Book Reviews

Iravani, Seyed M.R. 2020. Operations Engineering and Management: Concepts, Analytics and Principles for Improvement. McGraw Hill. 874 pp. $138.00.

Since I entered an industrial engineering program as a freshman in 1980, the content and structure of the textbooks on operations have changed remarkably. The two textbooks that, from my perspective, represent the evolution of the field, are the Johnson and Montgomery (1974) classic Operations Research in Production Planning, Scheduling and Inventory Control and the very different Factory Physics (Hopp and Spearman 2008), which was published almost a quarter of a century later. The earlier book presents a variety of mathematical models, ranging from deterministic lot sizing and linear programming models to stochastic inventory models and time series for demand forecasting—very different tools for very different problems arising in the manufacturing environment. Services and queueing models are conspicuous by their almost complete absence. The reason I have taught out of the Hopp and Spearman (2008) book for more than 20 years is the eponymous second section (chapters 6–9). For the first time in my experience, a book presented a unified way to view production and service systems through the idea of queueing, used this view to derive insights into their behavior, and gave a clear sense of the data needed to analyze these systems. The book reviewed here is a recent (2020) addition to the stable of books serving industrial engineering and management students in operations courses at the undergraduate and beginning graduate levels. It is supported by a variety of online resources, including PowerPoint slides, supplements for several chapters that provide additional material and associated problems, and appendices that present mathematical details and derivations. Each chapter is also followed by a substantial set of problems—a valuable resource for instructors. It is immediately apparent that this book is a labor of love; the author’s passion for the subject and deep-seated desire to engage the reader and convey the insights are evident on every page and are a major asset of the finished product.

Introductory courses on operations, whether in engineering or management programs, generally have two rather different objectives. The first is to give students from varied backgrounds and levels of preparation a unified perspective for approaching these problems; chapters 1–11 accomplish this admirably. The second is to provide them with a basic vocabulary of commonly used methods and concepts that they are likely to encounter when they enter the workforce. These include topics such as material requirements planning (MRP), lean manufacturing/services, theory of constraints, and Six Sigma, which are covered in chapters 12–17. The degree to which each objective will be addressed in a single class is a difficult decision, and the book is designed to allow instructors extensive flexibility in choosing topics to achieve their preferred learning outcomes.

Iravani’s book represents an interesting evolution over its predecessors in distinguishing between concepts, analytics, and principles. Concepts provide an idiom for describing operations problems; analytics stipulate the calculations of various quantities of interest defined by the concepts that quantify how the system is operating; and principles provide managerial guidelines gained from the concepts and the analytics. The book begins with a series of “preliminary” chapters that lay out the problem space and the library of concepts that support the “improvement” chapters (e.g., addressing improvements to throughput, inventory, quality, flow time, operations with flexibility, and lean tools). The “analytics” chapters link concepts and calculations (i.e., mathematical models) to support the improvement chapters.

The four preliminary chapters begin by discussing the role of operations in supporting the firm’s value proposition to its customers, which, in turn, defines both the nature and the volume of demand to be served. This material is important in situating operations engineering and management within the business needs of the firm and is often underemphasized in engineering courses. The second preliminary chapter presents the process flow framework for operations. Giving it its own chapter, which discusses real issues, such as the need to define units of flow and process boundaries clearly and consistently, is an excellent choice, leading to the definition of performance measures, such as throughput and inventory, and their relationship via Little’s law. A particularly useful aspect is the explicit discussion of how to handle multiple products, which is often confusing to students. The third preliminary chapter, on demand forecasting, provides a solid but fairly conventional treatment of forecasting models for time series and a useful discussion of the need to monitor the performance of forecasting systems on an ongoing basis.

Chapter 4 introduces the fraught subject of capacity, which is often oversimplified in mathematical models, teaching, and industrial practice (Elmaghraby 1991, 2010). The book follows Hopp and Spearman (2008) in basing its analysis on the effective processing time, the time required to produce one saleable flow unit, allowing a distinction between theoretical and effective capacity. The chapter then extends these ideas to systems with multiple products and stages, the latter leading to the definition of bottlenecks as the stage with the longest effective processing time. The focus on mean performance measures leads to a deterministic analysis that is complemented by the subsequent chapter. Chapter 5 addresses variability in demand, considering both demand per unit time and the interarrival times between demands as well as strategies to reduce demand variability, such as demand aggregation and reduction of batch sizes in the case of bulk demand arrivals. It then proceeds to discuss variability in processing times and processing flows, using the coefficient of variation of the effective processing time as the basic measure of variability. The natural variability of the processing time is augmented by the variability because of time- and task-based interruptions, corresponding to the preemptive and nonpreemptive interruptions of Hopp and Spearman (2008). An obvious question is that of how to analyze variability when both time- and task-based interruptions are present. A brief discussion of this issue, which would allow the interested reader to form an overall picture of process variability, together with an illustrative example, would pull chapter 5 together very nicely. An extremely important benefit of the process flow perspective based on effective processing times is that the equations in chapters 4 and 5 provide a clear idea of what data need to be collected to analyze system performance.

