Improving Projects Evaluations in Analyzing Transport Models Ability to Explain the Past

GLOBAL, the RATP transportation model, was recently updated to its 9th version. This was the opportunity to reconsider its methodology, and, more generally, the relevance of four-stage models for medium-to-long term planning studies. Indeed, a model's validity is generally assessed through its ability to replicate traffic counts. Considering that a good replication of the present leads to an accurate forecast of the future is an important assumption. It mainly implies a hypothesis of stability of mobility behavior, which could be questionable, and even more when many parameters are used, as it is the case for a large and complex network. The issue was addressed by carrying out the exercise in reverse, that is to say by analyzing the model's ability to explain the past. GLOBAL 9th version, estimated using the 2010/2011 Parisian household travel survey has been applied on all the past household survey years (1976, 1983, 1991 and 2001). This way, estimates could be compared with survey results in terms of demand volumes, mode shares and also with past public transport counts on main railway lines.The analysis of the discrepancies between the estimates and survey results clearly shows the limits of the mobility behavior stability hypothesis. Indeed, this assumption gives the model a high inertia, which is important to keep in mind when examining and interpreting modelling results. However, the discrepancies of predicted demand flows versus observed ones stay within an acceptable range. The exercise was then a good illustration of the uncertainties associated with the modeling results.