CPU task scheduling using genetic algorithm

This paper addresses p-processes single processor scheduling problem with a common deadline, to minimize the total execution time and reduce the penalty costs. Process scheduling is one of the most essential factor on which the efficiency and the performance of the work done by the CPU depends. Earliness and tardiness of the processes degrades the efficiency of the processor as they carry penalty costs with them. Thus, the scheduling problem of minimizing the total sum of earliness and tardiness with a common deadline on a single processor is important and competitive. Scheduling is particularly one of the subset of combinatorial optimization problems, which are in fact NP-Hard problems. This problem can be solved using heuristic and meta-heuristic approach such as genetic algorithm for optimal results. In the experiment performed, the Genetic Algorithm outperforms the results in comparison with a simple heuristic method.