Analytical and Comparative Study Between Neural Network Approach, and Metaheuristics for Solving Task Assignment Problem

Metaheuristics and neural network approaches have demonstrated their effectiveness in solving a variety of combinatorial optimization problems (COPs). The Task Assignment Problem (TAP), especially with non-uniform communication costs, is a well-known NP-Hard COP. Exact methods for solving TAP require exponential time, making them impractical for large-scale instances, Therefore, alternative strategies, such as metaheuristics and neural network approaches, are necessary. This study compares and analyses metaheuristic algorithms: Genetic Algorithm (GA), Ant Colony Optimization (ACO), Simulated Annealing (SA), and neural networks approach: Continuous Hopfield Networks (CHNs), on solving the TAP. Experimental results show that GA outperforms the other approaches in terms of effectiveness. Furthermore, SA and CHN achieved acceptable solutions within minimal execution time compared to ACO and GA.