A hybrid search combining interior point methods and metaheuristics for 0–1 programming

Abstract

Our search deals with methods hybridizing interior point processes and metaheuristics for solving 0-1 linear programs. This paper shows how metaheuristics can take advantage of a sequence of interior points generated by an interior point method. After introducing our work field, we present our hybrid search which generates a diversified population. Next, we explain the whole method combining the solutions encountered in the previous phase through a path relinking template. Computational experiments are reported on 0-1 multiconstraint knapsack problems.

Combining Metaheuristics and Exact Algorithms in Combinatorial Optimization: A Survey and Classification

“…Plateau et al [31] combine interior point methods and metaheuristics for solving the multiconstrained knapsack problem. The first part is an interior point method with early termination.…”

A novel application of crossover operator to a hybrid optimization framework: Investigation into cutting problems

“…Interior Point Algorithms and metaheuristics are combined to solve the one-dimensional Knapsack Problem in [3]. Firstly, the interior point method with early termination is executed to obtain an initial solution.…”

Optimised Search Heuristics: A Mapping of Procedures and Combinatorial Optimisation Problems

“…A mixed integer program is solved to find the remaining ones. (Plateau et al 2002) This procedure is applied to the multiconstrained knapsack problem. It starts by executing an interior point method with early termination.…”