Cities and their planning seem to be unpredictable, when looking back in the past 40 years
Yet we feel like we can build models to make predictions about complex systems
In the past 75 years, science has moved from a stance where we believed in 'good' prediction to prediction is now 'difficult'
Coronavirus has proven that even short term events may no longer be predictable
Because of a model's uncertainty, to ensure models are useful one must use many models
Transportation modeling difficult because we continually invent and reinvent how cities operate by changing the conditions, and this creates numerous feedback loops that is impossible to model
It is often accepted that shorter term forecasting is more accurate than long term forecasting, ==as if the amount of change is dependent on time - it is not - it is dependent on the kind of change==
Change is never smooth, always volatile
==Remove the illusion that cities can absorb the change easily with little disruption==
==This is probably because the use of continuous values in models==