Dependable Dynamic Routing for Urban Transport Systems Through Integer Linear Programming

Abstract

Highly automated transport systems play an important role in the transformation towards a digital society, and planning the optimal routes for a set of fleet vehicles has been proved useful for improving the delivered services. Traditionally, routes are planned beforehand. However, with the advent of autonomous urban transport systems (e.g. autonomous cars), possible obstructions of tracks due to traffic congestion or bad weather conditions need to be handled on the fly. In this paper we tackle the problem of dynamically computing routes of vehicles in urban lines in the presence of potential obstructions. The problem is formulated as an integer linear optimization problem. The proposed algorithm will assign routes to vehicles dynamically, considering the track segments that are no longer available and the positions of the vehicles in the urban area. The recomputed routes guarantee the minimal waiting time for passengers. Safety of the computed routes is also guaranteed.