Optimizer: IBM's Multi-Echelon Inventory System for Managing Service Logistics
Abstract
IBM recently implemented Optimizer, a system for flexible and optimal control of service levels and spare parts inventory, in its US network for service support. It is based upon recent research in multi-echelon inventory theory to address the IBM network. The inherent complexity and very large scale of the basic problem required IBM to develop suitable algorithms and sophisticated data structures and required large-scale systems integration. Optimizer has greatly improved IBM's US service business. The implementation of Optimizer has made it possible to make strategic changes to the configuration and control of the parts distribution network. It resulted in simultaneously reducing inventory investment and operating costs and improving service levels. Most important, however, Optimizer has proven to be a highly flexible planning and operational control system.

