A DIMMA-Based Memetic Algorithm for 0-1 Multidimensional Knapsack Problem Using DOE Approach for Parameter Tuning
Multidimensional 0-1 Knapsack Problem (MKP) is a well-known integer programming problems. The objective of MKP is to find a subset of items with maximum value satisfying the capacity constraints. A Memetic algorithm on the basis of Design and Implementation Methodology for Metaheuristic Algorithms (DIMMA) is proposed to solve MKP. DIMMA is a new methodology to develop a metaheuristic algorithm. The Memetic algorithm is categorized as metaheuristics and is a particular class of evolutionary algorithms. The parameters of the proposed algorithm are tuned by Design of Experiments (DOE) approach. DOE refers to the process of planning the experiment so that appropriate data that can be analyzed by statistical methods will be collected, resulting in valid and objective conclusions. The proposed algorithm is tested on several MKP standard instances from OR-Library. The results show the efficiency and effectiveness of the proposed algorithm.