Operations Research (OR) aims to apply information technology and mathematics for informed decision making. Analytical and numerical techniques are used to develop and manipulate mathematical and computer models for prediction and improvement of system performance. These systems may be composed of people, machines or procedures. OR tools have been applied in a diverse set of areas including defense, health care, design of communications systems, computer operation, design and networking, transportation, finance, inventory planning, and manufacturing.

Operations Research is practiced throughout IBM. Research in OR, as well as the application of tools in OR, has resulted in significant dollar savings, improved operations and more competitive strategies, for both IBM and IBM's customers.

Our research focuses on both the development and the application of OR tools. Much of the effort in the area of algorithm development concentrates on optimization, queuing theory, and agent-based systems. Recent application areas include supply chain modeling and optimization, manufacturing planning and scheduling, service industry resource planning and scheduling, airline optimization, and forecasting. New initiatives include modeling and software environments for collaborative planning and electronic commerce.

Recently, the Asset Management Tool (AMT) was developed. AMT is an advanced supply chain optimization and simulation tool that helps identify optimal inventory policies, by considering logistics, customer service levels targets, supply network configuration, channel assembly, parts commonality, fulfillment policies, lead time reduction, and forecast accuracy. AMT uses simulation, animation, graphical process modeling, analytical performance optimization, and activity-based costing to provide quantitative analysis of extended supply chains. It aids in evaluating supply chains on the basis of financial tradeoffs associated with various configurations and operational policies. Both the core OR tools incorporated in AMT and its production level software, which is currently being used worldwide, were developed by IBM's research laboratories. AMT has been used within IBM business units as well as in external reseller organizations. In 1999, INFORMS, the international OR professional society, awarded the AMT project with the prestigious Franz Edelman Award for outstanding application of OR methodology.

The IBM Journey Management Library was developed as part of our focus on using OR to develop industry solutions. It provides a set of reusable modules and templates that work together with a simulation tool to describe passenger processes and related new technologies. It helps airlines to improve the travel experience by finding ways of reducing waiting time, eliminating system bottlenecks, and improving customer service. We teamed with Air Canada to develop simulation models of ticketing, check-in, and baggage handling. These models were used to preview the effect of new technologies on customer service, and to identify potential improvements on system performance and customer wait times.

In addition to developing applications, we also engage in research projects that seek to advance the understanding of core OR methodology. For example, we are developing tools that are incorporated in advanced mathematical software products. Work in this area focuses on three main topics. The first is high-performance computational kernels on various serial and parallel hardware platforms, which consist of basic scientific algorithmic building blocks, such as matrix multiplication or fast Fourier transform. The second activity involves advanced optimization software for solving linear, integer and nonlinear programming problems. The third area is advanced mathematical software involving sparse linear systems, which has resulted in Watson Symmetric Sparse Matrix Package (WSSMP).

Over the years, our work on OR methodology and software systems has resulted in significant advances in OR techniques, and has yielded substantial financial gains for IBM and its customers.

Please contact Paridhi Verma to obtain copies of the Computer Science Brochure