Modeling for Insight E-Book

Stephen G. Powell

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 621

Veröffentlichungsjahr: 2011

Beschreibung

Praise for Modeling for Insight "Most books on modeling are either too theoretical or too focused on the mechanics of programming. Powell and Batt's emphasis on using simple spreadsheet models to gain business insight (which is, after all, the name of the game) is what makes this book stand head and shoulders above the rest. This clear and practical book deserves a place on the shelf of every business analyst." --Jonathan Koomey, PhD, Lawrence Berkeley National Laboratory and Stanford University, author of Turning Numbers into Knowledge: Mastering the Art of Problem Solving Most business analysts are familiar with using spreadsheets to organize data and build routine models. However, analysts often struggle when faced with examining new and ill-structured problems. Modeling for Insight is a one-of-a-kind guide to building effective spreadsheet models and using them to generate insights. With its hands-on approach, this book provides readers with an effective modeling process and specific modeling tools to become a master modeler. The authors provide a structured approach to problem-solving using four main steps: frame the problem, diagram the problem, build a model, and generate insights. Extensive examples, graduated in difficulty, help readers to internalize this modeling process, while also demonstrating the application of important modeling tools, including: * Influence diagrams * Spreadsheet engineering * Parameterization * Sensitivity analysis * Strategy analysis * Iterative modeling The real-world examples found in the book are drawn from a wide range of fields such as financial planning, insurance, pharmaceuticals, advertising, and manufacturing. Each chapter concludes with a discussion on how to use the insights drawn from these models to create an effective business presentation. Microsoft Office Excel and PowerPoint are used throughout the book, along with the add-ins Premium Solver, Crystal Ball, and Sensitivity Toolkit. Detailed appendices guide readers through the use of these software packages, and the spreadsheet models discussed in the book are available to download via the book's related Web site. Modeling for Insight is an ideal book for courses in engineering, operations research, and management science at the upper-undergraduate and graduate levels. It is also a valuable resource for consultants and business analysts who often use spreadsheets to better understand complex problems.

Leseprobe

CONTENTS

Preface

Using This Book

Acknowledgments

Acknowledgments for Cases

About the Authors

Part I

1: Introduction

1.0 Models and Modeling

1.1 Well-Structured versus Ill-Structured Problems

1.2 Modeling versus Problem Solving

1.3 Modeling for Insight

1.4 Novice Modelers and Expert Modelers

1.5 Craft Skills in Modeling

1.6 A Structured Modeling Process

1.7 Modeling Tools

1.8 Summary

2: Foundations of Modeling for Insight

2.0 Introduction

2.1 The Modeling Process

2.2 Tools for Modeling

2.3 Presentation Skills

2.4 Summary

3: Spreadsheet Engineering

3.0 Why Use Spreadsheets?

3.1 Spreadsheet Engineering

3.2 Summary

Part II

4: A First Example—The Red Cross Problem

4.0 Introduction

4.1 The Red Cross Problem

4.2 Bringing Blood Quality into the Analysis

4.3 Improving and Iterating

4.4 Summary

5: Retirement Planning Problem

5.0 Introduction

5.1 Retirement Planning (A)

5.2 Retirement Planning (B)

5.3 Retirement Planning (C)

5.4 Presentation of Results

5.5 Summary

6: Technology Option

6.0 Introduction

6.1 Technology Option (A)

6.2 Technology Option (B)

6.3 Additional Refinements

6.4 Presentation of Results

6.5 Summary

Part III

7: MediDevice

7.0 Introduction

7.1 MediDevice Case (A)

7.2 Revising the Model

7.3 MediDevice Case (B)

7.4 Presentation of Results

7.5 Summary

8: Draft Commercials

8.0 Introduction

8.1 Draft Commercials Case

8.2 Frame the Problem

8.3 Diagram the Problem

8.4 Ml Model and Analysis

8.5 M2 Model and Analysis

8.6 M3 Model and Analysis

8.7 M4 Model and Analysis

8.8 Presentation of Results

8.9 Summary

9: New England College Skiway

9.0 Introduction

9.1 New England College Skiway Case

9.2 Frame the Problem

9.3 Diagram the Problem

9.4 Ml Model and Analysis

9.5 Analyzing Mountain Capacity

9.6 M2 Model and Analysis

9.7 M3 Model and Analysis

9.8 Considering Uncertainty

9.9 Presentation of Results

9.10 Summary

10: National Leasing, Inc.

