Skip to main content

2000 | Buch

Scheduling: Control-Based Theory and Polynomial-Time Algorithms

verfasst von: Konstantin Kogan, Eugene Khmelnitsky

Verlag: Springer US

Buchreihe : Applied Optimization

insite
SUCHEN

Über dieses Buch

This book presents a first attempt to systematically collect, classify and solve various continuous-time scheduling problems. The classes of problems distinguish scheduling by the number of machines and products, production constraints and performance measures. Although such classes are usually considered to be a prerogative of only combinatorial scheduling literature, the scheduling methodology suggested in this book is based on two mathematical tools - optimal control and combinatorics. Generally considered as belonging to two totally different areas of research and application, these seemingly irreconcilable tools can be integrated in a unique solution approach with the advantages of both. This new approach provides the possibility of developing effective polynomial-time algorithms to solve the generic scheduling problems. This book is aimed at a student audience - final year undergraduates as well as master and Ph.D. students, primarily in Operations Research, Management, Industrial Engineering and Control Systems. Indeed, some of the material in the book has formed part of the content of undergraduate and graduate courses taught at the Industrial Engineering Department of Tel-Aviv University, the Logistics Department of Bar-Ilan University and the Technology Management Department of Rolon Center for Technological Education, Israel. The book is also useful for practicing engineers interested in planning, scheduling and optimization methods. Since the book addresses the theory and design of computer-based scheduling algorithms, applied mathematicians and computer software specialists engaged in developing scheduling software for industrial engineering and management problems will find that the methods developed here can be embedded very efficiently in large applications.

Inhaltsverzeichnis

Frontmatter

Introduction

Frontmatter
1. Introduction to Control-Based Scheduling
Abstract
Modern production systems are characterized by resources and the products these resources are designed to manufacture in response to market demand. When demand for products at a time horizon is given, scheduling generally means assigning individual resources to individual products in order to trace the demand along the horizon. Since typically, a large number of possible schedules can be constructed, the problem of scheduling is to select the schedule which ensures optimal functioning of the production system in terms of a certain performance measure. Due to the great variety of production systems, of performance measures as well as of production demand forms, a vast number of scheduling models has been developed to mathematically formalize the scheduling problem.
Konstantin Kogan, Eugene Khmelnitsky
2. Mathematical Background
Abstract
The methodology suggested in this book is based on three mathematical tools — optimal control, combinatorics and mathematical programming — which are traditionally related to separate areas of research and application. The methodology involves, first, investigating a continuous-time problem with the aid of the maximum principle and then reducing it to a discrete combinatorial or mathematical programming problem solvable in polynomial time. The following sections present selected combinatorics, the maximum principle and a constructive approach for integrating both mathematical tools.
Konstantin Kogan, Eugene Khmelnitsky

One-Machine, One-Product-Type Scheduling

Frontmatter
3. Unlimited Capacity Problems
Abstract
This section addresses the scheduling problems characterized by demand as a function of time specified over a finite planning horizon. The problems differ by inventory-related constraints that allow or disallow backlogs and by the form of the objective function.
Konstantin Kogan, Eugene Khmelnitsky
4. Limited Capacity Problems
Abstract
This section addresses the capacitated scheduling problems characterized by demand as a function of time specified over a finite planning horizon. The problems differ by inventory related constraints, allowing or disallowing, backlogs and by the form of the objective function.
Konstantin Kogan, Eugene Khmelnitsky

Multiple Product-Type Scheduling

Frontmatter
5. One-Machine Problems
Abstract
Scheduling multiple product types with unlimited production capacity is straightforwardly decomposed into a number of problems, each of which deals with a single product. The number of such problems is equal to the number of products to be scheduled. Such one-machine, one-product-type problems have been discussed in the previous part of the book. In contrast, scheduling multiple product types, with limited capacity, cannot be decomposed. This is because different products compete for the common capacity in a different, changing-in-time manner Because the multi-product, capacitated scheduling problem is much harder than the problems considered in the previous chapters, the objective of this chapter is to study the optimal behavior of production systems under only linear performance measures.
Konstantin Kogan, Eugene Khmelnitsky
6. Parallel Machine Problems
Abstract
This chapter deals with scheduling of multiple machines and products under both limited and unlimited production capacity. Since the unlimited capacity problems can be decomposed into a number of one-machine, one-product problems, they are considered only in Section 6.1, for general demands over a finite planning horizon. Because the capacitated, multi-machine, scheduling problems are much harder to solve than the problems considered in the previous chapters, the objective of this chapter is to study the optimal behavior for only some special cases of scheduling parallel machines with a linear performance measure.
Konstantin Kogan, Eugene Khmelnitsky
7. Serial Machine Problems
Abstract
This book is devoted to classifying and solving scheduling problems in the production systems of simple machine configurations. Although such configurations as job-shop, open-shop, and re-entrant lines are not within the book’s scope, the maximum principle applied to scheduling problems in complex machine configurations may still result in efficient combinatorial algorithms. This chapter deals with scheduling of multiple tandem machines with limited production capacity and demand concentrated at a due date. Among the various scheduling problems, this one is chosen to illustrate the solution methodology suggested in this book for those cases when configuration of the production system is not as simple as a parallel-machine cell.
Konstantin Kogan, Eugene Khmelnitsky
Backmatter
Metadaten
Titel
Scheduling: Control-Based Theory and Polynomial-Time Algorithms
verfasst von
Konstantin Kogan
Eugene Khmelnitsky
Copyright-Jahr
2000
Verlag
Springer US
Electronic ISBN
978-1-4615-4675-7
Print ISBN
978-1-4613-7116-8
DOI
https://doi.org/10.1007/978-1-4615-4675-7