A Design Model for Rapid Transit Networks Considering Rolling Stock's Reliability and Redistribution of Services During Disruptions

Abstract

This paper presents a model for designing a public transit network system combining the traditional approach of transport demand coverage in bimodal scenarios of operation with the recovery of possible disruptions due to limited reliability of the rolling stock. The model balances construction and operational costs with the benefits to the users for the optimization of their travel times. Two transportation modes have been considered, public and private transport and the proportion of the users choosing one mode or the other is assumed to obey to a bimodal logit choice model. While construction costs are a first stage decision, user travel costs and recovery action costs are scenario dependent. Two types of scenarios are taken into account: a) the scenarios of normal operation and b) disruption scenarios which are associated to a link's breakdown of the network. The disruptions in the links are assumed to follow a probability disruption model accordingly to the number of services that operate on them. The model can be used to analyze the influence of the rate of failures of the units on the reliability of the designed RTN. The proposed model can be considered as a two recourse stochastic programming model with a bi-level structure where the probabilities of failure are an implicit function of the number of services and the routing of the transit lines of the transport system. A heuristic solution method is examined for small to medium networks demonstrating the computational viability of the approach. © Springer International Publishing Switzerland 2015.