A Variables Clustering Based Differential Evolution Algorithm to Solve Production Planning Problem

The scenario tree based multistage stochastic programming model is popular to describe production planning problem. However, as its high-dimensional variables and large-scale solution space, the addressed model is hardly to be solved in an acceptable time. To deal with this challenge, we propose a variables clustering based differential evolution algorithm which combines two novel strategies i.e. the children cluster based parallel evolution operations and the entirely randomized parameters for each child-individual. A case of weapons production planning is studied to validate the proposed algorithm. The results show that this algorithm has the fastest convergence and the best global searching capability in 6 test instances with different scales of the variables and the solution space, compared with classical differential evolution algorithm, genetic algorithm and particle swarm optimization algorithm. (Received, processed and accepted by the Chinese Representative Office.)