International Series in Operations Research & Management Science – serie

The 5th edition of this classic textbook covers the central concepts of practical optimization techniques, with an emphasis on methods that are both state-of-the-art and popular. One major insight is the connection between the purely analytical character of an optimization problem and the behavior of algorithms used to solve that problem. End-of-chapter exercises are provided for all chapters.The material is organized into three separate parts. Part I offers a self-contained introduction to linear programming. The presentation in this part is fairly conventional, covering the main elements of the underlying theory of linear programming, many of the most effective numerical algorithms, and many of its important special applications. Part II, which is independent of Part I, covers the theory of unconstrained optimization, including both derivations of the appropriate optimality conditions and an introduction to basic algorithms. This part of the book explores the general properties of algorithms and defines various notions of convergence. In turn, Part III extends the concepts developed in the second part to constrained optimization problems. Except for a few isolated sections, this part is also independent of Part I. As such, Parts II and III can easily be used without reading Part I and, in fact, the book has been used in this way at many universities.New to this edition are popular topics in data science and machine learning, such as the Markov Decision Process, Farkas’ lemma, convergence speed analysis, duality theories and applications, various first-order methods, stochastic gradient method, mirror-descent method, Frank-Wolf method, ALM/ADMM method, interior trust-region method for non-convex optimization, distributionally robust optimization, online linear programming, semidefinite programming for sensor-network localization, and infeasibility detection for nonlinear optimization.

This book is an expository introduction to the methodology of sensitivity analysis of model output. It is primarily intended for investigators, students and researchers that are familiar with mathematical models but are less familiar with the techniques for performing their sensitivity analysis. A variety of sensitivity methods have been developed over the years. This monograph helps the analyst in her/his first exploration of this world. The main goal is to foster the recognition of the crucial role of sensitivity analysis methods as the techniques that allow us to gain insights from quantitative models.  Also, exercising rigor in performing sensitivity analysis becomes increasingly relevant both to decision makers and modelers. The book helps the analyst in structuring her/his sensitivity analysis quest properly, so as to obtain the correct answer to the corresponding managerial question. The first part of the book covers Deterministic Methods, including Tornado Diagrams; One-Way Sensitivity Analysis; Differentiation-Based Methods and Local Sensitivity Analysis with Constraints.  The second part looks at Probabilistic Methods, including Regression-Based methods, Variance-Based Methods, and Distribution-Based methods. The final section looks at Applications, including capital budgeting, sensitivity analysis in climate change modelling and in the risk assessment of a lunar space mission.

This volume reflects the theme of the INFORMS 2004 Meeting in Denver: Back to OR Roots. Emerging as a quantitative approach to problem-solving in World War II, our founders were physicists, mathematicians, and engineers who quickly found peace-time uses. It is fair to say that Operations Research (OR) was born in the same incubator as computer science, and it has spawned many new disciplines, such as systems engineering, health care management, and transportation science. Although people from many disciplines routinely use OR methods, many scientific researchers, engineers, and others do not understand basic OR tools and how they can help them. Disciplines ranging from finance to bioengineering are the beneficiaries of what we do — we take an interdisciplinary approach to problem-solving. Our strengths are modeling, analysis, and algorithm design. We provide a quanti- tive foundation for a broad spectrum of problems, from economics to medicine, from environmental control to sports, from e-commerce to computational - ometry. We are both producers and consumers because the mainstream of OR is in the interfaces. As part of this effort to recognize and extend OR roots in future probl- solving, we organized a set of tutorials designed for people who heard of the topic and want to decide whether to learn it. The 90 minutes was spent addre- ing the questions: What is this about, in a nutshell? Why is it important? Where can I learn more? In total, we had 14 tutorials, and eight of them are published here.

