The goal of this research is to develop a systematic and science-based methodology for rapid design of warehousing systems.  Currently, design practice is based either on ad-hoc expertise and experience of design engineers or on detailed simulation models of equipment and material flow through the warehouse.  However, the current business climate does not allow for the long amount of time needed to develop such simulation models.  Third-party logistics providers are routinely faced with two-week deadlines to respond to a request-for-proposals with a bid that includes a design and price over a horizon of several years.  While expert practitioners have their place, they are a scarce resource, and design results from different experts can be radically different.

There is an urgent need for a rigorous, science-based methodology for design that requires less data and modeling time than simulation approaches.  A software implementation of such a methodology is called a rapid prototyping tool.  In this research, the warehouse is viewed as having to satisfy a number of high-level objectives and constraints.  Many of the constraints are elastic, i.e., they can be violated with a penalty.  We believe that the design problem can be formulated as a multi-period, multi-commodity, capacitated network flow problem whose capacities are determined by binary configuration variables.  There are costs associated with the continuous flow and storage variables, as well as the binary configuration variables.  As such, this problem is a large-scale mixed-integer linear programming problem that is difficult to solve.  Hence, our focus also is on developing efficient heuristic solution procedures.  Initial work has focused on specifying the overall problem structure, and on developing specific formulations and procedures to support design of small-parts storage systems.
 

Contact: marc.goetschalckx@isye.gatech.edu