The purpose of this paper is the application of Genetic Algorithms to solve the Signal Setting Design at a single junction. Two methods are compared: the monocriteria and the multicriteria optimisations. In the former case, three different objectives functions were considered: the capacity factor maximisation, the total delay minimisation and the total number of stops minimisation; in the latter case, two combinations of criteria were investigated: the total delay minimisation and the capacity factor maximisa-tion, the total delay minimisation and the total number of stops minimisation. Furthermore, two multicriteria genetic algorithms were compared: the Goldberg’s Pareto Ranking (GPR) and the Non Dominated Sorting Genetic Algorithms (NSGA-II). Conclusions discuss the effectiveness of multicrite-ria optimisation with respect to monocriteria optimisation, and the effec-tiveness of NSGA-II with respect to the GPR.
Signal Setting Design at a Single Junction through the Application of Genetic Algorithms
CANTARELLA, Giulio Erberto;DE LUCA, STEFANO;DI PACE, ROBERTA;MEMOLI, SILVIO
2014-01-01
Abstract
The purpose of this paper is the application of Genetic Algorithms to solve the Signal Setting Design at a single junction. Two methods are compared: the monocriteria and the multicriteria optimisations. In the former case, three different objectives functions were considered: the capacity factor maximisation, the total delay minimisation and the total number of stops minimisation; in the latter case, two combinations of criteria were investigated: the total delay minimisation and the capacity factor maximisa-tion, the total delay minimisation and the total number of stops minimisation. Furthermore, two multicriteria genetic algorithms were compared: the Goldberg’s Pareto Ranking (GPR) and the Non Dominated Sorting Genetic Algorithms (NSGA-II). Conclusions discuss the effectiveness of multicrite-ria optimisation with respect to monocriteria optimisation, and the effec-tiveness of NSGA-II with respect to the GPR.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