Around the world, liberalization and privatization in the electricity industry have lead to increased competition among utilities. At the same time, utilities are now exposed more than ever to risk and uncertainties, which they cannot pass on to their customers through price increases as in a regulated environment. Especially electricity-generating companies have to face volatile wholesale prices, fuel price uncertainty, limited long-term hedging possibilities and huge, to a large extent, sunk investments. In this context, Uncertainty in the Electric Power Industry: Methods and Models for Decision Support aims at an integrative view on the decision problems that power companies have to tackle. It systematically examines the uncertainties power companies are facing and develops models to describe them - including an innovative approach combining fundamental and finance models for price modeling. The optimization of generation and trading portfolios under uncertainty is discussed with particular focus on CHP and is linked to risk management. Here the concept of integral earnings at risk is developed to provide a theoretically sound combination of value at risk and profit at risk approaches, adapted to real market structures and market liquidity. Also methods for supporting long-term investment decisions are presented: technology assessment based on experience curves and operation simulation for fuel cells and a real options approach with endogenous electricity prices.

Real-Time Management of Resource Allocation Systems focuses on the problem of managing the resource allocation taking place within the operational context of many contemporary technological applications, including flexibly automated production systems, automated railway and/or monorail transportation systems, electronic workflow management systems, and business transaction supporting systems. A distinct trait of all these applications is that they limit the role of the human element to remote high-level supervision, while placing the burden of the real-time monitoring and coordination of the ongoing activity upon a computerized control system. Hence, any applicable control paradigm must address not only the issues of throughput maximization, work-in-process inventory reduction, and delay and cost minimization, that have been the typical concerns for past studies on resource allocation, but it must also guarantee the operational correctness and the behavioral consistency of the underlying automated system. The resulting problem is rather novel for the developers of these systems, since, in the past, many of its facets were left to the jurisdiction of the present human intelligence. It is also complex, due to the high levels of choice – otherwise known as flexibility – inherent in the operation of these environments.

For over a decade, there has been an increasing interest in the use of supply chain methods to improve performance across the entire business enterprise. Numerous industries have recognized the importance of efficient supply chain integration, and, as a result, supply chain management has become a standard part of business practice. The Practice of Supply Chain Management: Where Theory and Application Converge is a must-have volume for users of supply chain management methods, supply chain management researchers, and students in supply chain management. The objective of the book is to provide an overview of this important practice-research cycle, and it is organized into three sections: Core Concepts and Practices; Emerging Supply Chain Practices; and Supply Chain in Action.  The focus of the book is on supply chain practice, but supply chain practice that has been heavily influenced by supply chain research. It is this synergy between research and practice that continues to simulate new directions for research.

The Theory and Practice of Revenue Management is a book that comprehensively covers theory and practice of the entire field, including both quantity and price-based RM, as well as significant coverage of supporting topics such as forecasting and economics. The authors believe such a comprehensive approach is necessary to fully understand the subject. A central objective of the book is to unify the various forms of RM and to link them closely to each other and to the supporting fields of statistics and economics. Nevertheless, the topics and coverage do reflect choices about what is important to understand RM. Hence, the book’s purpose is to provide a comprehensive, accessible synthesis of the state of the art in Revenue Management.

This book grew out of an effort to salvage a potentially useful idea for greatly simplifying traditional quantitative risk assessments of the human health consequences of using antibiotics in food animals. In 2001, the United States FDA’s Center for Veterinary Medicine (CVM) (FDA-CVM, 2001) published a risk assessment model for potential adverse human health consequences of using a certain class of antibiotics, fluoroquinolones, to treat flocks of chickens with fatal respiratory disease caused by infectious bacteria. CVM’s concern was that fluoroquinolones are also used in human medicine, raising the possibility that fluoroquinolone-resistant strains of bacteria selected by use of fluoroquinolones in chickens might infect humans and then prove resistant to treatment with human medicines in the same class of antibiotics, such as ciprofloxacin. As a foundation for its risk assessment model, CVM proposed a dramatically simple approach that skipped many of the steps in traditional risk assessment. The basic idea was to assume that human health risks were directly proportional to some suitably defined exposure metric. In symbols: Risk = K × Exposure, where “Exposure” would be defined in terms of a metric such as total production of chicken contaminated with fluoroquinolone-resistant bacteria that might cause human illnesses, and “Risk” would describe the expected number of cases per year of human illness due to fluoroquinolone-resistant bacterial infections caused by chicken and treated with fluoroquinolones.

