A flow-based integer programming approach to design an interurban shared automated vehicle system and assess its financial viability. (July 2021)

• Record Type: Journal Article
• Authors: Santos, Gonçalo Gonçalves Duarte; Correia, Gonçalo Homem de Almeida
• Is Part Of: Transportation research. Volume 128(2021)
• Publication Date: 2021-07
• Subjects: Shared automated vehicle -- Fleet size -- Optimization -- Interurban transport -- Linear programming; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2021.103092
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: We propose a flow-based integer programming model to design interurban SAV systems. The optimal results were the basis to produce a financial viability assessment. The case study was an interurban transport network in the central area of Portugal. A mixed fleet of cars and minibuses is the best option for profit maximization. The studied SAV system can replace up to 18 private vehicles per SAV vehicle. Abstract: Shared automated vehicles (SAV) are expected to become part of on-demand transport systems in the future. In an interurban context, SAV systems may potentially contribute to significantly improve the availability and quality of public transport. In this paper, we propose a flow-based integer programming approach to design and assess the financial viability of a SAV interurban system. An integer programming model, based on profit maximization and allowing vehicle pooling, was developed to optimize the fleet size and vehicle movements of a typical working day. This model and its optimal results were the basis for assessing the financial viability of an interurban transport network in the central area of Portugal. Different levels of demand were tested and three different fleet compositions were considered: cars, minibuses, and a mixed fleet of both. The results showed that a mixed fleet of vehicles is the best option for profit maximization. The optimal share of each vehicle type, in this fleet composition, asymptotically converges to approximately 1/3 of cars and 2/3 of minibuses, as demand increases. The introduction of a SAV system in such a region for interurban trips can replace up to 18 private vehicles per SAV vehicle, and should be done with sequential expansions to guarantee a profit at the different stages. The break-even point, for the conditions of the case study, is between 1.5 and 2 passengers per vehicle, for a fleet of cars, and between 4 and 5 passengers per vehicle, for a fleet of minibuses.