Chapter 6 builds on the discussion of capacity and variability to examine ways to improve process throughput. It begins by discussing causes of throughput loss, such as lack of supply (starvation), limited buffer space, and balking/abandonment, and distinguishing between demand- and capacity-constrained systems. The focus on improving bottleneck processes is important, but in practice, improvements in bottleneck processes that have been identified as such may be hard to come by either because the system has already been the focus of extensive improvement efforts or because of cost considerations. Such situations often provide opportunities to improve the performance of the bottleneck by reducing the variability of flow to the bottleneck, which is the subject of the next chapter.

Variability comes into its own in chapter 7, whose focus is flow time (i.e., cycle time in Hopp and Spearman (2008) terminology) and whose primary vehicle for the analysis is the heavy traffic approximation for the G/G/1 queue because of Kingman (1961) and its multiserver extensions with additional consideration of limited buffer sizes. This chapter supports the next chapter on flow-time improvement, which considers both system design issues (e.g., increase capacity, reduce variability in processing, pool resources) and behavioral actions, such as providing arriving customers with clear information on what to do when they arrive.

Chapter 9 provides an introduction to the concepts underlying inventory models with emphasis on deterministic demand and, hence, the economic order quantity model and the trade-offs in lot sizing, concluding with the economic production quantity model. The use of the (Q, R) policy as a framework for the basic decisions (e.g., when to order? how much to order when I do order?) makes a lot of sense and does a good job of setting up the stochastic inventory models in the next chapter. Conspicuous by its absence is the deterministic dynamic lot-sizing problem and the classical Wagner–Whitin algorithm. The value of this latter in a book of this nature is debatable; its inclusion in many MRP packages suggests it is a useful concept for students to be familiar with, but the dynamic programming solution is something of an outlier in terms of analytics. Hence, its omission is justified despite the sense of sacrilege one feels at seeing a venerable result written out of a basic text. Chapter 10 covers single-stage stochastic inventory models, clearly and concisely presented with interesting discussions of the trade-offs involved. Chapter 11 on inventory improvement then introduces a set of tools for reducing inventories, including reducing the mean and variance of replenishment lead time, various forms of inventory pooling and coordination of multistage systems using contracts to account for potentially conflicting incentives among the interacting entities, and a discussion of the bullwhip effect. Overall, the discussion of inventory systems covers all the basic concepts and does a good job of connecting them to each other and to the analytics.

Chapter 11 concludes what I would call the core of the book: the development of the basic perspective leading to the primary performance measures of throughput, flow time, and inventory, the mathematical models (analytics) supporting the diagnosis and improvement of production and service systems, and the approaches for improving them. Chapters 12–17 present additional topics that supplement the discussion in the previous chapters but are loosely connected with it. Chapters 12 and 13, on aggregate planning and operations scheduling, cover the topics that were central to books of the Johnson and Montgomery (1974) generation, that is, the use of deterministic optimization models for allocating demand to capacity over time (chapter 12) and order release and job scheduling on the shop floor and in a project management context (chapter 13). The latter chapter includes the widely used MRP logic as well as examples of well-solved (i.e., polynomially solvable) single and parallel machine-scheduling problems together with a brief introduction to assembly line balancing. Chapter 14 addresses a variety of ways in which resource flexibility can be exploited to the advantage of operations, whereas chapter 15 on lean operations is essentially a self-contained short course on lean manufacturing. Chapters 16 and 17, on quality control and Six Sigma, are also in the nature of introductions of broad areas to build awareness and direct the interested student to more specialized sources and courses. The inclusion of these additional chapters allows the book to be used in business schools in which, contrary to many engineering programs, some of these topics form part of the core operations courses. They also allow instructors to combine selected chapters into course packages to serve a variety of elective courses in related areas.

In summary, this book is a lineal descendant of the Hopp and Spearman (2008) book, which extends it in several meaningful ways: the organization into concepts, analytics, and principles; the explicit discussion of improvement; the heavy use of cases in the core chapters to engage readers and convince them the problems are complex but amenable to analysis; and the additional chapters that provide flexibility to configure quite different courses based on audience and learning objectives. I would certainly consider using this textbook in the entry-level graduate classes and senior-level undergraduate courses I have taught, especially because the book will serve as a valuable reference for students going into operations-related positions after graduation. The author is to be commended for providing a book that is simultaneously comprehensive, readable, and insightful—an achievement whose difficulty can only be appreciated by those who have tried to accomplish it.

Reha Uzsoy, PhD, PEEdward P. Fitts Department of Industrial and Systems Engineering, North Carolina State University, Raleigh, North Carolina 27695 ruzsoy@ncsu.edu