It is not an exaggeration that much of what people devote in their hfe re­ solves around optimization in one way or another. On one hand, many decision making problems in real applications naturally result in optimization problems in a form of integer programming. On the other hand, integer programming has been one of the great challenges for the optimization research community for many years, due to its computational difficulties: Exponential growth in its computational complexity with respect to the problem dimension. Since the pioneering work of R. Gomory [80] in the late 1950s, the theoretical and methodological development of integer programming has grown by leaps and bounds, mainly focusing on linear integer programming. The past few years have also witnessed certain promising theoretical and methodological achieve­ ments in nonlinear integer programming. When the first author of this book was working on duality theory for n- convex continuous optimization in the middle of 1990s, Prof. Douglas J. White suggested that he explore an extension of his research results to integer pro­ gramming. The two authors of the book started their collaborative work on integer programming and global optimization in 1997. The more they have investigated in nonlinear integer programming, the more they need to further delve into the subject. Both authors have been greatly enjoying working in this exciting and challenging field.

"Daprima importa sappere di che cosa si tratta." ("The ?rst thing is to know what the talk is about".) Attributed to Vilfredo Pareto. This book results from my continuous and deep interest in multiple criteria decision making (MCDM). Eleven years ago I wrote in my p- vious monograph: "This work results from my interest in the ?eld of vector optimization. I stumbled ?rst upon this subject in 1982 [ ... ]. I was attracted then by a gap between vector optimization used to serve as a formal model for multiple objective decision problems and the - cision problems themselves, the gap nonexistent in scalar optimization. Roughly speaking, vector optimization provides methods for ranking - cisions according to a partial order whereas decision making requires a linear ordering of decisions." This declaration is still valid and nothing needs to be changed. To be more speci?c, this book is a fruit of my dissatisfaction with the current state-of-the-art of MCDM. MCDM is a branch of science, whose declared ultimate goal is to provide practical tools. However, we cannot say, and this is regrettable, that all present MCDM methods and algorithms are in popular use by those who make complex decisions and for that purpose are in need of methodological or computational support.

Formal decision and evaluation models are sets of explicit and well-defined rules to collect, assess, and process information in order to be able to make recommendations in decision and/or evaluation processes. They are so widespread that almost no one can pretend not to have used or suffered the consequences of one of them. Our earlier companion volume, Evaluation and Decision Models, heavily criticised formal models but also argued that they could be useful. On the other hand, Evaluation and Decision Models with Multiple Criteria is a guide aimed at helping the analyst to choose a model and use it consistently. We propose a sound analysis of techniques and our presentation can be extended to most decision and evaluation models as a "decision aiding methodology".This volume is intended for the enlightened practitioner, for anyone who uses decision or evaluation models---for research or for applications---and is willing to question his practice, to have a deeper understanding of what he does.

Mass Customization: Challenges and Solutions defines the parameters of the emerging business strategy, mass customization. The book will cover the main categories of the area with a systematic examination of the following themes: manufacturing systems and mass customization, supply chain management and mass customization, and information systems and mass customization. In addition to being tightly organized by themes and a carefully selected group of contributors, the Editors will write introductory material to the sections along with introductory and concluding chapters. The result is a book that provides the field with the conceptual framework of mass customization, its tools, its solutions, and real-world examples of successful implementations of the business strategy. An important book for academics working in production operations, production manufacturing, industrial engineering, logistics, supply chain management, and industry practitioners.

The primary reason for producing this book is to demonstrate and commu­ nicate the pervasive nature of Discrete Optimisation. It has applications across a very wide range of activities. Many of the applications are only known to specialists. Our aim is to rectify this. It has long been recognized that ''modelling" is as important, if not more important, a mathematical activity as designing algorithms for solving these discrete optimisation problems. Nevertheless solving the resultant models is also often far from straightforward. Although in recent years it has become viable to solve many large scale discrete optimisation problems some problems remain a challenge, even as advances in mathematical methods, hardware and software technology are constantly pushing the frontiers forward. The subject brings together diverse areas of academic activity as well as di­ verse areas of applications. To date the driving force has been Operational Re­ search and Integer Programming as the major extention of the well-developed subject of Linear Programming. However, the subject also brings results in Computer Science, Graph Theory, Logic and Combinatorics, all of which are reflected in this book. We have divided the chapters in this book into two parts, one dealing with general methods in the modelling of discrete optimisation problems and one with specific applications. The first chapter of this volume, written by Paul Williams, can be regarded as a basic introduction of how to model discrete optimisation problems as Mixed Integer Programmes, and outlines the main methods of solving them.

