A two-stage mathematical-programming method for the multi-floor facility layout problem

Abstract

We present a two-stage method using mathematical-programming techniques for finding high-quality solutions to the multi-floor facility layout problem. The first stage is a mixed-integer linear program that assigns departments to floors such that the total of the vertical interaction costs between departments on different floors is globally minimized. The second stage finds a locally optimal layout for each floor. Two versions of the proposed approach are considered. The first solves the layout of each floor independently of the other floors, and is suitable for up to one elevator location. The second solves the layout of all floors simultaneously and can handle multiple elevator locations. Preliminary computational results show that both versions of the proposed method can efficiently provide a good variety of high-quality solutions in a short amount of time for medium and large-scale problem instances.

Keywords:

• facilities planning and design
• facility layout
• convex programming
• global optimization

^†Research partially supported by an Ontario Graduate Scholarship.

^††Research partially supported by the Natural Sciences and Engineering Research Council of Canada, and by a Humboldt Research Fellowship.

^†Research partially supported by an Ontario Graduate Scholarship.

^††Research partially supported by the Natural Sciences and Engineering Research Council of Canada, and by a Humboldt Research Fellowship.

Acknowledgements

We thank two anonymous referees for their constructive comments that helped improve the paper.