Gabriel Fagúndez

This article presents the application of the NSGA-II multiobjective evolutionary algorithm to the problem of distributing passengers traveling from the same origin to different destinations in several taxis. A new problem formulation is presented, accounting for two quality of service metrics from the point of view of the users: minimize the total cost of the trips and the time of travel for each passenger from the origin to its destination. The proposed method follows a fully multiobjective… Show more

This article presents the application of the NSGA-II multiobjective evolutionary algorithm to the problem of distributing passengers traveling from the same origin to different destinations in several taxis. A new problem formulation is presented, accounting for two quality of service metrics from the point of view of the users: minimize the total cost of the trips and the time of travel for each passenger from the origin to its destination. The proposed method follows a fully multiobjective approach, and it is designed to provide an accurate and efficient way to solve realistic instances of the problem with high practical applicability. The experimental analysis compares the solutions found using the proposed algorithm versus those computed using a previous parallel micro evolutionary algorithm (following a linear aggregation approach to combine the problem objectives) and two greedy heuristics. The results show that the multiobjective evolutionary algorithm is able to efficiently reach significant improvements in both problem objectives in short execution times. Show less

Este trabajo presenta la aplicación de un micro algoritmo evolutivo paralelo para resolver una variante multiobjetivo del problema de planificación de viajes compartidos en taxis. Los objetivos considerados en la variante abordada del problema son el costo total del viaje y la demora en llegar a destino por parte de cada pasajero. Esta versión del problema de viajes compartidos en taxis contempla situaciones realistas con alta aplicabilidad en la práctica, y toma en cuenta los criterios más… Show more

Este trabajo presenta la aplicación de un micro algoritmo evolutivo paralelo para resolver una variante multiobjetivo del problema de planificación de viajes compartidos en taxis. Los objetivos considerados en la variante abordada del problema son el costo total del viaje y la demora en llegar a destino por parte de cada pasajero. Esta versión del problema de viajes compartidos en taxis contempla situaciones realistas con alta aplicabilidad en la práctica, y toma en cuenta los criterios más considerados por los usuarios. El problema se resuelve con un micro algoritmo evolutivo paralelo que utiliza el modelo de subpoblaciones distribuidas y una estrategia de descomposición de dominio para contemplar las diferentes ponderaciones de los objetivos. El análisis experimental realizado sobre un conjunto de instancias reales del problema muestra que el algoritmo propuesto es ca- paz de encontrar significativas mejoras en ambos objetivos cuando se lo compara con estrategias ávidas intuitivas para resolver el problema. Show less

This article presents the application of evolutionary algorithms for solving the problem of distributing a group of passengers travelling from the same origin to different destinations in several taxis, with the goal of minimizing the total cost of the trips. The experimental analysis compares the quality of the solutions found using the proposed algorithm versus those computed using an intuitive greedy heuristic similar to those previously proposed in the related literature to solve the… Show more

This article presents the application of evolutionary algorithms for solving the problem of distributing a group of passengers travelling from the same origin to different destinations in several taxis, with the goal of minimizing the total cost of the trips. The experimental analysis compares the quality of the solutions found using the proposed algorithm versus those computed using an intuitive greedy heuristic similar to those previously proposed in the related literature to solve the problem, showing that the evolutionary algorithm is able to reach significant improvements in the trips' total cost, outperforming the greedy heuristic in up to 41.7 % in the best case, and up to 36.4% on average. Two applications are also presented: a web-based user interface and a mobile application for devices using the iOS operating system. Show less

This article presents the application of a parallel micro evolutionary algorithm to the problem of distributing passengers traveling from the same origin to different destinations in several taxis, with the goal of minimizing the total cost of the trips. The proposed method is designed to provide an accurate and efficient way to solve the problem. The experimental analysis compares the solutions found using the proposed algorithm versus those computed using a sequential evolutionary algorithm… Show more

This article presents the application of a parallel micro evolutionary algorithm to the problem of distributing passengers traveling from the same origin to different destinations in several taxis, with the goal of minimizing the total cost of the trips. The proposed method is designed to provide an accurate and efficient way to solve the problem. The experimental analysis compares the solutions found using the proposed algorithm versus those computed using a sequential evolutionary algorithm and an intuitive greedy heuristic. The results show that the parallel evolutionary algorithm is able to efficiently reach significant im- provements in the total cost, outperforming the greedy heuristic in up to 36.1% (18.2% on average), and the sequential evolutionary algorithm in up to 8.5% (4.3% on average). Show less