Handbook of Production Scheduling concentrates on real-world production scheduling in factories and industrial settings. It includes industry case studies that use innovative techniques as well as academic research results that can be used to improve real-world production scheduling. Its purpose is to present scheduling principles, advanced tools, and examples of innovative scheduling systems to persons who could use this information to improve production scheduling in their own organization.The intended audience includes: production and plant managers, industrial engineers, operations research practitioners, advanced undergraduate/ graduate students and faculty studying and doing research in operations research and industrial engineering.

Operations Research began with the mathematical scheduling of a massive project—logistically supplying Europe with military equipment and goods during the WWII. Today project scheduling research continues growing in a variety of its theoretical models, in its magnitude and application. As the world becomes more interrelated and complex, the wider its research is applied to an increasing number of project scheduling problems. Project Scheduling: Surveying the State-of-the-Art surveys the current state-of-the-art in operations research with chapters written by the respective leading experts on each topic. It covers the range of the key models in the field, including deterministic, probabilistic, single- and multi-mode, single- and multi-objective, and a general model on discrete-continuous resources. Recent solution algorithms are systematical examined. The book summarize sthe current developments and theoretical achievements in the field, including project uncertainty and grid resource management.

A classical queueing model consists of three parts - arrival process, service process, and queue discipline. However, a vacation queueing model has an additional part - the vacation process which is governed by a vacation policy - that can be characterized by three aspects: 1) vacation start-up rule; 2) vacation termination rule, and 3) vacation duration distribution. Hence, vacation queueing models are an extension of classical queueing theory.Vacation Queueing Models: Theory and Applications discusses systematically and in detail the many variations of vacation policy. By allowing servers to take vacations makes the queueing models more realistic and flexible in studying real-world waiting line systems. Integrated in the book's discussion are a variety of typical vacation model applications that include call centers with multi-task employees, customized manufacturing, telecommunication networks, maintenance activities, etc. Finally, contents are presented in a "theorem and proof" format and it is invaluable reading for operations researchers, applied mathematicians, statisticians; industrial, computer, electrical and electronics, and communication engineers; computer, management scientists; and graduate students in the above disciplines.

This edited volume contains 16 research articles. It presents recent and pressing issues in stochastic processes, control theory, differential games, optimization, and their applications in finance, manufacturing, queueing networks, and climate control. One of the salient features is that the book is highly multi-disciplinary. The book is dedicated to Professor Suresh Sethi on the occasion of his 60th birthday, in view of his distinguished career.

From the large overall scope of planning and management of technology, Technology Portfolio Planning and Management: Practical Concepts and Tools will focus specifically on the concepts and tools for the planning and management of an investment portfolio by a government or a business organization for either the development or the application of technologies. A portfolio is a coordinated combination of technologies that achieve a common objective for the decision-maker. Thus, the book will focus on those concepts and tools for selecting and modifying a combination of technologies that will be either developed by a technology supply organization, such as a national laboratory or a corporate research center, or adopted by a technology application organization, such as a government administrative office or a corporation management department, to either advance public goals (space exploration or disease eradication) or enhance corporate strategies (improving productivity or increasing competitiveness).

