, by the way. Lots of people in the SD community are talking about causal loop diagrams. It’s basically teh same thing, We just liked the word Balanced Scorecard, as it sounds a little more easy to understand. Expert language, you know. There must be a clue in making things sound more complicated as they are (thanks to Robert Koshinskie from North Carolina for his fine review and contribution to this article at this point). Let’s see this paragraph as a contribution to the (exceptional) expert reader, so we don’t lose or start confusing any of them ahead of time.
The team around Jay W. Forrester were working on something better. They were developing an approach to not only look at the sheet of paper. They began to separate the things on that sheet from each other. Circles with a single or multiple arrows starting from them, but no incoming arrows, were typically describing input parameters. Like steering wheels for everything that follows. Circles with arrows going through them were forming formulas. Like, A+B = C, where A and B are Input Values, C is the Target, the result, and the whole thing is a Formula. Circles where all those arrows were ending in, are forming the main Targets of Interest, then. The things to optimize.
And so, these people were developing a methodology to quantify basically any kind of Balanced Scorecard. Then, they put their methodology, their procedure, their algorithms in software, and developed the first specific programming language suitable to describe complex problems: DYNAMO.
After working a while on DYNAMO, they were the first team in the world to feel the capabilities of the approach. So they decided to work on something great. In a time when about everyone around, every country, every nation in the world were industrializing and re-building the world destroyed by World War II as quickly as possible, these people started to develop World Models. Model I was just a prototype. The results were poor, and about noone took interest in the idea. So they developed World Model II. A much better approach, so good that Forrester himself wrote a first book about it, World Dynamics. Apart from a few researchers though, nobody took notice.
So they developed World Model III.
Landscape, Resources, Food per Capita, per Head. And prediction of developments, as good as possible. And results were showing that the proceedings of mankind were pretty similar to what Algae do with a fresh garden’s pond. They populate it, then they get greedy, others are dying, then Oxygen gets low, and in the very end, almost all Algae are dying from greed, their cells falling to the bottom, and the garden’s pond is clean again. Only very few Algae remain to re-populate it as soon as resources and food would ever return.
Available Resources. Reduction of Resources by Mining and Oil Production. Worldwide pollution of air, rivers, garbage hills, and ground water. The results hit them like a thunderstroke. And when nobody took notice in the U.S., Forrester decided to speak to a few people forming a discussion circle, an early “think tank”, in Switzerland.
The Club of Rome.
And when the Club of Rome became aware of the results and had verified them, they wrote their own book about the limits of growth, of world-wide exaggeration. At a time when the first oil crisis was just through, the world was ready for a new message. Bigger, Higher, more Spectacular. Limited.
The Limits to Growth became a world wide bestseller in the last 1970s.
There’s one important thing to know about these kind of predictions, though:
they are made to predict scenarios that are not going to become true! So don’t you fear. This is because the system just shows a
The simulation gives everyone transparency about the situation and its consequences. Then, as with any complex scenario there’s always a hundred or more Input Parameters influencing the outcome of the simulation. In this case, since the calculation goes over more than 100 years, there’s enough time to solve the problem.
So that’s what we mean by transparency as the central benefit, and not the Target Values, as such. In every System Dynamics calculation, there are various Input Parameters, so in the end, all Targets are adjustable. The outcome, your optimal solution, depends on what you think realistic to achieve, what you are willing to do, and eventually, which of your Targets you are willing to sacrifice, to optimize a more important one.
At Dynamic Applications, we felt it’s a little sad that such a fine approach is nowadays almost forgotten, or never really made it past science. The world around us is changing, and things like the Internet have been invented, that haven’t been thought of in these days. Some people say progress is accelerating, as much as computing power does. People start looking for orientation. For understanding. So wedecided to develop simulation models for problems of public interest, like, for example, a forecast on your own bank account’s development perspectives. A free platform to run them and get to know how to solve any kind of complex problem, whether in School, at University, or in Business. Our Vision.
Our platform, Perfect Desire, can be used for Return-on-Invest (ROI) calculations, prognosis and simulation of system and value developments over time, and amortization calculations. Any operative system that features Input Parameters, Formulas, and Target Values calculated by these formulas is suitable for simulation. This may as well be a production cycle, any interaction of things in nature, business and finance, or a critical situation with limited possibilities, a strategic invest decision with long-year consequences, measured by numbers as good as possible.
Combining all the knowledge understood about single aspects in one simulation.
Example: Photovoltaic System.
A good example of System Dynamics is to calculate the amortization calculation of a Photovoltaic System (PV System) over time. Amortization against a projection net price developments of electricity, as described here.
It can’t give a final answer, as we all do not know the net price for electricity in the year 2040, when a today-built PV system will reach its end-of-lifecylcle.
But it allows us to get a pretty good picture of the interactions of all those input parameters, like sunshine, battery, and average household usage profiles.
In all these situations, we are typically looking for optimal behaviour – formally spoken, a parameterization of adressable input values to get a maximum, minimum, or balanced behaviour of a specific, interesting target value over time.
