A variable neighborhood descent heuristic for a multi-item lot-sizing problem with remanufacturing

Abstract

Objective: to implement the neighborhood descent heuristic approach into solving the lot-sizing problem.Methods: a variable neighborhood descent heuristic approach and a framework of lot-sizing problem.Results: dynamic lot-sizing problem is supposed to be an NP-hard problem - a problem for which there is no known polynomial algorithm, so the time to find a solution grows with a problem size. In the article, we explain the basic lot-sizing, lot-sizing in closed-loop, describe methods used by other researchers, and provide reasoning for using a variable neighborhood descent heuristic method. Implementation of the method is presented in a numerical example. Application of the method significantly reduces estimation time when searching for lot-sizing problem solution. Application of the method significantly reduces estimation time when searching for lot-sizing problem solution.Scientific novelty: the method may significantly reduce time expenses during the solution of the lot-sizing problem. Practical significance: the method may be implemented for solving the lot-sizing problem in various industries, such as pulp and paper, consumer goods industry and heavy industry.

Keywords

• neighborhood descent
• lot-sizing problem
• remanufacturing
• multi-item lot-sizing
• heuristic approach
• closed-loop
• reverse logistics