10.0 Introduction

10.1 National Leasing Case

10.2 Frame the Problem

10.3 Diagram the Problem

10.4 Ml Model and Analysis

10.5 M2 Model and Analysis

10.6 M3 Model and Analysis

10.7 M4 Model and Analysis

10.8 Presentation of Results

10.9 Summary

11: Pharma X and Pharma Y

11.0 Introduction

11.1 The Pharma X and Pharma Y Case

11.2 Frame the Problem

11.3 Diagram the Problem

11.4 Expected Value or Simulation?

11.5 Ml Model and Analysis

11.6 M2 Model and Analysis

11.7 M3 Model and Analysis

11.8 Presentation of Results

11.9 Summary

12: Invivo Diagnostics, Inc.

12.0 Introduction

12.1 Invivo Diagnostics Case

12.2 Frame and Diagram the Problem

12.3 Ml Model and Analysis

12.4 M2 Model and Analysis

12.5 M3 Model and Analysis

12.6 M4 Model and Analysis

12.7 Presentation ol Results

12.8 Summary

Appendix A: Guide to Solver®

Appendix B: Guide to Crystal Ball®

Appendix C: Guide to the Sensitivity Toolkit

Index

Copyright © 2008 by John Wiley & Sons, Inc. All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New JerseyPublished simultaneously in Canada

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., Ill River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data

Powell, Stephen G.Modeling for insight : a master class for business analysts /Stephen G. Powell, Robert J. Batt.p. cm.Includes bibliographical references and index.ISBN 978-0-470-17555-2 (cloth)1. Decision making. 2. Business planning. 3. Business—Computer simulation. 4. Electronic spreadsheets. I. Batt, Robert J. II. Title.HD30.23.P69 2008658.4’0352—dc22

2008009579

For Nick and Annika

PREFACE

STEPHEN G. POWELL

Origins of the Book

This book has its origins in a class I took in the mid-1970s from Dick Smallwood and Pete Morris in Stanford University’s Department of Engineering-Economic Systems. Smallwood and Morris taught a course they called “The Art of Modeling,” in which they taught modeling as a craft. It was an eye-opening experience for me. All my other courses were devoted to formal mathematical techniques, and although there was a certain abstract beauty to the quantitative methods I was learning, it was hard to see how I could apply them to practical problems. Modeling as Smallwood and Morris taught it could be applied anywhere. Moreover, it was a creative, sometimes playful, enterprise. It was the most challenging course I had ever taken, and it struck a deep chord.

Some years later, as a professor at the Tuck School teaching management science to MBAs, I created my own version of the course. I borrowed the “Art of Modeling” title, although it was unpopular with some of my colleagues, who thought it sounded more like a New Age seminar than a serious course for a business school. I tried to keep the mathematical level modest and added an emphasis on presentation of quantitative insights. Even so, it was just as challenging for Tuck MBAs as the Stanford course had been for engineering PhD students. Nonetheless, many of the students loved it, which was more than sufficient reason to persist with it.

Smallwood and Morris taught their course in the style of an art studio. That is, there was almost no modeling theory to learn, just repeated practical exercises. We worked for a week, solo or in pairs, on each modeling challenge. At the end of the week, we handed in a paper describing our work and our recommendations to the client. Later we would listen as Smallwood and Morris described their approach to the problem, which was always far more clever and insightful than our own.

Each week in this course we were forced to make the entire modeling journey from developing a problem statement, through modeling and analysis, and concluding with recommendations to the client. The problems Smallwood and Morris posed for us were not the kind of homework problems encountered in other courses, which always had a well-defined solution and only one approach (that was sure to be whatever mathematical topic we were studying that week). These were ill-defined problems, typical of consulting practice, with no right answer. And where homework problems ended with the numerical solution, the hard work of interpretation and presentation in these problems began with the numerical solution.

