Greedy heuristics for determining a product family bill of materials

Radwan El Hadj Khalaf (1) , Bruno Agard (2) , Bernard Penz (1)

1 G-SCOP_ROSP - Recherche Opérationnelle pour les Systèmes de Production
2 EPM - École Polytechnique de Montréal

When designing a new product family, designers and manufacturers have to define simultaneously the product structure and its supply chain. This leads to a complex optimization problem to solve in order to satisfy diversified customers' requirements with various options and variants. The paper focuses on the first step of this design problem. It consists in selecting a set of modules that will be manufactured in distant facilities and shipped in a nearby location plant for a final assembly operation under time limits. The objective is to determine the set of modules able to define the bill of materials of each finished product in order to minimize assembly costs. We propose in this paper two heuristic strategies to solve the problem. We provide experiments on small instances which we compare with optimal solutions and we provide also experiments on big instances to compare performance of each heuristic.

Mots clés: Product family design; bill of materials; integer programming; greedy heuristic; assembly costs

Domaines: Recherche opérationnelle [math.OC]

Radwan El Hadj Khalaf, Bruno Agard, Bernard Penz. Greedy heuristics for determining a product family bill of materials. 38th International Conference on Computers & Industrial Engineering, Nov 2008, Beijing, China. ⟨hal-00337896⟩