A Conditioned Model for Choice of Mode Under Information

Abstract

This paper examines the influence of time and cost information obtained from different sources on choice of mode of Leeds' long distance travellers. The choice of mode was investigated through modal attributes provided by at least two different information sources which might provide contrary or corroborating information rather than on actual attributes. The experiment included telephone administered questionnaire including RP (Revealed Preference) questions and an SP (Stated Preference) exercise dealing with the choice of modes conditioned by the information received from various sources. Information on travel time and cost was provided from two different information sources for each mode to facilitate the conditioning of mode choice on corroborating/contradictory information. The research employs a wide range of modelling methodologies and examines a range of traditional and newly developed calibration and estimation procedures including Mixed Logit models with individual specific parameters and the newly developed RRM (Random Regret Minimisation) framework. The study confirms that the market share of the modes increases when information sources show decreased travel time and cost values and shows that the maximum shares are achieved when different information sources give the same information to the travellers. The study found that pre-trip time information has more influence on mode choice when derived from websites than when derived from other sources. Pre-trip information on costs was, however, less influential when derived from websites than when derived from other sources.

Keywords

• Choice Modelling
• Travel Information
• Mixed Logit
• Random Regret Model
• Mode Choice