In the early years of my Tuck modeling course, I followed the same principles (described in Powell, 1998 ). I told my students that modeling was an art and a craft, and therefore, I could not teach them how to do it, but I could coach them to develop their modeling skills by giving them appropriate problems and by helping them during the modeling process. To this end, I required them to come to my office halfway through the week to show me their work in progress. This meeting gave me an invaluable opportunity to observe them at work, to learn about the challenges they faced in modeling the problem of the week, and of course, to assist them over the rough spots. One skill I developed during these appointments was to not answer when they asked if they were “on the right track” but instead to ask what their goals were in the problem at hand. Then, instead of showing them how I had approached the problem, I tried to help them achieve their goals. These mid-week interviews were more like tutoring sessions because, with just one or two students to deal with, I could tailor my interaction precisely to what they needed. I could further challenge students who had already made great progress or provide detailed guidance to those who were struggling.

Through this process, I began to see how well trained our students were to believe that every problem had one right answer (known only to the professor) and that their task was to find that answer. I espoused a radically different philosophy in the course — that is, the problems we worked on had many credible approaches and the students were free to pick their own approach as long as it led to sound insights. It took some students weeks to finally accept that they could set their own goals and be creative in achieving those goals. I even went so far as to avoid discussing my own solution in class, so as to further discourage the notion that the professor has the one right answer. This way the class saw many different approaches to a problem, and sometimes different recommendations, all developed by students and each as credible as the next.

Modeling Heuristics

Through this experience of coaching students, I began to develop some rules of thumb for good modeling. These were pithy statements of techniques I found myself using over and over, and I tried to develop a shorthand way of stating them to help students remember. I eventually identified six tricks of the trade or heuristics and wrote a short article in Interfaces (Powell, 1995) to describe them. They are as follows:

1. Decomposition: Divide and Conquer

2. Prototyping: Get Something Working

3. Sketch a Graph: Visualize

4. Parameterization: Call It Alpha

5. Separate Idea Generation from Evaluation: Quiet the Critic

6. Model the Data: Be Skeptical

These heuristics were my first attempt to reduce certain aspects of the modeling craft to principles in order to eliminate some of the mystery of modeling. However, at that time, I had no idea how to use these heuristics in teaching, except to repeat them endlessly and hope students somehow learned to translate them into effective modeling behaviors.

Research on Modelers and Modeling

In the mid-1990s, I came under the influence of Chris Jernstedt, a learning theorist at Dartmouth College. Jernstedt opened my eyes to the idea that if we want our students to learn a skill, we first must understand how novices acquire that skill. This understanding requires study of and reflection on how novices, including both strong and weak students, proceed step-by-step from an initial state of incompetence to a state of competence. Although this principle may not seem revolutionary, it shocked me into admitting how little I knew about how novices learn modeling skills. I had been convinced for so many years that modeling was an art acquired only through practice that I had not really given much thought to breaking it down into learnable stages.

A great deal of research had been done by psychologists on the differences between novices and experts in such skills as swinging a golf club. Some research had been done specifically on novice and expert problem solving, such as Chi’s work on solving physics problems (Chi et al., 1981 ). No research had been done on understanding how expert modelers approach their task until Tom Willemain took on the challenge in the early 1990s. He interviewed a small number of recognized modeling experts and taped their conversations while they worked on a modeling problem for an hour. Willemain’s papers on this research (Willemain, 1994 , 1995 ) revealed how experts shift their attention frequently from one aspect of problem solving to another and how little time they actually spend building a model compared with the time they spend evaluating whether their work meets the needs of the client.

Several years ago, I joined forces with Tom to extend his research on expert modelers to novice modelers. We presented some of the same problems he had used with experts to Tuck student volunteers, some of whom had taken my “Modeling” course. Despite my many years teaching modeling to MBAs, I was somewhat shocked to read transcripts of these sessions. I had no idea how much students floundered when they first encountered an ill-structured problem. This was almost as true of students who had just completed my course as it was of those who had not taken it. We identified the following five behaviors in our subjects that impeded their progress toward a functioning model (Powell and Willemain, 2007 ; Willemain and Powell, 2007 ):

• Over-reliance on given numerical data

• Taking shortcuts to an answer

• Insufficient use of abstract variables and relationships

• Ineffective self-regulation

• Overuse of brainstorming relative to structured problem solving

In many ways, this book is my response to our findings in this research.

