A hybrid evolutionary approach to the management of Bicycle Renting Systems

摘要

Urban Bicycles Renting Systems (UBRS) require the planification of minimal-cost routes for several vehicles used in servicing a set of bicycle base stations distributed across a city. We address the UBRS management problem modeled as a vehicle routing type problem using a hybrid evolutionary approach. Individuals encode a set of routes satisfying the constraints related to vehicle capacity and node (i.e. bicycle base station) demands. The main features of the proposed evolutionary algorithm refer to an asynchronous insertion of new individuals (generated via specific search operators) into the population and a local search stage relying on 2-opt heuristic. This hybrid evolutionary algorithm is engaged for the UBRS problem using data from the cities of Barcelona and Valencia with promising results.