Purpose: This volume is intended as a textbook for a first year graduate or a senior undergraduate course on Production Systems. The aim is two-fold. First, it is to present material that is practical and applicable to any production system in large volume manufacturing industries, such as automotive, electronics, appliances, etc. Second, it is to present this material at the same level of rigor as that in other engineering disciplines, such as Electrical Engineering, Mechanical Engineering, etc. Therefore, we use the title Production Systems Engineering (PSE).
Main Topic: Production systems are machines and material handling devices arranged so as to produce a desired product. Each machine is characterized by three attributes:
Each material handling device is characterized by:
This textbook does not address the issues of operation technology of either machines or material handling devices. Addressed are the issues of parts flow through a production system. The nature of parts flow is affected by production capacity and reliability of the machines and storing capacity of the material handling devices. In general, numerous issues related to parts flow could be addressed. Included in this volume are only those, which have a clear practical significance. Thus, the rigorous engineering study of practical issues related to parts flow in production systems with unreliable machines and finite buffers is the main topic of this textbook.
Origin: The origin of this volume is also two-fold. First, it is based on 20+ years of industrial studies, which one of the authors (S.M. Meerkov), together with his students (one of whom is the other author, J. Li), conducted in various automotive plants. Every problem considered in this book originated on the factory floor and, after appropriate conceptualization and analysis, ended up as an application on the factory floor. The case studies included in this volume describe some of these applications.
Second, this textbook is based on a course on production systems, which we have been teaching at the University of Michigan for many years. The audience included both full time university students and part time students from manufacturing industries. (In most cases, this course was available through the University of Michigan distance learning system to practicing engineers throughout the US and indeed the world.) Reading materials for this course consisted of journal papers, which described results of our investigations. Both groups of students encouraged us to summarize these results in a textbook. We also felt that such a book would be a contribution to the literature. That is why we undertook the arduous effort of writing this textbook, which lasted for almost five years.
Problems addressed: Three groups of problems are considered:
This textbook provides analytical methods for solving these problems and illustrates them by case studies.
Main difficulties: Unreliable machines and finite buffers make the problem of parts flow in production systems difficult since the former makes it stochastic and the latter nonlinear. In more specific terms, the difficulties in investigating production systems arise due to mutual interferences of the machines because of breakdowns. Indeed, a breakdown of one machine affects all other machines in the system – by blocking those upstream and starving those downstream. Buffers are supposed to alleviate these perturbations, serving as "shock absorbers" for parts flow. However, having them "infinite" and, hence, efficient for "shock absorbing", creates economic problems. Thus, the buffers must be finite and the machines, clearly, cannot be absolutely reliable. These features lead to mathematical models of production systems, which are nonlinear and stochastic. Typically, they are difficult to analyze. Thus, the unreliability of the machines and the finiteness of the buffers are the main sources of difficulties in production systems analysis, continuous improvement, and design.
Method of investigation: Often, production systems are investigated quantitatively based on queuing theory methods. In this book, however, we use a different approach. It is based on recursive equations, which describe production systems at hand. These equations are derived using exact analysis of the simplest systems and subsequent approximate analysis of more complex ones. The properties of the resulting equations characterize the flow of parts and expose the laws, which govern production systems behavior. Thus, a system-theoretic analysis, based on recursive equations that describe production systems at hand, is the method of investigation employed in this textbook.
Relation to other engineering disciplines: Many engineering disciplines are also concerned with flows. In Electrical Engineering this is the flow of current in a circuit. In Mechanical Engineering, it is fluid or gas flow. Focusing our attention on parts flow makes Production Systems Engineering similar to other engineering disciplines. Therefore, rigorous quantitative methods, typical for modern engineering, become possible and, moreover, necessary in this area of technology.
As in other disciplines, flows are characterized by their statics (i.e., stationary regimes) and dynamics (i.e., transient regimes). Although this book centers on the statics, results on the dynamics and transients are also included.
Solution paradigm: The solutions of the problems considered in this textbook are given in three classes of results. The first class consists of results proved analytically; they are referred to as Theorems. The second class is comprised of properties, which are shown to exist numerically and approximately; they are referred to as Numerical Facts. Finally, the third class includes results called Improvability Indicators; they are also justified numerically and are intended as guides for production systems improvement in the framework of the so-called Measurement-based Management. Converting the Numerical Facts and Improvability Indicators into Theorems could be a fruitful direction for future research.
Intended audience, outcomes, and prerequisites: This textbook is intended for senior and beginning graduate students in all engineering disciplines. In addition, business and management students may be interested in a course based on this textbook. Finally, practicing production engineers and managers may find it useful to read this book.
As an outcome of this course, the students are expected to acquire rigorous and practical knowledge on design and management of production systems.
No specific prerequisites are assumed; however, prior exposure to Probability Theory may be beneficial (although all probability facts, necessary for this course, are described in this textbook).
