Sponsor:

Medline Industries, Inc.

Research Team:

Russell D. Meller, Lisa Thomas

Universities Involved:

University of Arkansas

Start Date:

08/15/11

End Date:

12/31/12

Summary:

Many organizations use U-shaped picking modules to increase the productivity of warehouse workers. Such modules concentrate the picking area, which increases pick density, and a variety of media are used (e.g., case flow and bin locations) to differentiate between faster and slower movers. Analytical models can be useful in evaluating such picking modules (for example, is the picking module performing well or not) and evaluating the tradeoff between picking efficiency and replenishments. This project developed statistical models to compare picking modules with each other as well as general models to establish a productivity goal for Medline’s operations.
Medline Industries manufactures and distributes over 125,000 different medical products and supplies to healthcare institutions and retailers including hospitals, surgery centers, long-term care facilities and physicians’ offices. Headquartered in Mundelein, IL, Medline currently has over 35 distribution centers (DCs) throughout the U.S. and Europe. Some customers order products in less-than-case quantities, and these orders are fulfilled in a separate piece-pick module. The design of a pick module involves choosing the length of the pick module and the size of case flow and bin locations, as well as determining which product to assign to each of the picking locations. The general design approach at Medline is to place as many items as possible within the U-shaped forward area and to apply rules for allocating items within it. Because the design of the pick module significantly impacts the operating costs related to fulfilling less-than-case-quantity product demand, Medline is interested in models to assist them in determining the best design of a pick module.

Many organizations use U-shaped picking modules to increase the productivity of warehouse workers. Such modules concentrate the picking area, which increases pick density, and a variety of media are used (e.g., case flow and bin locations) to differentiate between faster and slower movers. Analytical models can be useful in evaluating such picking modules (for example, is the picking module performing well or not) and evaluating the tradeoff between picking efficiency and replenishments. This project developed statistical models to compare picking modules with each other as well as general models to establish a productivity goal for Medline’s operations. Our results provided value to Medline and they are using the Tool.
Our recommendation is for Medline to utilize the tool that we developed to design new facilities or reconfigure existing branches. The tool incorporates travel-time models for the following operations: order picking from Zone 30 locations, order picking from Zone 30 bottom-level pallet locations outside of the U-shaped module, order picking from Zone 39, replenishment to case flow locations, replenishment to endcaps and bin locations, and replenishment to Zone 30 bottom-level pallet locations. Our experience with the tool indicates that significant labor savings can be achieved (more than 20%) if Medline considers the length of the pick module in designing new pick modules, as well as alternate sub-slot size allocations. Finally, we recommend that Medline consider a smaller days-on-hand inventory level to slot items in smaller locations such that more items can be picked within the forward area, which is based on the observation that, for Medline data, the replenishment cost is small compared to the cost of picking. Further, we recommend that Medline fulfill Zone 30 pick lines in two separate batches, one for picks within the U-shaped module and one for picks from the bottom-level pallet locations outside of the U-shaped module. This will allow more lines (higher pick density) for batches within each of the respective Zone 30 areas.

The Medline operations engineer involved with this project was trained to use the tool (by analyzing an existing facility) and developed a query specifically for obtaining the input data necessary to run the tool. Further, he trained other Medline engineers in using the documentation and tool, as well as in running the query.