DEM: A discrete electromagnetism-like mechanism for solving discrete problems

In this paper a novel model based on electromagnetism-like mechanism (EM) is proposed which is highly compatible with discrete space problems. The proposed method utilizes the EM operators to move particles towards an optimal or near optimal solutions. In fact, the proposed algorithm exploits the crossover operator to calculate forces on particles and move them according to these forces. To keep the algorithm from getting sock on local maxima a mutation approach is used. To show the performance of the proposed method it is applied on a discrete space problem called multidimensional knapsack problem (MKP). To compare our results, some standard test benches of the MKP are used and the results are compared with other methods. Experimental results showed that the proposed method outperforms some new methods in terms of the number of needed iterations to find best known solutions (about 73% in average) and CPU time (about 77%). Furthermore, the DEM is applied on some bigger MKPs and the experiments showed that the algorithm works very well and had very small error from best known optimums (0.17% error in average).