How relevant is the aggregation bias with regards to mode choice?

Abstract

Most travel demand models used in practice still apply a four-step approach which describes travel demand in an aggregated manner. Aggregation bias refers to the assumption that group characteristics are shared by all the individuals who are members of that group as compared to an approach that evaluates travel behaviour on the level of individuals. The use of average values applied to aggregated populations across spatial zones and time periods distorts a model’s sensitivity to investment and policy alternatives (Castiglione u. a. 2015).

Our paper examines the relevance and impact of aggregation bias with regards to trip- and tour-based mode choice. The paper starts with a short literature review and a theoretical description of the problem. The main part of the paper consists of two case studies which we have developed to quantitatively assess the relevance of the aggregation bias.

The first case study is based on synthetic data that reflects typical rates of mobility tool ownership and travel times by different modes for a given travel relation in an urban, agglomeration and rural context in Switzerland. The relevance of the aggregation bias is assessed by different ways to consider mobility tool ownership and systematic adaptations of the considered travel times.

The second case study is based on the implementation of an activity-based and disaggregated demand model for the German city of Halle based on which an aggregated model is derived that shared exact same transport network and overall travel demand. We use the two models to assess the relevance of the aggregation bias based on three scenarios with different transport planning measures.