The attention paid to computer-integrated manufacturing (CIM) has increased dramatically in the last ten years. Intense industrial activity to try to harvest the productivity gains from automation has been balanced by equally heavy academic and research activity. It is now generally recognized that formal training for engineers is incomplete without at least an introduction to CIM. Earlier research papers, presentations, and case studies have been supplemented by texts and other full-length presentations on CIM.

The publications selected for this review have copyright dates of 1990 or 1991. They were chosen because they include “CIM” or “computer-integrated manufacturing” in their titles, and they cover a broad spectrum of topics. Books with a narrow focus, such as economic justification or engineering design, were omitted. The books covered at least approach the concept of integration, rather than just the use of computers in manufacturing. Using these criteria, I selected 6 books from a list of more than 50.

The SME CASA wheel was chosen as a basis for comparing these books (see Table 2) because it is a widely understood representation of the CIM functions. It provides a common list of functions that are generally accepted as necessary and sufficient to claim integration of manufacturing enterprises with computers. It does not address the associated people and planning issues or the other topics compared in Table 3. Table 4 indicates the appropriate audience for each book.

Bedworth, Henderson, and Wolfe

The focus of this text is on computer-aided design (CAD) and computer-aided manufacturing (CAM). It does not give a complete view of CIM. The audience is fourth-year and graduate students. Each chapter highlights key definitions and includes exercises and references. The structure is good for a student text. Each chapter includes a mathematical formalism for one of the concepts it covers. These formalisms seem to have been forced into the information flow by a desire to require linear algebra as a prerequisite for the text. They are overly simplistic representations of the real mathematical models and serve no real purpose.

Chapter 1 is a history of CAD and an overview of CAM. Chapter 2 introduces geometric modeling of surfaces and solids. Chapter 3 covers CAD. Chapter 4 covers an important topic: concurrent engineering. The topics are the standard list of methods (such as Boothroyd’s), but many views are omitted, including serviceability, testability, and reliability. Four coding schemes for group technology are described in chapter 5. Variant and generative process planning are described in chapter 6. Chapter 7 is an attempt to encompass other CIM topics; it includes materials requirements planning (MRP), capacity requirements planning (CRP), master production scheduling, capacity planning, shop floor scheduling, quality assurance, facility control, and just-in-time. The formulas for linear regression are also included. Chapter 8 is a collection of topics related to manufacturing control; it also includes a section on converting between hexadecimal and decimal notation. Chapter 9 provides extensive coverage of numerical control programming. Most of chapter 10 discusses two robotics languages, but SIMAN is introduced as a simulation tool. Chapter 11 covers measurement systems, and chapter 12 presents information systems concepts including relational databases and communication network protocols.

This work looks like a good book for the intended audience, but it is not a book on CIM; it is a book on two of the many aspects of CIM. It does not provide an integrated view of manufacturing or of manufacturing information systems.

Bertain

Each chapter in this compendium of papers on CIM implementation is by a different author. Each author seems to feel that it is necessary to justify CIM. As a result, the book contains a lot of repetitive material, composing as much as a quarter of the book.

Part 1 consists of three papers. The first is a brief history of CIM, whose main message is that technology is not the answer, people are. It also discusses computer-aided logistic support (CALS) and CIM and the focus on concurrent engineering; the paper again emphasizes the need to consider people and their skills. The next paper provides guidelines for CIM implementation, including upper management support, simplicity, education and training, and communication (in order to draw on the knowledge of the shop workers). The need for a centralized database and a trial implementation is also noted. The last paper in this part discusses strategic planning. The focus is on how to justify investing in CIM, based on indirect costs and savings; the main message is to improve business processes first.

Part 2 comprises six papers on CIM tools. The first proposes event map diagramming as a successful tool for CIM implementation. This method is really just a way to think about shop floor problems, not a general CIM tool. The next paper describes the ESPRIT Computer Integrated Manufacturing–Open Systems Architecture (CIM-OSA) and robotics projects, but gives no technical detail. Another paper describes the current state of data exchange for CAD/CAM. Overviews of the Initial Graphics Exchange Standard (IGES) and Product Design Exchange Specification (PDES) are provided. The next paper describes alternate coding schemes for group technology and a successful implementation of group technology. It includes descriptions of the integrated definition method for processes (IDEF0) and integrated definition method for data (IDEF1X), but lacks references. The last paper in this part is a case study of integrated modeling, design, and operations tools.

