管理决策
2009-9
清华大学出版社
GREGORY P.PRASEACOS
542
Globalization, the economic uncertainties, and the explosive growth of information andcommunication technologies have transformed the business world radically. A constant pres-sure on companies to reduce costs while improving quality, efficiency, service and innova-tion, is on the daily agenda. In this complex and continuously evolving environment, acompany's ability to react to the challenges becomes of vital importance.In this new business reality, managers must have the ability to make decisions quicklyand thoroughly, by taking evidence into account, and evaluating consequences and alterna-tive plans in terms of growth, profitability and acceptable risk. In today's environment, be-ing able to make good and fast decisions in all areas of business, including financial and riskmanagement, marketing strategy and customer understanding, production and logistics, pro-curement and supply chain management, human resources management and development,IT-based process redesign, is becoming strategically important.Management Science is of paramount importance in this effort. It provides decisionmakers with an extensive range of methodologies, skills, models and software tools that arenecessary in order to effectively address a wide range of decision problems. More specifical-ly, Management Science has been very helpful in management and decision making by con-tributing: a） A rational, systematic and robust methodology that can be followed when ad-dressing a decision problem, b） A rich set of models, techniques and tools that can be usedto structure, understand and solve the problem, and c ） A rich interface environment thathelps the decision maker understand a solution, evaluate alternative solutions and come upwith the preferred one. In all of these steps, the decision maker is supported by specializedeasy-to-use software, which often come as additions （add-ins） to popular software like Ex-cel, or are part of corporate Business Intelligence Systems or Decision Support Systems.
Management Science is of paramount importance in this effort. It provides decision makers with an extensive range of methodologies， skills， models and software tools that are necessary in order to effectively address a wide range of decision problems. More specifically， Management Science has been very helpful in management and decision making by contributing： a） A rational， systematic and robust methodology that can be followed when addressing a decision problem， b） A rich set of models， techniques and tools that can be used to structure， understand and solve the problem， and c ） A rich interface environment that helps the decision maker understand a solution， evaluate alternative solutions and come up with the preferred one. In all of these steps， the decision maker is supported by specialized easy-to-use software， which often come as additions （add-ins） to popular software like Excel， or are part of corporate Business Intelligence Systems or Decision Support Systems.
作者：(希腊)GREGORY P.PRASEACOS
PAPTⅠ THE FUNDAMENTALS OF MANAGERIAL DECISION MAKINGCHAFFER 1 INTRODUCTION TO MANAGEMENT SCIENCE1.1 Introduction to Managerial Decision Making1.2 Trends Affecting Decision Making Today1.3 Key Characteristics of Management Science for Decision Making1.4 The Use of Models for Making Decisions1.5 The Use of Software in Decision Making1.6 Applications of Management Science in BusinessReferencesCHAFFER 2 AVOIDING BAD DECISIONS: The Methodology ofDecision Making2.1 Introduction2.2 Decision Making Traps2.3 Decision Making Tips2.4 The Rational Methodology for Decision Making2.5 Identification of the Problem2.6 Analysis of the System2.7 Formulation of the Objectives2.8 Initial System Design2.9 Detailed System Design2.10 Solution Implementation and MonitoringReferencesPAPTⅡ MODELS IN MANAGERIAL DECISION MAKINGCHAPTER 3 LINEAR PROGRAMMING3.1 Introduction3.2 Characteristics of LP Problems3.3 A Maximization Problem3.4 A Trial-and-Error Approach in Solving LP Problems3.5 Graphical Solution of a LP Problem3.6 A Minimization Problem3.7 General Formulation and Assumptions of LP Models3.8 Solving LP ProblemsProblemsReferencesCHAPTER 4 USING SOLVER TO SOLVE LINEAR PROGRAMMINGPROBLEMS4. 