Modeling for Insight

About the time I began thinking about how novice modelers learn, I was fortunate to have Jay Goldman, then of Strategic Decisions Group, come to Tuck—first to teach the modeling course on his own and then to coteach it with me. Jay is both a master modeler and a natural teacher, so he was enthusiastic when I told him what I was learning from the theorists, and he was eager to work with me to develop a step-wise approach to modeling. Jay coined the phrase “modeling for insight” to stress that the ultimate goal in modeling ill-structured problems is not to find the “solution” but to generate insights that will help a client understand the problem and take action. Jay also convinced me of the power of influence diagrams, the importance of spreadsheet hygiene (which I have renamed spreadsheet engineering), and the usefulness of strategy analysis. He also introduced several excellent cases, some of which appear here. To a large degree, this book is a direct descendant of the modeling course we developed together; Jay’s influence on this book is profound.

STEPHEN G. POWELL

Hanover, NHFebruary 2008

ROBERT J. BATT

Steve and I met when I was a first-year MBA student in his Decision Science class at Tuck. I was struck by how Steve refused to allow the course to be simply about how to use Excel. Most of the assignments in the class were business cases that required more than a simple, numeric answer. They required a recommended course of action and an understanding of the business implications of that recommendation. Students were regularly called on to present their findings to the class. These were not presentations of a model but, rather, of the thinking behind the model and the relevance of the results it produced. Steve was pushing us to model for insight. Later that year, I also took Steve’s “Applications of Simulation” course, which similarly was not just about methodology and techniques but, rather, about how to use simulation to gain insight into the problem at hand.

The next year, Steve asked me to help him redesign the simulation course. We decided that in order to allow the students more time to work on bigger, more realistic cases, we would cancel half the lectures. Instead of class time we required office hours where he and I met with the students to review their modeling progress and to help coach them along the way. The days the class did meet, class time was primarily devoted to students presenting their work. While working together on this course, I saw the power of the modeling methods Steve had developed.

After graduating from Tuck, I joined the school as a research fellow. Steve had already drafted the first few chapters of this book, and he enlisted my help in reviewing them. Together we refined and improved the conceptualization and delivery of the core principles of modeling for insight. We went on to share the load of writing and editing the rest of the book to such an extent that Steve was generous enough to allow me to be a full coauthor of the book. Without exception, the roots of the ideas presented in this book are Steve’s. It has been my great pleasure to come alongside and help give voice and expression to these ideas.

ROBERT J. BATT

Hanover, NHFebruary 2008

USING THIS BOOK

ORGANIZATION OF THE BOOK

This book is organized into three parts. Part I consists of Chapters 1 – 3 . Chapter 1 introduces the ideas of models and modeling, describes the class of problem this book addresses, and explains what “modeling for insight” means. Chapter 2 lays out the modeling and presentation methods on which the book rests. Chapter 3 describes the basics of spreadsheet engineering: how to design, build, test, and analyze a spreadsheet model with efficiency and effectiveness. If you are an experienced spreadsheet user, some of these ideas may be familiar to you, but you are sure to find something new here. In particular, the sensitivity analysis tools we discuss in this chapter are vital throughout the book and are likely to be new to many readers. We recommend that you skim familiar sections of this chapter but study new concepts carefully.

In Part II, which consists of Chapters 4 – 6 , we take you through a series of increasingly complex modeling cases and show how the procedures and tools of modeling for insight work. Chapter 4 presents a first example, worked out as we would actually carry out a modeling assignment. This chapter demonstrates the type of problem that modeling for insight was developed to deal with and illustrates the kinds of results you can develop with these methods. If you want to know whether this book is for you, read Chapter 4 now.

Chapters 5 and 6 address different modeling problems, beginning in each case with the simplest version of the problem. Expert modelers begin by reducing a problem to its essentials. We believe learning how to do this is critical for novices, so we begin each chapter with a radically simplified version of the problem and we show how to analyze this version effectively. Then we introduce a few complexities so that the problem becomes one step more realistic and we modify our model and analysis and develop additional insights. This process is repeated several times until we find ourselves modeling the problem in all its real - world richness. Part II is the heart of the book and is essential for all readers. We recommend you do not attempt the later chapters before you have worked carefully through Part II of the book.

