An Enumeration-Based Deterministic Algorithm for Solving optimization Problem
Changlin Yang · 2023
An enumeration-based deterministic algorithm is proposed for solving optimization problem. Discretization of the feasible region for each decision variable is applied to establish search tree so that all the solutions could be enumerated. A new search region is established from the optimization of previous literation. Meanwhile, adjustment of search step with exponential decay function is conducive to reducing the search scope until the convergence of result is realized. The proposed algorithm is applied for three kinds of typical optimization problem, including Non-linear Programing Problem, Job-Shop Scheduling Problem and Traveling Salesman Problem. The results obtained by the new algorithm are compared with results of swarm intelligent optimization algorithm. Results indicate that for the combinational optimization, high convergence rate could be realized by adjusting the feasible region for each decision variable. For non-linear programming problem, the accuracy of results obtained from deterministic algorithm is better than that from self-organizing migrating algorithm, which infers that enumeration-based deterministic algorithm could be used to solve NP hard problem.