Part 3 includes three case studies from partial CIM implementations by US firms. Part 4 is a case study of a CIM implementation at a Japanese firm. Part 5 is an overview of the PDES standard for product data exchange. No references are included to allow the reader to find additional information. Part 6 is a bland report on a survey of CIM planning.

ESPRIT Consortium

A consortium of companies from several European countries produced this research report. It includes the CIM-OSA architectural concepts and framework; the modeling levels; the parts of the framework and their relationships; the CIM-OSA life cycle, business process design, and execution; and the results of standardization work.

In a summary of future needs of the manufacturing industry, the authors emphasize the ability to manage change, real-time control of the total manufacturing process, adaptability, an explicit description of the processes in a form that can be executed, availability of information, more effective deployment of information technology, and the ability to combine equipment from different vendors. To achieve the last goal, they define the architecture in terms of generic building blocks and partial models that can be customized into particular models for an application. Requirements are gathered in a structured fashion and transformed into particular solutions. The transformations are formally defined and can presumably be automated.

The authors define the objective of CIM as “appropriate integration of exchange of information within a company with the help of computer aided tools.” The CIM Reference Architecture is a cube consisting of modeling levels (from top to bottom), architectural models (from left to right), and views (from front to back). The four views are function, data, resources, and organization. The authors introduce the concepts of “build time” and “runtime.” Traversal (instantiation, derivation, and generation) of the cube from generic building blocks and requirements leads to a particular implementation that can be released for use. The method provides a hierarchical decomposition of business processes so that vendors who are CIM-OSA compliant can compete with standard solutions. The work is still precompetitive research. Given the growth potential for European manufacturing, compliance with this standard may make the difference between success and failure for CIM vendors, particularly in the European market.

I recommend this book for CIM architects who will be designing either particular implementations or vendor offerings for the CIM market. It is required reading for researchers as a formal way to express their work and to demonstrate how it fits with other CIM activities.

Foston, Smith, and Au

The preface claims that this book is “virtually an encyclopedia on the factory of the future.” It is not an encyclopedia, but it is a pretty good introductory book for manufacturing students. It contains numerous figures, but they are inconsistent in style and have no common thread. Typically, the same concept is shown in at least three different forms. It looks as though the authors borrowed as many graphics as they could find and used them all. The text is clear, and I found no glaring errors. Each chapter ends with exercises and references. The exercises are not challenging, but the references are adequate.

The book is divided into two parts: “Fundamentals of Product Processes and Operations” (chapters 1 through 6) and “Manufacturing Systems Integration Techniques and Strategies” (chapters 7 through 10). Part 1 is better than Part 2.

Chapter 1 is an introduction to manufacturing from an organizational perspective. CIM is not defined until much later, in chapter 7. Chapter 2 discusses computers in manufacturing, including automation, computer control, plant communications (in amazing detail, including the color standards for leads on RS-232 connectors), and manufacturing databases. Chapter 3 is an already outdated discussion of computer-aided engineering, including lots of pictures of computer hardware. Less than one page is devoted to each of the following: finite element analysis, geometric modeling, wire frame modeling, and solid modeling. Similarly brief sections present process planning, numerical control, and mold and tool design. This chapter will be useful only to someone totally lacking in knowledge of manufacturing.

Chapter 4 covers computer-aided production systems. It reads like wishful thinking: we are told that the computer does things, with no mention of the complexities of applications code and database development needed to get the computer to perform even the simplest of the tasks mentioned. This chapter is another flyover for people with no knowledge of manufacturing. Chapter 5 discusses the integration of CAD and CAM technologies. The discussion leads the reader to the conclusion that distributed database technology is readily available for all CIM applications. The chapter includes a good section on interprocess information flows. Chapter 6 covers computer-aided business applications, including production management and MRP.

Chapter 7 presents the benefits of CIM. The variety of graphics in this chapter is particularly confusing. Along with the benefits, the author presents such topics as hierarchical computer systems and database management. Chapters 8, 9, and 10 are brief summaries of CIM planning, human factors, and implementation.

