A parallel implementation of a multi-objective evolutionary algorithm

Abstract

The use of Evolutionary Algorithms (EAs) in difficult problems, where the search space is unknown, urges researches to find ways to exploit their parallel aspect. Multi-objective Evolutionary Algorithms (MOEAs) have features that can be exploited to harness the processing power offered by modern multi-core CPUs. Modern programming languages offer the ability to use threads and processes in order to achieve parallelism that is inherent in multi-core CPUs. This thesis presents a parallel implementation of a MOEA algorithm and its application to the de novo drug design problem. Drug discovery and De novo Drug design is a complex task that has to satisfy a number of conflicting objectives, where a MOEA finds a suitable problem to be used on. Further more such a task needs high amount of execution time. The aim is to minimize this time by the use of a parallel MOEA. The results indicate that using multiple processes that execute independent tasks of a MOEA can reduce significantly the execution time required and maintain comparable solution quality thereby achieving improved performance.