Automatic Tuning of GRASP with Path-Relinking Heuristics with a Biased Random-Key Genetic Algorithm
نویسندگان
چکیده
GRASP with path-relinking (GRASP+PR) is a metaheuristic for finding optimal or near-optimal solutions of combinatorial optimization problems. This paper proposes a new automatic parameter tuning procedure for GRASP+PR heuristics based on a biased random-key genetic algorithm (BRKGA). Given a GRASP+PR heuristic with n input parameters, the tuning procedure makes use of a BRKGA in a first phase to explore the parameter space and set the parameters with which the GRASP+PR heuristic will run in a second phase. The procedure is illustrated with a GRASP+PR for the generalized quadratic assignment problem with n = 30 parameters. Computational results show that the resulting hybrid heuristic is robust.
منابع مشابه
Automatic Tuning of GRASP with Evolutionary Path-Relinking
Heuristics for combinatorial optimization are often controlled by discrete and continuous parameters that define its behavior. The number of possible configurations of the heuristic can be large, resulting in a difficult analysis. Manual tuning can be time-consuming, and usually considers a very limited number of configurations. An alternative to manual tuning is automatic tuning. In this paper...
متن کاملRandomized heuristics for the family traveling salesperson problem
This paper introduces the family traveling salesperson problem (FTSP), a variant of the generalized traveling salesman problem. In the FTSP, a subset of nodes must be visited for each node cluster in the graph. The objective is to minimize the distance traveled. We describe an integer programming formulation for the FTSP and show that the commercial-grade integer programming solver CPLEX 11 can...
متن کاملGrasp with Path Relinking for the Three-index Assignment Problem
Abstract. This paper describes variants of GRASP (greedy randomized adaptive search procedure) with path relinking for the three index assignment problem (AP3). GRASP is a multi-start metaheuristic for combinatorial optimization. It usually consists of a construction procedure based on a greedy randomized algorithm and of a local search. Path relinking is an intensification strategy that explor...
متن کاملGRASP with Path Relinking for Three-Index Assignment
T paper proposes and tests variants of GRASP (greedy randomized adaptive search procedure) with path relinking for the three-index assignment problem (AP3). GRASP is a multistart metaheuristic for combinatorial optimization. It usually consists of a construction procedure based on a greedy randomized algorithm and of a local search. Path relinking is an intensification strategy that explores tr...
متن کاملRestart strategies for GRASP with path-relinking heuristics
GRASP with path-relinking is a hybrid metaheuristic, or stochastic local search (Monte Carlo) method, for combinatorial optimization. A restart strategy in GRASP with path-relinking heuristics is a set of iterations {i1, i2, . . .} on which the heuristic is restarted from scratch using a new seed for the random number generator. Restart strategies have been shown to speed up stochastic local se...
متن کامل