Application of user preference choice models in the city of Johannesburg demand model

Abstract

Transport models and choice models are often developed in isolation without the benefit of combining it into a single dynamic model. This paper aims to demonstrate that, based on the City of Johannesburg (CoJ) Model, choice models developed from stated preference surveys can be even more valuable when designed to be incorporated into a transport model. This paper describes how the choice models, developed from the stated preference data, supported with reveal preference data from the Household Travel Surveys, was applied in the CoJ Model. The mode choice model includes the intrinsic characteristic of city travellers which are either captive to public transport, mainly due to affordability, or to private transport, mainly due to lifestyle perceptions. An iterative procedure is used whereby mode-specific performance statistics on in-vehicle travel time, walking time, waiting time, travel cost, number of transfers and seat availability from the spatially represent transport model is used to calculate modal splits which can be assigned to the transport network. The transport model allows for the following modes; namely private vehicles, taxis, buses, Bus Rapid Transit (BRT) buses, rail (Metrorail) and Gautrain. The paper shows outputs from hypothetical scenarios based on road tolls, fuel price, BRT headways, and rail improvements to demonstrate the sensitivity in modal shift.