Principles of Mathematics in Operations Research is a comprehensive survey of the mathematical concepts and principles of industrial mathematics. Its purpose is to provide students and professionals with an understanding of the fundamental mathematical principles used in Industrial Mathematics/OR in modeling problems and application solutions.All the concepts presented in each chapter have undergone the learning scrutiny of the author and his students. The conceptual relationships within the chapter material have been developed in the classroom experience working with the students' level of understanding. The illustrative material throughout the book (i.e., worked-out problems and examples of the mathematical principles) was refined for student comprehension as the manuscript developed through its iterations, and the chapter exercises are refined from the previous year's exercises. In sum, the author has carefully developed a pedagogically strong survey textbook of OR and Industrial Mathematics.

xv Global Supply Chain: General Strategies and Framework There are five chapters on the general strategies and framework. In "On the Globalization of Operations and Supply Chain Strategies - A Conceptual Framework and its Applications," Panos Kouvelis and Julie Niederhoff de­ scribe the forces that shape globalization and a framework to develop strate­ gies. The authors illustrate how the framework can be used with the case of Acrilan, an acrylic fiber manufacturer. As the emerging economies begin to mature, and the supply and demand points in a supply chain begin to shift, companies need to re-optimize the de­ sign of their global supply chain, so as to make the best use of their global resources. This is the subject of "Globalization and Emerging Markets: The Challenge of Continuous Global Network Optimization," by Peter Koudal and Douglas A. Engel. Many emerging economies have lower direct labor costs, and are so attrac­ tive offshoring locations. But there could be many hidden costs. The decision to offshore has to be based on a sound comprehensive analysis of the total landed costs, the tradeoffs of associated risks, and the business strategies of the company. David Pyke provides us with the approach to tackle this deci­ sion, and share with us his personal experience in helping companies to make such decisions, in his chapter titled "Shanghai or Charlotte? The Decision to Outsource to China and Other Low Cost Countries.

Intense global competition in manufacturing has compelled manufacturers to incorporate repetitive processing and automation for improving productivity. Modern manufacturing systems use robotic cells – a particular type of computer-controlled system in cellular manufacturing. Throughput optimization in robotic cells provides practitioners, researchers, and students with up-to-date algorithmic results on sequencing of robot moves and scheduling of parts in robotic cells. It brings together the structural results developed over the last 25 years for the various realistic models of robotic cells. After describing industrial applications of robotic cells and presenting fundamental results about cyclic production, several advanced features, such as dual-grippers, parallel machines, multi-part-type production, and multiple robots, are treated. Important open problems in the area are also identified.This book is an excellent text for use in a graduate course or a research seminar on robotic cells.

Network Science, Nonlinear Science and Infrastructure Systems has been written by leading scholars in these areas. Its express purpose is to develop common theoretical underpinnings to better solve modern infrastructural problems. It is felt by many who work in these fields that many modern communication problems, ranging from transportation networks to telecommunications, Internet, supply chains, etc., are fundamentally infrastructure problems. Moreover, these infrastructure problems would benefit greatly from a confluence of theoretical and methodological work done with the areas of Network Science, Dynamical Systems and Nonlinear Science. This book is dedicated to the formulation of infrastructural tools that will better solve these types of infrastructural problems.

A number of methodologies have been employed to provide decision making solutions to a whole assortment of financial problems in today's globalized markets. Hidden Markov Models in Finance by Mamon and Elliott will be the first systematic application of these methods to some special kinds of financial problems; namely, pricing options and variance swaps, valuation of life insurance policies, interest rate theory, credit risk modeling, risk management, analysis of future demand and inventory level, testing foreign exchange rate hypothesis, and early warning systems for currency crises. This book provides researchers and practitioners with analyses that allow them to sort through the random "noise" of financial markets (i.e., turbulence, volatility, emotion, chaotic events, etc.) and analyze the fundamental components of economic markets. Hence, Hidden Markov Models in Finance provides decision makers with a clear, accurate picture of core financial components by filtering out the random noise in financial markets.

Network ?ow optimization problems may arise in a wide variety of important ?elds, such as transportation, telecommunication, computer networking, ?nancial planning, logistics and supply chain management, energy systems, etc. Signi?cant and elegant results have been achieved onthetheory,algorithms,andapplications,ofnetwork?owoptimization in the past few decades; See, for example, the seminal books written by Ahuja, Magnanti and Orlin (1993), Bazaraa, Jarvis and Sherali (1990), Bertsekas (1998), Ford and Fulkerson (1962), Gupta (1985), Iri (1969), Jensen and Barnes (1980), Lawler (1976), and Minieka (1978). Most network optimization problems that have been studied up to date are, however, static in nature, in the sense that it is assumed that it takes zero time to traverse any arc in a network and that all attributes of the network are constant without change at any time. Networks in the real world are, nevertheless, time-varying in essence, in which any ?ow must take a certain amount of time to traverse an arc and the network structure and parameters (such as arc and node capacities) may change over time. In such a problem, how to plan and control the transmission of ?ow becomes very important, since waiting at a node, or travelling along a particular arc with di?erent speed, may allow one to catch the best timing along his path, and therefore achieve his overall objective, such as a minimum overall cost or a minimum travel time from the origin to the destination.