Part III, Chapters 7 – 12 , consists of a series of modeling cases worked out in detail. Each problem is presented in all its complexity so the modeling challenge is as realistic as possible. We offer our own approach to each problem, but we offer you many chances to sketch out your own approach along the way. Our solutions are intended to show one way to approach each problem, not to present the perfect solution. Thus, we share our thoughts about the right approach to take and occasionally present work that leads down a dead end. Although we do not want to burden you with our mistakes, we do want you to understand how competent modelers iterate their way to a good final result through a process of intelligent trial and error.

REQUIRED BACKGROUND

This book is intended primarily for modelers in the business and nonprofit sectors, although the techniques are applicable to engineering and other technical disciplines. Accordingly, most of our examples originate in actual business situations. Some background in business and economics will be helpful, but none of the examples assumes detailed knowledge of any particular business sector or functional area. Where necessary, we explain enough of the background to a problem to make it accessible for general readers.

We assume you have considerable experience with Microsoft Excel ® and basic skills in the following areas:

• Windows operations

• Workbook navigation

• Entering text and data

• Editing cells

• Formatting cells

• Writing formulas

• Using functions (e.g., SUM, IF, NPV)

• Creating graphs

We do not assume you have extensive experience in creating Excel models from scratch, but some prior experience in this area will be highly beneficial. When we use advanced Excel tools, we explain them thoroughly in the text. We make routine use of Solver ® for optimization, Crystal Ball® for simulation, and the Sensitivity Toolkit for sensitivity analysis; short appendices are provided that describe these tools. If you have never used them, we recommend you study the appendices independently before tackling this book.

You do not need knowledge of advanced mathematics to understand this book, but you should be comfortable with basic algebra and probability. Familiarity with simple functions, like the straight line and the power function, as well as basic probability concepts, such as expected value, conditional probability, and distributions is required.

If you need to bolster your background, you will find extensive coverage of many of the basic skills in Management Science: The Art of Modeling with Spreadsheets (Powell and Baker, 2007 ). Chapter 3 on basic Excel skills and Chapters 5 and 6 on spreadsheet engineering and analysis will be helpful, as will Chapters 9 and 10 on optimization (Solver) and Chapter 15 on simulation (Crystal Ball).

SOFTWARE

All models in this book were built in Excel, specifically in Excel 2007 for Windows. We also rely heavily on three Excel add - ins: Premium Solver, Crystal Ball, and the Sensitivity Toolkit. Trial versions of Solver and Crystal Ball can be downloaded from <www.solver.com> and <www.decisioneering.com>, respectively. The Sensitivity Toolkit is available at <>http://mba.tuck.dartmouth.edu/toolkit/>. The spreadsheets referred to in the book are available at <www.modelingforinsight.com> along with additional materials related to modeling.

HOW TO LEARN TO MODEL FOR INSIGHT

No one can become an expert modeler merely by reading a book, even this one. However, you can dramatically improve your modeling skills if you work though this book in a dedicated and organized fashion. Furthermore, the skills you learn here will provide a firm foundation for your efforts to refine your modeling skills in school or on the job.

The key to using this book effectively is to read actively . Instead of reading large chunks of the text with your feet up, read in small chunks and stop frequently to consider what you have read. When you see “To the reader” and the stopsign icon, stop and take some time to do the exercise we suggest. You will internalize what you have learned better if you try it first yourself. You may feel clumsy initially, but try the problems before you read our suggested approaches. Do not be discouraged if you do not make much progress with these problems at first. As you make your way through the book and learn the tools and approaches we develop here, you will become more comfortable and skilled in developing your own creative approach to the problems.

