A hybrid method for solving large-scale supply chain problems

Authors:
Steffen Wolf;Peter Merz
Affiliations:
Distributed Algorithms Group, University of Kaiserslautern, Germany;Distributed Algorithms Group, University of Kaiserslautern, Germany
Venue:
EvoCOP'07 Proceedings of the 7th European conference on Evolutionary computation in combinatorial optimization
Year:
2007

Citing 3
Cited 0

Integer and combinatorial optimization

Integer and combinatorial optimization
Approximating the two-level facility location problem via a quasi-greedy approach

SODA '04 Proceedings of the fifteenth annual ACM-SIAM symposium on Discrete algorithms
Dynamic multi-commodity capacitated facility location: a mathematical modeling framework for strategic supply chain planning

Computers and Operations Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

The strategic supply chain design problem which allows capacity shifts and budget limitations can be formulated as a linear program. Since facilities are allowed to be opened or shut down during the planning horizon, this problem is in fact a mixed integer problem. Choosing the optimal set of facilities to serve the customer demands is an NP-hard combinatorial optimization problem. We present a hybrid method combining an evolutionary algorithm and LP based solvers for solving large-scale supply chain problems, which takes its power from filtering out infeasible solutions. The EA incorporating these filters is shown to be faster than the MIP solver ILOG CPLEX in most of the considered instances. For the remaining instances it finds feasible solutions much faster than the MIP solver.