Discrete Particle Swarm Optimization with a Search Decomposition and Random Selection for the Shortest Path Problem

This paper proposes a discrete particle swarm optimization (DPSO) for the solution of the shortest path problem (SPP). The proposed DPSO termed as DPSO_SPP adopts a new solution representation which incorporates a search decomposition procedure and random selection of priority value. The purpose of this representation is to reduce the searching space of the particles, leading to a better solution. Detailed descriptions of the new solution and the DPSO_SPP algorithm are elaborated. Computational experiments involve an SPP dataset from previous research and road network datasets. The DPSO_SPP is compared with a genetic algorithm (GA) using naive and new solution representation. The results indicate that the proposed DPSO_SPP is highly competitive and shows good performance in both frequency of obtaining an optimal solution and rate of convergence in comparison with the GA_SPP, PSO, and GA. In particular, DPSO_SPP with the use of inertia weight had shown better solution to SPP compared to constriction coefficient (CF). The quality of the solution achieved through DPSO_SPP for all datasets indicated higher potential in achieving the optimum results for SPP, serving as a good ground to further test the algorithm on larger datasets.