A thermodynamical selection rule for the genetic algorithm

The genetic algorithm (GA), an optimization technique based on the process of evolution, suffers from a phenomenon called premature convergence. That is, the system often loses diversity of the population at an early stage of searching. The authors propose a novel method called the Thermodynamical Genetic Algorithm (TDGA), which adopts the concepts of the temperature and the entropy in the selection rule, getting a hint from the method of simulated annealing (SA) to maintain diversity of the population. Comparison of the TDGA with the Simple GA is carried out taking a knapsack problem as an example