Mitchell

The focus of this book is on two manufacturing sectors: machine tools and small electronic assemblies. It claims to be an introduction for students. Assignments are included with each chapter. Mitchell defines CIM as follows: “CIM refers to the application of system concepts to produce a manufacturing enterprise that can best achieve performance objectives in today’s market setting.”

Chapter 1, “Applying Open Systems Concepts,” has nothing to do with the open systems referenced in computer science or other CIM literature; instead, it refers to open-minded problem solving. It starts by describing how to do systems analysis, but the reader would have to be experienced in systems analysis to gain much from this discussion. Modeling views are discussed, but they do not match the CIM-OSA standard or constitute a complete set of primitives for modeling. The CIM-OSA work is mentioned and dismissed, as the author did not understand the power of that formalism. Petri nets, PERT charts, and graph trees get only a paragraph each.

Chapter 2 is a history of CIM, but the examples focus on transportation systems and post offices. It makes no mention of major corporate efforts in the US or the competitive pressure from Japan. Chapter 3, “Manufacturing System Design,” spends four pages on IDEF modeling but omits constraints and mechanisms. The rules for modeling are not covered. This chapter introduces the author’s own system environment simulation (SES) model, which is supplied on an accompanying diskette. The model is a gross oversimplification of the manufacturing process, and the program seems to have been designed for a junior high school audience. Statistical quality control is dismissed in two pages.

Chapter 4, “Implementing System Design Concepts,” emphasizes the role of a planning group to help evolve the system. It makes no mention of establishing goals and objectives, and does not acknowledge the use of existing data to aid in implementation design. Chapter 5, “Software for Modeling Support,” is a random collection of vendor-supplied screen snapshots. It provides no consistent view of a problem or model sequence. Chapter 6 is another random collection of vendor-supplied drawings and photos of manufacturing equipment. The author does not discuss how or whether this equipment might be included in a CIM system.

Chapter 7 is another strange collection of historical information about computers. It is inconsistent and incomplete. Chapter 8 is a brief overview of a collection of organizational theories. Chapters 9 and 10 are no better than the others. This book is a random collection of readily available material on manufacturing and CIM. The presentation is neither coherent nor consistent. The book should not be used.

Scheer

The basis for this book is CIM implementations in Germany; some US implementations are also discussed. The architecture used as a basis for discussion is the Y-CIM model (a Y-shaped representation of CIM functions), with a strong focus on the need for one common database. Scheer defines CIM as the integration of the process and the data, which starts with the design of the product and extends through all manufacturing processes, including office automation. A complete CIM implementation is described by showing the data flows among key components.

The discussion of production planning demonstrates a good understanding of the complexity of the problem and why simple-minded approaches will not help. This discussion includes a summary of the current state of automation of production planning and control. The author also summarizes the current state of CAD as well as work on data interchange standards. He briefly discusses computer-aided quality.

The section on implementation of CIM includes a good review of simultaneous product development (which most US texts call concurrent engineering) and assesses the problems realistically. Scheer includes intercompany communication and financial control systems. This book contains many clear diagrams. Levels of integration are described, and communications protocols are introduced. The author describes several successful CIM implementations.

Sections describe the German CIM Technology Transfer Centers and the ESPRIT CIM architecture projects. The book includes extensive references and a good index. I recommend it as supplemental reading for any CIM course, as it provides an alternate data architecture and justifies the assertions with successful examples. I also recommend it for engineering and manufacturing staff who desire a concise understanding of the whole problem.

Comparison

I cannot heartily recommend any of these books for upper-level undergraduate or graduate use as a complete textbook on CIM. The Bedworth book covers design and manufacturing of products, but it is not a complete CIM reference. Foston is the best of the books for this audience, but it would need consistent graphics in order to be recommended. Experienced manufacturing and CIM personnel will benefit from the Bertain and Scheer books. Both are easy to read and worth the time invested. The ESPRIT book is required reading for CIM architects and researchers. Several hours of concentrated study will be required. Mitchell’s book should not be used for any purpose related to CIM.

Numerous texts and supplemental books on CIM are available. None of the books I reviewed is a perfect match for classroom needs, but an adequate set can be compiled from the available works. A good writer/publisher combination would have the opportunity to take over the market for a CIM textbook. Such a book would probably require collaboration between an academic and a practicing manufacturing integrator to produce a quality student text.