activitysim

activitysim copied to clipboard

@guyrousseau - For instance in mode choice, a sub-routine could be added in the source code that allows the user to enter calibration target values, and the program could adjust coefficients until the maximum number of iterations is reached, or the results fall within a user specified closure criteria. Such sub-routine could also summarize the available person trips and compute an implied mode share using observed targets. This could help shed some light on the target values, as well as potential issues with models that are upstream of mode choice. It could indicate certain market segments where perhaps the mode choice model has to compensate for relative inaccuracies occurring upstream, just as an example.

Thanks Ben for adding this, hence allowing us to potentially explore the option of adding a built-in, self-calibration routine in mode choice calibration, which could systematically adjust constants to match observed transit shares.

The SEMCOG team (@JilanChen and others) is prototyping some cool functionality for this based on improvements made for estimation integration