Optimization of tardiness related objectives in single machine environments via multirecombined evolutionary algorithms

Tardiness related objectives are of utmost importance in production systems when client satisfaction is a main goal of a company. These objectives measure the system response to the client requirements and rate manager´s performance In scheduling problems with diverse single or multiple objectives and environments Evolutionary algorithms (EAs) were successfully applied. Latest improvements in EAs have been developed by means of multirecombination, a method, which allows multiple exchange of genetic material between individuals of the mating pool. These individuals can be provided by the current population or by an external source. The performance of the algorithm depends o the number of individuals in the mating pool and their mating frequency. MCMP-SRI and MCMP-SRSI are multirecombined evolutionary approaches using the concept of the stud (a breeding individual), random immigrants and/or seeds, to avoid premature convergence and adding problem-specific- knowledge. Here, both methods applied to tardiness related problems in single machine environmen are discussed and contrasted against conventional heuristics. Eje: Inteligencia artificial Red de Universidades con Carreras en Informática (RedUNCI)