A Self-Adaptive Heuristic Algorithm for Combinatorial Optimization Problems

Abstract

This paper introduces a new self-tuning mechanism to the local search heuristic for solving of combinatorial optimization problems. Parameter tuning of heuristics makes them difficult to apply, as parameter tuning itself is an optimization problem. For this purpose, a modified local search algorithm free from parameter tuning, called SelfAdaptive Local Search (SALS), is proposed for obtaining qualified solutions to combinatorial problems within reasonable amount of computer times. SALS is applied to several combinatorial optimization problems, namely, classical vehicle routing, permutation flow-shop scheduling, quadratic assignment, and topological design of networks. It is observed that self-adaptive structure of SALS provides implementation simplicity and flexibility to the considered combinatorial optimization problems. Detailed computational studies confirm the performance of SALS on the suit of test problems for each considered problem type especially in terms of solution quality.

Modified self-adaptive local search algorithm for a biobjective permutation flow shop scheduling problem

“…If the algorithm is allowed to run enough, Θ approaches 1 and forces the search process to find better solutions. A detailed analysis of Θ can be found in [28].…”

“…The same moving mechanisms are also utilized by SALS when the permutation representation is adapted [28,30,42]. MSALS selects one of the moving types randomly and then generates a random neighbor using the selected moving mechanism at each iteration, allowing repetitions of the neighbors.…”

“…On the other hand, these algorithms still include some parameters that need to be initialized. The work in [28] proposed a simple self-adaptive local search (SALS) and applied it to several single-objective problems including quadratic assignment, PFSS, classical vehicle routing, and topological optimization of backbone networks. The work in [29] also represented a detailed implementation of SALS for topological optimization problem.…”

“…Trial solutions obtained during the search process are accepted or rejected depending on the current value of the acceptance parameter. It was shown that the tuning of the single parameter of SALS is independent of the problem type or its instances [28].…”

Self-Tune Linear Adaptive-Genetic Algorithm for Feature Selection

“…In recent years, appropriate self-tuning procedures has been implemented into other machine learning techniques. For instance, a unique Self-Adaptive Local Search (SALS) has been introduced for multi-objective optimisation purposes [23]. The SALS performs an automatic heuristic variable learning process to eliminate complicated parameter fine-tuning and to deliver performance enhancement.…”