Service Parts Logistics

Friday, September 13, 2002 - 9:30am - 10:15am
Keller 3-180
Jon Lee (IBM)
Large spare-parts inventory and distribution logistics operations involve over 10K part numbers, with annual volume in the several M. In recent years, market forces have changed the way these services are sold and delivered. These changes are being made to address the needs to improve customer satisfaction and to drive down costs associated with service delivery. For example, the IBM/ITS North American maintenance organization focused on specific areas in order for IBM to drive down costs and improve the customer experience. Specifically, (i) a value-based pricing scheme tied to service levels, and (ii) a call screening process enabling remote diagnosis of problems. These changes led to challenging logistics requirements involving the deployment of labor and parts in order to satisfy contracted service levels while containing inventory and transportation costs.

The Optimization Center of the Mathematical Sciences Department at IBM Research and IBM's ITS/SPS organization, in collaboration with the Lehigh University Department of Industrial and Manufacturing Engineering have been developing a next generation logistics system designed for flexible and optimal control of spare parts inventories. IBM has migrated to a Parts Procurement Time (PPT) performance measure, which monitors whether the frequency at which parts are delivered to a customer location within a contractually determined time interval aggregated across machine service groups and geographies meet acceptable thresholds. I will describe the optimization model and algorithms that are currently being successfully deployed which have led to increased service, and reduced transportation costs, and dramatically reduced inventory costs.