Solving the response time variability problem by means of a psychoclonal approach

Abstract

Abstract. The Response Time Variability Problem (RTVP) is a combinatorial scheduling problem which has recently appeared in the literature. This problem has a wide range of reallife applications in, for example, manufacturing, hard real-time systems, operating systems and network environment. Originally, the RTVP occurs whenever products, clients or jobs need to be sequenced in such a way that the variability in the time between the instants at which they receive the necessary resources is minimized. Since RTVP is hard to solve, heuristic techniques are needed for solving it. In a previous study, three metaheuristic algorithms (a multi-start, a GRASP and a PSO algorithm) were proposed to solve the RTVP. These three metaheuristic algorithms have been the most efficient to date in solving non-small instances of the RTVP. We propose solving the RTVP by means of a psychoclonal algorithm based approach. The psychoclonal algorithm inherits its attributes from the need hierarchy theory proposed by Maslow and the artificial immune system (AIS) approach, specifically the clonal selection principle. In this paper we compare the proposed psychoclonal algorithm with the other three metaheuristic algorithms previously mentioned and show that, on average, the psychoclonal algorithm strongly improves the obtained results.

Solving the response time variability problem by means of a genetic algorithm

“…We adapt the GA by defining a suitable representation of the solutions and genetic operators for the RTVP. The proposed GA algorithm is compared with the most efficient procedure for solving non-small instances published in the literature, which is a psychoclonal algorithm proposed in García-Villoria and Pastor (2008). On average, the proposed GA algorithm improves the best previous results reported in the literature by more than 20%.…”

“…As introduced in Section 1, the psychoclonal algorithm proposed in García-Villoria and Pastor (2008) is the best procedure to date for solving the RTVP. This algorithm is an evolutionary metaheuristic (as is the GA) that was first proposed in Tiwari et al (2005).…”

“…The basic scheme of the psychoclonal metaheuristic is the following: (1) an initial population of solutions is generated and a function is given to evaluate the fitness of a solution; (2) the best solutions are selected and cloned in a number proportional to their fitness; (3) the generated clones are hypermutated (hypermutation is similar to the mutation operator of genetic algorithms, but the difference lies in the fact that the modification rate of the hypermutation is inversely proportional to the fitness of the solution); (4) the new generation is formed by the best clones and by new solutions generated at random; (5) steps 2-4 are repeated until a stop condition is reached. This metaheuristic was adapted to solve the RTVP (for a more detailed explanation, see García-Villoria and Pastor (2008)). …”

“…The electromagnetism-like mechanism was proposed to solve the RTVP in García-Villoria and Pastor (2009b). Finally, the best results recorded to date were obtained with a psychoclonal algorithm (García-Villoria and Pastor, 2008).…”

An adaptive search for the response time variability problem

“…Seven new heuristics are also given by Corominas et al (2009). Metaheuristics for the RTVP were recently proposed in García-Villoria and Pastor (2010a, 2010b, 2010c and these include an electromagnetism-like mechanism (EM) algorithm, a psychoclonal algorithm and a genetic algorithm (GA) respectively.…”