Constrained multinomial Probit route choice modeling for passengers in large-scaled metro networks in China

Abstract

Abstract Considering that a large scaled metro network provides the opportunity of multiple route choice, it is necessary to consider integrating the impacts of routes set and the interdependency among alternative routes on route choice probability into route choice modeling. Therefore, a constrained multinomial probit (CMNP) route choice model in the large scaled metro network is proposed in this paper. The utility function is formulated to be composed of the following three components: the compensatory component is a linear function of level of service variables and route direction measurement, such as in-vehicle travel time, number of transfers, transfer time, congestion level and revised angular cost; the non-compensatory component represented by the logarithm function of a binary probit equation denoting the relationship between the constrained attributes and the corresponding thresholds measures the impact of considered probability of one route on the route’s utility; following a multivariate normal distribution, the covariance of the error component is structured into two parts, that is, the part measuring the correlation among routes, and the part denoting the unobserved variance distributed independently by route. Based on the surveyed revealed preference data in the Guangzhou Metro system, the estimations show that the proposed CMNP model shows the superiority of goodness- of-fit to data over traditional models. Meanwhile, the results also indicate that the non-compensatory component in the CMNP model works well to explain the impact of routes set on route choice probability.