Cost-of-crowding model for light rail train and platform length

Abstract

With light rail transit (LRT) and other similar rail-based commuter transit systems, train and associated station platform length provides an added dimension of flexibility not available to buses. Train and platform lengths are important factors in the planning and expansion phases of a network. Existing cost models that determine optimal headway by combining passenger and operational costs provide headways that are small and close to a logistical minimum (2–3 min); this type of standard waiting cost model is not sensitive to train and platform length. In this paper, on-board crowding is used as a cost factor and a cost-of-crowding model is developed from supporting psychological research. Two models are proposed and optimized with respect to train length to determine the optimal train and platform length for a many-to-one peak period commuter LRT system. Data from the C-Train network in Calgary, Alberta is used for numerical analysis of the model. The model demonstrated that crowding has an effect on optimal train length. The model produced feasible results when applied to a real-world scenario.