Age-Layer-Population-Structure with Self-adaptation in Optimization

Dynamic optimization problems are a class of optimization problems where the objective function, constraints, or both can change over time. In this work, we address the dynamic travelling salesman problem (TSP) by using the age-layered population structure (ALPS). To enhance different layer’s behaviour, we introduce the self-adaptation strategies to adjust the mutation and crossover rate in each layers, which are stationary in the conventional ALPS. The proposed strategies are compared with the stationary strategy over 7 different dynamic TSPs. The experimental results shows that the proposed strategy, convex strategy, yields much better results than the stationary strategy on complex problems.