A mixed frontier model for urban bus performance evaluation

Bus demand fluctuation at different time periods is one of the most intangible problems for bus operators, which seriously affects operational efficiency. A good knowledge of performance variation at different periods is helpful in developing an effective transit schedule. The aim of this study is to evaluate the efficiency of an urban bus operation at different periods using a mixed frontier model and to explore a better operating strategy which is adaptive to bus demand fluctuation. A mixed data envelopment analysis–stochastic frontier analysis (DEA–SFA) model is proposed, in which multiple outputs are aggregated by the DEA model, and then the technical efficiency is calculated by SFA. The model deals with multiple outputs while considering the stochastic factor simultaneously, so as to ensure model practicability. Using field data from Shanghai Transit Route 55, this study evaluates efficiency at different periods by analysing various aspects. The results indicate that the model effectively describes the characteristics of the bus line efficiency with time period changes, which provides an innovative perspective to solve the operational problem of the urban transit system.