Seminars

ME/IE 8773-8774
Industrial Engineering Division Seminar
Operations Research
Host: William L. Cooper

Solving MRP Production Planning Problems by Mixed Integer Programming
by

Andrew J. Miller
Assistant Professor
Department of Industrial Engineering
University of Wisconsin-Madison

Wednesday, February 4, 2004
3:30-5:00 p.m.
Room 1130 ME
Coffee and cookies will be available at 3:15 p.m. before the seminar

Abstract: Many manufacturing companies encounter production planning problems which can be broadly classified as multi-level lot-sizing problems. The ability to solve these problems effectively could result in substantial improvements in lead times, on-time order fulfillment, and other efficiency measures. Because of the size and difficulty of these problems, however, the most prevalent methods for solving them remain rough approaches such as Material Requirements Planning (MRP) or Manufacturing Resources Planning (MRP II). These methods simplify the relationships between capacity, setup times, inventory, and demand to such an extent that the resulting production plans are often of poor quality or even not feasible. Our talk will focus on mixed integer programming (MIP) approaches that have led to recent progress in solving multi-level lot-sizing problems the kinds of problems that MRP and MRP II systems try to solve. We will discuss recent theoretical results concerning the polyhedral structure of simple production planning models, as well as computational progress in using valid inequalities for simple submodels to solve more complicated production planning problems efficiently. We will also discuss some significant challenges that remain in solving these problems. This talk discusses research carried out with Kerem Akartunali, George L. Nemhauser, Yves Pochet, Martin W.P. Savelsbergh, and Laurence A. Wolsey, among others. We will discuss recent and ongoing research by other authors as well.

Bio: Andrew J. Miller joined the UW-Madison Department of Industrial Engineering in the fall of 2001. His research focuses on integer programming, mixed integer programming, and combinatorial optimization, and on the application of methods in these areas to solve problems that arise in logistics, production planning, and supply chain design. He has published several journal articles and has worked on research projects involving a number of manufacturing and software companies, including Philips Electronics, Procter and Gamble, BASF, and DASH Optimization. Prior to his arrival in Madison, Dr. Miller spent two years as a visiting research fellow at the Center of Operational Research and Econometrics (CORE) in Louvain-la-Neuve, Belgium. He received his Ph.D. from Georgia Institute of Technology in 1999. Outside of academia, he has worked for Philips Electronics North America and InterTrans Logistics Solutions (since acquired by i2 Technologies).

Informal Faculty Luncheon: Wednesday, February 4, 2004, 12:00 noon. Meet in 1100 ME and walk to lunch with other faculty. Prof. Andrew Miller will be able to attend.