Squaring the Circle
If we can find this parametrization and the relevant formulas of the describing model are correct, process of application can be adjusted accordingly, and the proposed results will occur in due time.
So by applying System Dynamics, we are solving optimization problems. Squaring the Circle, that’s what we do here. A very typically solution. In most cases, the best solution is not either extreme.
It’s the optimally weighted point between them.
And the quantification of Balanced Scorecards will help us to find a measure.
We’re aiming at the best possible measure we can get.
The special thing about a System Dynamics Application is that it allows you to change any single input parameter over time, even in combination, and take a look at results beforehand. To reflect that in nature about every existing thing will change over time, and conclude from that.
In the most easy approach, this can be done from past to present, so the outcome is alredy known – a verification of approach, to make sure the formulas are working. However, the same approach can be used to describe the changes of input parameters (personal and material cost, revenue, sales) anticipated from present to future. The remaining guess, the uncertainity here is that we can’t usually forecast things not yet invented.
As the software always works in the same way, the results will be correct in both cases. So, to predict future changes, it is just necessary to anticipate all influential parameters, correctly. The good thing is that whenever we are planning something, we use to guess upon the future change of driving parameters anyway.
Just most people would combine these in their mind, or search for a compromise in discussion. Theory says that a standard person is able to combine 2-3 different, changing parameters in their mind. A genious like Albert Einstein could have done 4. About noone can combine the results of five or more changing parameters in complex interaction in their mind.
But a Computer can do it.
A System Dynamics software application will allow you to anticipate as many changing parameters as you want, and still be able calculate all formulas, resulting target values, and consequences correctly, restricted only by the accuracy and detailedness of the underlying formula.
A collection of typical basic interactions has been researched to describe common issues.
The list of System Archetypes, as found on Wikipedia, describes typical problem patterns:
- Growth but under-Investment.
- Shifting the Burden.
- Tragedy of all Common Goods.
- Unintended Opposition.
- Eroding Goals.
- Fixes that Fail.
- Success to the Successful.
- The Limits to Growth.
- Balancing a Process with Delay.
Wikipedia has a pretty nice explanatory diagram for each of these System Archetypes.
At Dynamic Applications, we feel this approach can very well be adapted to a typical company, as it covers all well-known patterns of typical Business Management situations.
To conclude, for a complex model as the one describing a photovoltaic installation, system dynamics is a very suitable approach. It allows for the necessary complexity reduction and calculates the consequences of the various described interactions part-by-part.
To deliver a very good result in the final formula, amortization. Balance. Transparency. Verify your own thoughts. Insert your own estimate on parameter proceedings and see the combined result. Fine-tune as many input parameters as you want.
See the proposed outcome, the consequences, immediately. It allows you to play through all kinds of possibilities. It allows you to verify that you haven’t overlooked an alternative. It allows you to find new alternatives, measure them, validate their consequences, and be ahead of competition anytime.
We think the general picture here, in Business, in Capitalism and free Entrepreneurship, has pretty much in common with playing the game of Chess.
There’s always these people around who swear on Tactics. Concentrate on the next move. Evaluate your situation, your history. How was the current situation developing? – what did our counterpart do? – and what is the next best move. There are people around who use days, weeks, months, or even years of their life to memorize the best known development openings of their favourite board game. And that’s pretty similar to today’s big data approach in business. Hunt for more data, more understanding of the past. And collect as many of them as possible. It seems intuitively right, as it addresses hunting and collecting in one concept. And the latest hype, Big Data Analytics, will even be able to show you statistical interactions within data of any kind, without specific programming of any kind. 75% of your customers older than 45 years will prefer product B instead of A. And then, all of these tactically driven companies start to develop key score criteria, just to be able to raise them from month to month, from year to year. Only a thinker would put that in question.
A System’s Thinker (basically a System Dynamics user, who learned to think over time) would focus on the long-term target instead, on developing either a survival Strategy, or a winning strategy. A System’s Thinker wouldn’t read many books, or take a look at all those BigData Charts. They are useful to deduct a couple of formula from them, as you can derive them from Google or Wikipedia on about every important thing invented so far.
The Strategical Analyst would naturally desire a System Dynamics application, and simulate all possibilities in advance. Move 7, 8, 9 input values, like fine-tuning sliders, right after each other. Preview the consequences. Test everything. Get an overview of possibilities. And then, decide for the best strategy that there is around. Pretty similar to playing chess, or any other board game around, isn’t it?
Strategy against Tactics.
If you want to see how that works, we recommend you download Startup Product Manager, our free System Dynamics simulation environment, and prepare some really advanced questions for your free consultancy day.
Developing a platform in System Dynamics, you become pretty good in thinking about new and possible scenarios, and evaluate possible alternatives even in the most difficult situation. And always remember:
No problem is too complex to be solved.
In case you’re interested, we offer an introductory, free consultancy day in our article About Dynamic Applications.
We’re pretty sure that after this one day, you’ll find a lot of more problems to be solved for the rest of our common lifetime. You win a free consultancy day, we win a customer, as sooner or later you are going to come back to us. Only if you want, of course.
Reference: Wikipedia on System Dynamics.