Appendix D Model audit checklist
Check if a model specification and detail is available for:
- The type of model used
- Geographic area covered by the model and the level of zonal disaggregation
- Tabulation of land use assumptions for all modelled years
- The transport networks (active, public transport ,highway, freight) included and their details
- The time periods modelled
- The vehicle types modelled
- How the external trips are modelled.
Check if a data source description and the source’s reliability are available for:
- Transport network data (link lengths, link types, free flow speeds, capacities, number of lanes)
- Travel data and collection methods (traffic counts, origin–destination surveys)
- Source and coverage of travel data and collection methods (household travel surveys, origin-destination surveys, stated preference surveys)
Check for details on:
- The description of each step in the development of the Reference (Base) Year and forecast trip matrices, including methods, assumptions, parameters and factors applied – include details of any matrix estimation procedures that have been used
- Evidence of matrix fit to observed data (screenlines, comparison with independent origin–destination flows)
- Evidence of sensitivity testing of input parameters or elasticities
- Detail on the matrix methods or techniques used – if variable matrix methods or growth constraint techniques have been used, provide details on the method and parameters adopted and the justification for the approach
- Description of the trip generation, trip distribution and mode choice modules by trip purpose, including a listing of all variables.
- Description of special trip generators (airports, ports).
- The basis for the development of a commercial vehicle matrix.
Check for details on the:
- Description of how the transport network was developed
- Assignment method used (incremental, equilibrium, volume-averaging, other)
- Generalised cost formulations used for route choice to include the methodology for incorporating tolls on route choice
- Volume delay functions adopted (equations, coefficients, calibration, validation)
- Any intersection modelling being undertaken and the basis it is being undertaken on.
- Outcomes of the model validation tests for each step of the 4-step modelling system.
Check for details on:
- Comparison of forecast year growth rates with historical trends (land use, household size, car ownership, traffic volumes, commercial vehicle volumes)
- Average growth across screenlines to ensure local growth is reasonable
- Comparisons with other forecasts (if any).