AN EVOLUTIONARY APPROACH WITH MULTIPLICITY, INCEST PREVENTION AD DECODERS TO SOLVE THE JOB SHOP PROBLEM

摘要

In the field of scheduling, the job shop scheduling problem (JSSP) is very complex in nature and quite hard to solve by conventional optimization techniques. During the last decades, there has been an increasing interest in evolutionary algorithms (EAs) to solve such kind of problems. Even if born to simulate evolution and natural adaptive systems behaviour, EAs have received considerable attention regarding their potential as an optimization technique. In a previous work we proposed evolutionary algorithms using a multiplicity feature to solve the JSSP. The algorithms applied multiple crossovers on multiple parents (MCMP) and its further combination with extended incest prevention (MCMPIP). The setting of the population size for both algorithms was different, because the EA with MCMPIP needed a bigger population to find individuals without common ancestors, for matting. In this work both algorithms are revisited to compare their performance for equal population sizes. Also the effect of increasing the population size for an algorithm with MCMP is analyzed.