Book organization: The textbook consists of five parts, each comprised of several chapters. The first part includes background and modeling material, while the rest are devoted to various classes of production systems of increasing complexity. Specifically, Part II is devoted to serial lines with so-called Bernoulli machines, Part III treats similar problems for exponential and general models of machine reliability, while Part IV addresses assembly systems. Finally, Part V includes a summary of the main facts of Production Systems Engineering, a description of the PSE Toolbox, which is a suite of C++ programs that implement all methods developed in this book, and the proofs of the theorems presented in this volume.
In each part, every chapter begins with a motivating comment, which provides a reason for considering its subject matter. This is followed by an overview, which outlines specific problems addressed and results obtained. Each chapter is concluded with a set of homework problems and an annotated bibliography. Most chapters include case studies based on industrial applications carried out by the authors in the automotive industry.
Special features: While other books on production systems are centered mostly on performance analysis, the present volume has the following special features:
Chapter contents: Chapter 1 places production systems in the broader context of manufacturing at large. Chapter 2 provides an overview of Probability Theory and derives several facts on machine reliability and performance used throughout this book. Chapter 3 describes the process of mathematical modeling of production systems.
Chapters 4 - 10 are devoted to serial lines with the simplest, i.e., Bernoulli, model of machine reliability. Specifically, Chapters 4, 5, and 6 address, respectively, the issues of analysis, continuous improvement, and design. Chapters 7.
and 8 are devoted to closed lines and product quality, respectively. Chapter 9 addresses the issue of customer demand satisfaction, and Chapter 10 discusses transient behavior of serial lines.
Chapters 11 - 15 are devoted to similar issues in the framework of serial lines with continuous time (exponential and non-exponential) models of machines reliability, while Chapters 16 and 17 consider assembly systems.
Chapter 18, summarizes the main facts of Production Systems Engineering. Chapter 19 describes the PSE Toolbox. Finally, Chapter 20 provides proofs of theorems and other formal statements included in this textbook.
Advice to Instructors: There are several ways to structure a semester-long course based on this textbook. If the audience has limited background in Probability Theory, the course could cover Parts I and II in detail and a brief overview of Parts III and IV (mainly, the ideas of the aggregation procedures and bottleneck identification techniques). For an audience with strong background in probability, the emphasis could be on Chapter 3 and Parts II and III; basic ideas of Part IV could also be covered.
Similar approaches can be used for a quarter-based course but, of course, with a less detailed coverage of the material. A two-semester course will cover the entire text.
The proofs of most mathematical statements included in this volume are typically not covered in class. However, doctoral-level students specializing in manufacturing may find it useful to study these proofs in order to develop their expertise for theoretical research; that is why the proofs are included in Chapter 20.
Finally, a solution manual of all problems included in this textbook is available from the Publisher upon instructor’s request.
Acknowledgements: The material of this textbook was developed by the authors and the Ph.D. students working under SMM’s supervision. These include (in chronological order) Drs. Jong-Tae Lim, Ferudun Top, David Jacobs, Chih-Tsung Kuo, Shu-Yin Chiang, Emre Enginarlar and Liang Zhang. The authors are particularly grateful to Liang Zhang who co-created the PSE Toolbox, helped writing Chapters 7, 8, 10, 11, 13, 17 and spent countless hours proofreading and correcting the manuscript. His contribution is difficult to overestimate.
In addition, the following graduate and undergraduate students contributed to this book: S. Ching, A. Hu, J.Z. Huang, A. Khondker, Y. Liu, B. Rumao, H. Shih and F. Xu. To all of them the authors express their deep gratitude.
The stimulating environments of the Department of Electrical Engineering and Computer Science at the University of Michigan, the General Motors Research and Development Center (with which JL was associated) and the University of Kentucky were very conducive to the research that led to this book.
Financial support of the National Science Foundation over the period of time when this material was developed and this textbook written was invaluable for the success of this work. Gratitude to the Division of Civil, Mechanical and Manufacturing Innovations of the National Science Foundation is in order.
Special thanks are due to many colleagues throughout the world for their advice and encouragement. First of all, the authors are indebted to Professor Pierre Kabamba from the University of Michigan for his technical advice, impeccable scientific taste, and countless hours spent discussing the material included in this textbook. He carefully read every chapter and made valuable suggestions. Professor Stanley Gershwin of the Massachusetts Institute of Technology was a source of inspiration from the very beginning of the authors’ work in the area of manufacturing. He truly can be viewed as the father of the system-theoretic approach to manufacturing, a small part of which – production systems – is addressed in this textbook. Professor Chrissoleon Papadopoulos of the Aristotle University of Thessaloniki, Greece, is acknowledged for both his personal input to our work and for a series of stimulating conferences, which he has organized for many years and which provided an outstanding opportunity to discuss the issues of production system research with numerous colleagues from all over the world.
Our industrial partners, who helped to test the methods described in this book in the framework of numerous projects on the factory floor, are gratefully acknowledged. This includes management and production system personnel from
While we thankfully acknowledge all of the individuals and organizations mentioned above, needless to say that all errors, which are undoubtedly present in this book, are due to the authors alone. The list of corrections will be maintained at
http://www.ProductionSystemsEngineering.com/book/corrections.html
Last, but not least, we are deeply indebted to our families for enduring countless hours missing us in family activities. Their love and support truly made this work possible.