PROCESS OPTIMIZATION: A Statistical Approach is a textbook for a course in experimental optimization techniques for industrial production processes and other "noisy" systems where the main emphasis is process optimization. The book can also be used as a reference text by Industrial, Quality and Process Engineers and Applied Statisticians working in industry, in particular, in semiconductor/electronics manufacturing and in biotech manufacturing industries.

Inventory management is concerned with matching supply with demand and a central problem in Operations Management. The problem is to find the amount to be produced or purchased in order to maximize the total expected profit or minimize the total expected cost. Over the past two decades, several variations of the formula appeared, mostly in trade journals written by and for inventory managers. A critical assumption in the inventory literature is that the demands in different periods are independent and identically distributed. However, in real life, demands may depend on environmental considerations or the events in the world such as the weather, the state of economy, etc. Moreover, these events are represented by stochastic processes - exogenous or controlled.In Markovian Demand Inventory Models, the authors are concerned with inventory models where these world events are modeled by Markov processes. Their research on Markovian demand inventory models was carried out over a period of ten years beginning in the early nineties.

Just-in-Time Scheduling: Models and Algorithms for Real Time Operating Systems is the first expository treatment surveying the theoretical work on computer systems models and algorithms utilized in just-in-time scheduling. With the impact of globalization and supply chains on manufacturing, there are immense amounts of material flowing through supply chains at any given time worldwide. Therefore the scheduling of all the stages of material arriving at different geographical points at precise times is a highly significant problem. Moreover, the theoretical work in this area has larger ramifications for operational scheduling in many application areas. The scheduling models and algorithms presented and illustrated in the book will be done so in the context of extensive use of computer systems in a "real time context." The book will survey and synthesize all the research in this topical area.

The foreign exchange market is one of the most complex dynamic markets with the characteristics of high volatility, nonlinearity and irregularity. Since the Bretton Woods System collapsed in 1970s, the fluctuations in the foreign exchange market are more volatile than ever. Furthermore, some important factors, such as economic growth, trade development, interest rates and inflation rates, have significant impacts on the exchange rate fluctuation. Meantime, these characteristics also make it extremely difficult to predict foreign exchange rates. Therefore, exchange rates forecasting has become a very important and challenge research issue for both academic and ind- trial communities. In this monograph, the authors try to apply artificial neural networks (ANNs) to exchange rates forecasting. Selection of the ANN approach for - change rates forecasting is because of ANNs’ unique features and powerful pattern recognition capability. Unlike most of the traditional model-based forecasting techniques, ANNs are a class of data-driven, self-adaptive, and nonlinear methods that do not require specific assumptions on the und- lying data generating process. These features are particularly appealing for practical forecasting situations where data are abundant or easily available, even though the theoretical model or the underlying relationship is - known. Furthermore, ANNs have been successfully applied to a wide range of forecasting problems in almost all areas of business, industry and engineering. In addition, ANNs have been proved to be a universal fu- tional approximator that can capture any type of complex relationships.

The objective of ONLINE OPTIMIZATION is to provide a systematic survey of the methodology. From the methodological survey, the book then covers a variety of applications of online optimization methods in the domain of Operations Research and Management Science. These applications include a range of problem types, which include the multiple scheduling complex transportation systems, optimizing financial decision problems in "real time", and complex production problems of all sorts (e.g., whether costs should be reduced or profits should be maximized or scarce resources should be used wisely, etc.). With online optimization the issue of incomplete data is an essential aspect of the scientific challenge. Hence, how well online algorithms can perform and how one can guarantee solution quality—even without knowing all data in advance—are the primary challenges of the online optimization methodology.