Rafael Pastor Moreno

Solving the response time variability problem by means of the electromagnetism-like mechanism

The response time variability problem (RTVP) is an NP-hard combinatorial scheduling problem that has been recently formalised in the literature. The RTVP has a wide range of real-life applications such as in the automobile industry, when models to be produced on a mixed-model assembly line have to be sequenced under a just-in-time production. The RTVP occurs whenever products, clients or jobs need to be sequenced so as to minimise variability in the time between the instants at which they receive the necessary resources. In two previous studies, three metaheuristic algorithms (a multi-start, a GRASP and a PSO algorithm) were proposed to solve the RTVP. We propose solving the RTVP by means of the electromagnetism-like mechanism (EM) metaheuristic algorithm. The EM algorithm is based on an analogy with the attraction-repulsion mechanism of the electromagnetism theory, where solutions are moved according to their associated charges. In this paper we compare the proposed EM metaheuristic procedure with the three metaheuristic algorithms aforementioned and it is shown that, on average, the EM procedure improves strongly on the obtained results.

Equilibrado de líneas de montaje con ventanas de visibilidad. Descripción del problema y procedimiento heurístico de resolución

RESUMEN: Los problemas de equilibrado de lineas (Assembly Line Balancing Problems, ALBPs) surgen al disenar o modificar una linea de montaje o de produccion y consisten, usualmente, en asignar las tareas a las estaciones de la linea de forma que se cumplan ciertos objetivos. Tradicionalmente se considera que en cada estacion hay exactamente una pieza y que la estacion tiene acceso a toda la pieza. En la practica, este supuesto puede ser demasiado restrictivo. Un ejemplo es el caso del ensamblaje automatizado de placas de circuitos impresos. En este tipo de lineas la longitud de la pieza puede ser mayor que el ancho de la estacion que la procesa. Como consecuencia, cada estacion solamente puede acceder a una porcion limitada de las piezas que esta dentro de su ventana de accesibilidad. En este trabajo se presenta la extension del ALBP al caso de estaciones con ventanas de accesibilidad (AWALBP). Se describe el problema y se propone un procedimiento de resolucion sencillo de implementar basado en i) un algoritmo para obtener una buena solucion inicial y ii) un algoritmo de recocido simulado para mejorar la solucion inicial. Finalmente se presentan los resultados computacionales que validan el procedimiento propuesto. Palabras Clave: equilibrado de lineas, ventanas de accesibilidad, heuristicas, recocido simulado

COCTEL DE HEURISTICAS PARA RESOLVER PROBLEMAS DIFICILES

In this work we introduce the concept of cocktail of heuristics. It refers to the possibility of solving a problem by all available quick heuristics instead of the practice of using only the heuristic that has the best average performance. The reasons are that the computing time of running all quick heuristics is affordable, and the best solution for a certain application of the problem can be obtained by any heuristic, even by the heuristic with the worst average performance. Thus, when a new quick heuristic is developed, it is not so important its average performance but how much the performance of the cocktail compound of the available heuristics is improved by adding this new heuristic.