Our final word of advice is to look for opportunities to apply modeling everywhere you go. Some problems in this book originated from articles in newspapers or public radio news programs. Modeling opportunities are everywhere, and you will increase your capacity to model well if you think about how you would model situations as they arise. Here are some examples:

The state legislature is considering raising the cigarette tax. Can you model the impacts on retailers and consumers?A local museum is considering a major capital campaign. Can you model the impact on the museum’s endowment in the future?Chrysler has amassed a multibillion - dollar cash fund, and major investors are demanding it be distributed to the shareholders. Can you use modeling to determine whether Chrysler should declare a large dividend, or save its cash to help maintain R & D spending during the next economic downturn?

Now, on to modeling!

REFERENCES

Chi M, Feltovich P, Glaster R. Categorization and representation of physics problems by experts and novices . Cognitive Science 1981;5:121–152.

Powell S. The teacher’s forum: Six key modeling heuristics . Interfaces 1995;25(4): 114–125.

Powell S. The studio approach to teaching the craft of modeling . Annals of OperationsResearch 1998;82:29–47.

Powell S, Baker K. Management Science: The Art of Modeling with Spreadsheets. 2nd edition. Hoboken, NJ: Wiley, 2007.

Powell S, Willemain T. How novices formulate models. Part I: qualitative insights and implications for teaching . Journal of the Operational Research Society 2007;58 983–995.

Willemain T. Insights on modeling from a dozen experts . Operations Research 1994; 42:213–222.

Willemain T. Model formulation: What experts think about and when. OperationsResearch 1995;43:916-932.

Willemain T, Powell S. How novices formulate models. Part II: a quantitative description of behavior . Journal of the Operational Research Society 2007;58:1271–1283.

ACKNOWLEDGMENTS

This book is not so much our invention as our interpretation of ideas developed by others. We therefore owe more than the usual authorial debts. Some of the most important ones are as follows:

Dick Smallwood and Peter Morris : who were among the first to recognize that mathematical technique is not enough for a modeler and to find a way to teach craft skills.

Mike Magazine, Steve Pollock , and Seth Bonder : all pioneers in teaching modeling.

Mike Rothkopf : who encouraged Steve to write a column for Interfaces on teaching and modeling and thereby helped stimulate additional thought on the topic.

Peter Regan and Jay Goldman : Peter taught “Art of Modeling” when Steve was on sabbatical and developed several new cases. He has since become a valued teaching colleague. Jay taught the course several times and developed new cases and new concepts. His ideas on modeling and presenting have strongly influenced the book.

Ken Baker and Rob Shumsky : Ken has been a colleague at Tuck for many years. Steve and Ken have taught management science together and have developed a Management Science textbook that contains some of the ideas presented here. Rob Shumsky joined Tuck in 2005 and has embraced the spirit of the modeling courses enthusiastically.

Tom Willemain : who did the first work on expert modelers and opened the way for joint research on novices.

Julie Lang and Jason Romeo : who shared their expertise on communication and presentation design.

Beth Golub : an editor at John Wiley & Sons, who encouraged Ken Baker and Steve to write the management science textbook. She has supported our efforts for many years and kept reminding us that the rest of the world would eventually see things our way if we just persevered.

Susanne Steitz - Filler : the editor of this book and an enthusiastic supporter throughout.

Anita Warren : who gave us valuable advice on the design of the book and did an excellent job of copyediting our drafts.

Bob Hansen and Dave Pyke : Associate Deans at the Tuck School who supported our work and made it possible for Bob to devote time to this project.

ACKNOWLEDGMENTSFOR CASES

Chapter 6, Technology Option.Developed by Jay Goldman, Senior Engagement Manager, Strategic DecisionsGroup, and Professor Stephen G. Powell of the Tuck School of BusinessAdministration, Dartmouth College, Hanover, NH.

Chapter 7, MediDevice.Developed by Jay Goldman, Senior Engagement Manager, Strategic DecisionsGroup, and Professor Stephen G. Powell of the Tuck School of BusinessAdministration, Dartmouth College, Hanover, NH.

Chapter 8, Draft Commercials.Developed by Professor Stephen G. Powell of the Tuck School of BusinessAdministration, Dartmouth College, Hanover, NH, and Professor ThomasWillemain of Rensselaer Polytechnic Institute, Troy, NY.

