主题:Scheduling Applications in Industry - Steelmaking

标题：钢铁工业中的调度研究

时间: 2026年1月5日早上9:30

地点:管理科研楼第二教室

主讲人: Michael Pinedo 纽约大学Stern商学院

Bio:Michael Pinedo is the Julius Schlesinger Professor of Operations Management at New York University's Stern School of Business.  He received an Ir. degree in Mechanical Engineering from Delft University of Technology in 1973 and a Ph.D. in Operations Research from the University of California at Berkeley in 1978. His research focuses on the modeling of production and service systems, and in particular planning and scheduling systems. He is author of Scheduling: Theory, Algorithms and Systems (Springer), and Planning and Scheduling in Manufacturing and Services (Springer), and coauthor of Queueing Networks: Customers, Signals and Product Form Solutions (Wiley).  Recently, his research has also focused on operational risk in financial services.  He is co-editor of Creating Value in Financial Services: Strategies, Operations, and Technologies (Kluwer), and co-editor of Global Asset Management – Strategies, Risks, Processes and Technologies (Palgrave/McMillan). Professor Pinedo has been actively involved in industrial systems development. He supervised the development and implementation of several scheduling systems for International Paper and participated in the development of systems at Goldman Sachs, Philips, Siemens, and Merck. Professor Pinedo is Department Editor of Naval Research Logistics, Department Editor of Production and Operations Management and Associate Editor of Annals of Operations Research. 

Abstract: Efficient scheduling of industrial systems typically has a major impact on  productivity levels.   In this presentation we focus on some scheduling applications in steelmaking with continuous casting. In steel production the steelmaking-continuous casting (SCC) process is typically a bottleneck. Its scheduling has become more challenging over the years. We first describe the modeling of the essential features of an SCC process, such as unrelated parallel machine environments, stage skipping, and maximum waiting time limits in between successive stages.The objective is the minimization of a weighted combination of the total weighted waiting time, total earliness, and total tardiness.The problem can be formulated as a mixed integer program. We first present an iterated greedy matheuristic in which the subproblems are solved to optimality using mathematical programming in order to find an overall near-optimal solution.We then describe the performance of a similar heuristic in which the mathematical programming procedures for the subproblems are replaced by constraint programming procedures.  Through numerical experiments, we compare the effectiveness of both types of heuristics with the performance of a multi-objective geneticalgorithm.We conclude our presentation with some other scheduling applications in industry that deserve research attention.