Priorização otimizada de redes de transporte público por ônibus através de análise multicritério e algoritmo genético em Araraquara - SP

Abstract

Prioritizing public transport modes allows reducing externalities caused by the intense use of individual motorized transport modes and increasing the social welfare of the population. Exclusive bus lanes are the simplest spatial prioritization model, allowing a reduction in travel times, an increase in the operational speed of buses, a reduction in fuel consumption and a decrease in pollutant emission levels. Large urban centres are the object of most studies and projects for implementing public transport prioritisation systems. However, medium-sized cities have the fastest growing population and in 2010 they already concentrated about a quarter of the Brazilian population. These cities have the unique opportunity to design public transport systems in time to avoid that the problems observed in large urban agglomerations reach them. This paper seeks to develop a decision-making model for the design problem of public transport network by bus, more precisely the determination of priority roads for implementing bus corridors or exclusive bus lanes. The scope of the work is medium-sized Brazilian cities, with the application of a case study in Araraquara, a medium-sized city in the interior of the State of Sao Paulo, Brazil. The proposed model is composed of the combination of two methods, a multi-criteria analysis through AHP - Analytic Hierarchy Process, and an optimization method with solution through a genetic algorithm. The former simplifies the city’s road network, considering socioeconomic criteria, land use, infrastructure and operation of the existing public transport system; the latter provides a set of solutions for implementing exclusive lanes. Results show that roads selected through the AHP method correspond to traditional transport axes in the city and form a connected network, a desired characteristic of transport networks. Optimization results show a significant reduction in travel times, ranging from 9.45% or 21.88 minutes, in the smallest network resulting from the optimization (17.87km) to 23.61% or 54.66 minutes, if the most extensive network, composed of 44.64 kilometres of exclusive lanes, is considered. The combination of both methods for the decision-making process is one of the major contributions of this work, besides being one of the first applications of an optimization method with solution through genetic algorithm in a real network of a Brazilian city.