Chapter 9, New England College Skiway.Developed by Professor Stephen G. Powell of the Tuck School of Administration, Dartmouth College, Hanover, NH. Adapted from an earlier case by Dick Smallwood and Peter Morris of Stanford University. The assistance of Don Cutter and Peter Riess of Dartmouth College, and ChristyBieber of the Tuck School, is gratefully acknowledged.

Chapter 10, National Leasing, Inc.Developed by Adjunct Professor Peter Regan and Professor Stephen G. Powell of the Tuck School of Business Administration, Dartmouth College, Hanover,NH, and Jay Goldman, Senior Engagement Manager, Strategic DecisionsGroup.

Chapter 11, Pharma X and Pharma Y.Developed by Adjunct Professor Peter Regan and Professor StephenG. Powell of the Tuck School of Business Administration, Dartmouth College,Hanover, NH.

Chapter 12, Invivo Diagnostics, Inc.Developed by Stephanie Bichet of Andersen Consulting and Professor StephenG. Powell of the Tuck School of Business Administration, Dartmouth College,Hanover, NH.

ABOUT THE AUTHORS

Steve Powell is a professor at the Tuck School of Business at Dartmouth College. His primary research interest lies in modeling production and service processes, but he has also been active in research in energy economics, marketing, and operations. At Tuck he has developed a variety of courses in management science, including the core “Decision Science” course and electives in the “Art of Modeling,” “Business Process Redesign,” and “Applications of Simulation.” He originated the Teacher’s Forum column in Interfaces , and he has written several articles on teaching modeling to practitioners. He was the Academic Director of the annual INFORMS Teaching of Management Science Workshops. In 2001 he was awarded the INFORMS Prize for the Teaching of Operations Research/Management Science Practice. Prof. Powell holds an AB degree from Oberlin College and MS and PhD degrees from Stanford University.

Bob Batt is a Tuck Fellow and researcher at the Tuck School of Business at Dartmouth College. His work is focused on operations management and finance. He has written case studies on Steinway & Sons and Hurricane Katrina. He also serves as a teaching fellow for the core MBA Corporate Finance course and the Corporate Financial Management course. Prior to coming to Tuck, Mr. Batt worked as a manufacturing engineer designing manufacturing and quality control processes. He holds a BA degree from Wheaton College (IL) and BE, MEM, and MBA degrees from Dartmouth College. He graduated from the Tuck School of Business at Dartmouth in 2006 as an Edward Tuck Scholar with High Distinction. Beginning in the fall of 2008, Mr. Batt will be a doctoral student in Operations Management at the Wharton School of the University of Pennsylvania.

PART I

CHAPTER 1

Introduction

1.0 MODELS AND MODELING

A model is a simplified representation of reality. We use models all the time, although often we are not aware of doing so. Any map is a model of some part of the natural or human landscape. A decision to merge two companies is likely to be based on a model of how the combined companies will operate. Predictions that the U.S. Social Security program will go bankrupt are based on models, as are decisions to recommend evacuation in the face of a hurricane. Models, in fact, are a ubiquitous feature of modern life, and everyone is affected by them.

Modeling is the process of building and using models. Some models are built for a single purpose. For example, someone might build a simple spreadsheet model to test the impact of various investment alternatives on their personal tax liabilities. Other models are built by modeling experts to help a group of managers make a one-time decision. An example of this might be a consulting company that builds a model to value a potential acquisition target for a large corporation. Still other models are used repetitively, as part of everyday operations. State legislatures use models routinely, for instance, to forecast the annual budget surplus based on tax receipts collected to date.

Just as there are many different types of models and purposes for modeling, there are also many different types of modelers. Professional modelers often specialize in a particular industry or modeling method. Professional modelers work on airline-crew scheduling problems, marketing media-selection problems, and stock-option valuation problems. Other professionals specialize in particular tools, such as econometrics, optimization, or simulation. Also, millions of modelers build and use models as part of their jobs or in their personal lives but do not think of themselves as modelers. Most spreadsheet users probably fall into this category. We refer to these individuals as End users often are professionals in a field other than modeling. They may be consultants, lawyers, accountants, marketing or financial analysts, plant managers, or hospital administrators. This book is designed to help end users become more like professional modelers, without investing the time to acquire an advanced degree or serve an apprenticeship.