1 Introduction4.2 Introducing the Model in Excel4.3 Solving the Problem4.4 Understanding and Analyzing the Solution-SOLVER Reports4.5 Solving Integer Programming Problems with SOLVER4.6 Solving Non-Linear Programming Problems with SOLVER4.7 ConclusionsProblemsReferencesCHAPTER 5 SENSITIVITY ANALYSIS IN LINEARPROGRAMMING5.1 Introduction5.2 An Example5.3 Dual Prices in LP5.4 Reduced Costs in LP5.5 Changes in the Objective Function's Coefficients5.6 Changes in the Right Hand Sides (RHS) of the Constraints5.7 Evaluation of a New Activity5.8 ConclusionsProblemsReferencesCHAPTER 6 INTEGER PROGRAMMING6.1 Introduction6.2 Formulating IP Problems with Binary Variables6.3 An Investment Example6.4 Formulating IP Problems with Fixed Costs and/or Discounts6.5 Solving IP Problems6.6 Heuristic Methods to Solve IP Problems6.7 ConclusionsProblemsReferencesCHAPTER 7 MULTI-CRITERIA DECISION MAKING7. I Introduction7.2 Empirical Methods7.3 Goal Programming7.4 The Analytical Hierarchy Process7.5 Using Expert Choice to Solve Multicriteria ProblemsProblemsReferencesCHAPTER 8 STATISTICAL METHODS IN DECISIONMAKING8.1 Introduction to Forecasting8.2 Key Concepts about Forecasting8.3 The Moving Averages Forecasting Method8.4 Exponential Smoothing Forecasting Method8.5 Other Forecasting Methods8.6 Linear Regression8.7 Multiple Regression8.8 Discriminant Analysis8.9 Using SPSS for Statistical AnalysisProblemsReferencesCHAPTER 9 DECISION ANALYSIS9.1 Introduction9.2 Key Concepts about Decision Analysis9.3 Criteria for Making Decisions under Uncertainty9.4 The Expected Value of Perfect Information9.5 Introduction to Decision Trees9.6 Calculating the Risk Profile a Strategy9.7 Sensitivity Analysis9.8 Using Precision Tree to Solve Decision Analysis ProblemsProblemsReferencesCHAPTER 10 SIMULATION10.1 Introduction10.2 Key Characteristics of Simulation10.3 Implementation of Simulation under Conditions of Uncertainty10.4 Simulation of Queuing Systems10.5 Simulation of an Inventory System10.6 Analysis of Simulation Results10.7 Using Simulation for Risk Management10.8 Using Simulation for Business Process ReengineeringProblemsReferencesPAPTⅢ IMPLEMENTING MANAGEMENT SCIENCE IN PRACTICECHAPTER 11 GETTING TO KNOW YOUR CUSTOMER11.1 Introduction11.2 Determining Customer Satisfaction11.3 Designing New Products11.4 Sales-Advertising Response Analysis11.5 Forecasting Sales of New Products11.6 Identifying Areas of Improvement11.7 Studying Product Positioning11.8 Identifying Market Segments11.9 Identifying Prospect CustomersProblemsReferencesCHAPTER 12 MARKETING AND SALES MANAGEMENT12.1 Introduction12.2 A Product Selection Problem12.3 Design of Sales Network12.4 Selection of Communication Media12.5 Selection of Location12.6 Design of New Product Marketing Strategy12.7 Design of Promotion Strategy12.8 Sales Strategy12.9 Evaluation of Customer Value for CRM Implementation……CHAPTER 13 PRODUCTION AND INVENTORY MANAGEMTCHAPTER 14 NETWORKS AND TRANSPORT PROBLEMSCHAPTER 15 LOGISTICS AND SUPPLY CHAIN MANAGEMENT
插图：When the decision variables can only take integer values, the condition of divisibility of Lin- ear Programming does not hold. For example, in the class of problems where the variables can only take values 0 or 1 （0/1 problems）, such as project selection problems, the decision maker has to choose among specific alternatives, and therefore, the decisions are either yes or no for each alternative no "partial" answers are acceptable. To the degree that there exists an objective, which the decision maker has to achieve, and the contributions of eachchoice to this objective can be calculated, these problems can be formulated as Integer Pro-gramming problems and can be solved using Solver.In order to use Solver in Integer Programming problems, we have to add the constraintsof integrality to the constraints of the model. We implement this through the "Add Con-straint" dialog box, in which we select INT （ for generally integer variables） or BIN （ for bi-nary variables of the type 0 or 1 ）. To input an Integer Programming problem to Excel, wefollow the same steps as in the case of Linear Programming, with the only difference beingthe definition of integrality of variables.Solving Non-Linear Programming Problems with SOLVERSolver can also be used to solve （with a certain approximation） Non-Linear Programmingproblems. This is done through a series of choices that have to do with the way the problemis solved. In order to define these choices, we select at the "Definition of Solver Parame-ters" of the problem （ table 4.8 ）, we left click at "Options" and in the appearing dialog boxwe define （left click again） the following choice:Hypothesis of Linear model
《管理决策:理论与实践》是由清华大学出版社出版的。
