1.4 OVERCOMING OUR SOMETIMES CONSTRAINED VIEW OF THE WORLD
Unfortunately, it is difficult to develop dynamic hypotheses in certain situations where cause and effect are not proximate in either time or space. This is exacerbated if our window on the world has a small aperture or does not allow us to see events over a sufficiently long time horizon, or from a highly informative perspective.
Imagine living in a basement that has only one small window through which we see the legs and feet of people in the street above. By observing the scurrying feet we can deduce patterns of behaviour, such as when activity levels are high or low. In the morning we wake from our sleep to hear the city noise, the traffic in the street and see the feet and legs scurrying by. If it is a week day, the scurrying will be hectic. If it is Saturday or Sunday the traffic sounds and scurrying feet will be less.
If inclined to do so, we might also observe how well individuals maintain their footwear. That will tell us something about those who wear the shoes. We might even be able to hypothesise about their individual wealth. Based on how quickly and freely they walk we might be able to hypothesise about their health. But, we cannot learn much about their world, why they scurry as they do, or where they go.
If we lived at street level with floor-to-ceiling glass windows that give us a panoramic view we would have a totally different view of the world. If we could fly around in a helicopter we could see the patterns of scurrying as people drive to work or travel by train, plane or boat. Imagine how much we could learn by taking a video of people’s movements during the day and replaying that in the fast-forward mode. In part at least, this is what we try to do when we build models in the time domain, and simulate them.
Time is of the essence. Developing dynamic hypothesis is an iterative process through which we build and review hypotheses of how time sequences produce reference modes of behaviour. The graphs over time developed for each of the important variables are the reference behaviour modes for the development of system dynamics models. One essential test of a system dynamics model is its ability to reproduce the reference modes identified at the outset (during problem conceptualisation). So, it makes a lot of sense to build models which enable our analysis in the time domain. We can now build models that accelerate timeframes, change our time horizons and extend our visualisation into the future, or replay what we might have observed in the past.
Further, even those of us who do not have advanced skills in mathematics can build models that provide powerful insights into the complexity we see around us. This book enables us to develop those insights by harnessing the power of system dynamics modelling.