Improving Initial Population for Genetic Algorithm using the Multi Linear Regression Based Technique (MLRBT)

Genetic algorithms (GAs) are powerful heuristic search techniques that used successfully to solve problems for many different applications. Seeding the initial population is considered as first step of GAs. In this work, a new method proposed, seeding called Multi Linear Regression Based Technique (MLRBT). That divides given large scale TSP problem into smaller sub-problems and technique works frequently until sub-problem size very small, four cities or less. Experiments were carried out using well-known Travelling Salesman Problem (TSP) instances they showed promising results in improving GAs' performance TSP.