A project for Manhattan Associates

These materials are provided by the Supply Chain & Logistics Institute at the Georgia Institute of Technology. You are welcome to use them so long as the copyrights remain intact, credit for authorship is acknowledged, and nothing is resold at profit.

The company

Manhattan Associates is one of the leading vendors of software to manage warehouses. In addition, they are moving aggressively to provide tools to manage outside the warehouse, indeed all along the supply chain.

The project

Suppose your distribution center (DC) distributes, among other things, flashlights and batteries. If almost everyone who orders a flashlight also orders batteries then you may be able to reduce travel by storing flashlights and batteries near each other.

More generally, there is a "setup" cost in picking a sku: The order-picker must travel to the location where the sku resides. This setup cost is typically sequence dependent because its value depends on the previously visited location. If skus that are frequently ordered together are also stored together then travel time may be reduced. In effect, the physical layout of the warehouse is tuned to match the pattern of customer orders.

Slotting Optimization® is a Manhattan Associates tool that produces slotting recommendations to increase picking efficiencies and drive down order selection costs. To use the software, the customer must supply information about the physical layout of their warehouse (racking, slot dimensions, etc.), the items that are stored in the facility (dimensions, grouping characteristics, etc.), and their slotting objectives (where certain item groups can and cannot be slotted, whether "golden zone" slotting or pickline balancing are to be used, etc.). An electronic representation of the customer's warehouse will be generated using this information. Once everything is properly configured, the system can be used to control the slotting of items in the facility, placing new items in the appropriate places, and finding better locations for items that are poorly slotted.

Slotting Optimization® currently can group items together, but the user is required to manually identify the characteristics the items share that makes them members of a group. Typically, items are grouped based upon product type, the vendor who supplied the item, or whether or not the items have a special handling characteristic (e.g., flammability). Slotting by affinity would be an automatic process. Slotting Optimization® would analyze actual customer order data, first attempting to identify items that need to be kept together and then finding neighboring picking locations for these items.

The goal of this project is to determine whether or not slotting by affinity is beneficial to the customer, and can be incorporated as new functionality to Slotting Optimization®. Assuming it is worthwhile, Manhattan Associates would also need to know how to decipher order data and determine which items are best slotted together.


The company data is copyrighted and proprietary. You may use it for the purposes of this course only. (If you would like to use it for something else, please contact me to discuss.)

We will be working with data from the following facilities.

Suggestions for getting started