MobiSim is a data efficient platform that avoids over-calibration (frequently leading to ad hoc models and finally to poor knowledge improvement). Four categories of input data are used in order to parameterize the platform for different study areas and to generate comprehensive scenarios.
Standard information is used in initializing MobiSim: socio-economic databases and fine scale geographic data. This kind of information is usually available for all European regions from administrative surveys or free open-source data.
Local databases giving information about local behaviours and practices of individuals are used to calibrate MobiSim. MobiSim is provided with a calibration dataset by default, which has to be modified for each study area using appropriate local data to allow for local specificities.
Ad hoc surveys
This kind of data is used to validate the model by measuring the differences between the results produced by MobiSim and actual observations on the ground. Those data must therefore be different from the data used for calibration in order to avoid the Snake biting its own tail problem.
Local development plans, forward-looking documents, interviews of local experts and stakeholders
This information is required for imagining realistic planning scenarios when using MobiSim in the companion modelling framework. It also includes political concerns about the desired planning policies for the future.