Demand-Driven Forecasting

CHARLES W. CHASE JR. is the Chief Industry Consultant in SAS's Manufacturing & Supply Chain Global Practice, where he is the principal architect and strategist for delivering demand planning and forecasting solutions to improve SAS customers' supply chain efficiencies. He has more than twenty-six years of experience in the consumer packaged goods industry, and is an expert in sales forecasting, market response modeling, econometrics, and supply chain management.

• Foreword xiPreface xvAcknowledgments xixAbout the Author xxChapter 1 Demystifying Forecasting: Myths versus Reality 1Data Collection, Storage, and Processing Reality 5Art-of-Forecasting Myth 8End-Cap Display Dilemma 10Reality of Judgmental Overrides 11Oven Cleaner Connection 13More Is Not Necessarily Better 16Reality of Unconstrained Forecasts, Constrained Forecasts, and Plans 17Northeast Regional Sales Composite Forecast 21Hold-and-Roll Myth 22The Plan that Was Not Good Enough 23Package to Order versus Make to Order 25“Do You Want Fries with That?” 26Summary 28Notes 28Chapter 2 What Is Demand-Driven Forecasting? 31Transitioning from Traditional Demand Forecasting 33What’s Wrong with The Demand-Generation Picture? 34Fundamental Flaw with Traditional Demand Generation 37Relying Solely on a Supply-Driven Strategy Is Not the Solution 39What Is Demand-Driven Forecasting? 40What Is Demand Sensing and Shaping? 41Changing the Demand Management Process Is Essential 57Communication Is Key 65Measuring Demand Management Success 67Benefits of a Demand-Driven Forecasting Process 68Key Steps to Improve the Demand Management Process 70Why Haven’t Companies Embraced the Concept of Demand-Driven? 71Summary 74Notes 75Chapter 3 Overview of Forecasting Methods 77Underlying Methodology 79Different Categories of Methods 83How Predictable Is the Future? 88Some Causes of Forecast Error 91Segmenting Your Products to Choose the Appropriate Forecasting Method 94Summary 101Note 101Chapter 4 Measuring Forecast Performance 103“We Overachieved Our Forecast, So Let’s Party!” 105Purposes for Measuring Forecasting Performance 106Standard Statistical Error Terms 107Specific Measures of Forecast Error 111Out-of-Sample Measurement 115Forecast Value Added 118Summary 122Notes 123Chapter 5 Quantitative Forecasting Methods Using Time Series Data 125Understanding the Model-Fitting Process 127Introduction to Quantitative Time Series Methods 130Quantitative Time Series Methods 135Moving Averaging 136Exponential Smoothing 142Single Exponential Smoothing 143Holt’s Two-Parameter Method 147Holt’s-Winters’ Method 149Winters’ Additive Seasonality 151Summary 156Notes 158Chapter 6 Regression Analysis 159Regression Methods 160Simple Regression 160Correlation Coefficient 163Coefficient of Determination 165Multiple Regression 166Data Visualization Using Scatter Plots and Line Graphs 170Correlation Matrix 173Multicollinearity 175Analysis of Variance 178F-test 178Adjusted R 2 180Parameter Coefficients 181t-test 184P-values 185Variance Inflation Factor 186Durbin-Watson Statistic 187Intervention Variables (or Dummy Variables) 191Regression Model Results 197Key Activities in Building a Multiple Regression Model 199Cautions about Regression Models 201Summary 201Notes 202Chapter 7 ARIMA Models 203Phase 1: Identifying the Tentative Model 204Phase 2: Estimating and Diagnosing the Model Parameter Coefficients 213Phase 3: Creating a Forecast 216Seasonal ARIMA Models 216Box-Jenkins Overview 225Extending ARIMA Models to Include Explanatory Variables 226Transfer Functions 229Numerators and Denominators 229Rational Transfer Functions 230ARIMA Model Results 234Summary 235Notes 237Chapter 8 Weighted Combined Forecasting Methods 239What Is Weighted Combined Forecasting? 242Developing a Variance Weighted Combined Forecast 245Guidelines for the Use of Weighted Combined Forecasts 248Summary 250Notes 251Chapter 9 Sensing, Shaping, and Linking Demand to Supply: A Case Study Using MTCA 253Linking Demand to Supply Using Multi-Tiered Causal Analysis 256Case Study: The Carbonated Soft Drink Story 259Summary 276Appendix 9A Consumer Packaged Goods Terminology 277Appendix 9B Adstock Transformations for Advertising GRP/TRPs 279Notes 282Chapter 10 New Product Forecasting: Using Structured Judgment 283Differences between Evolutionary and Revolutionary New Products 284General Feeling about New Product Forecasting 286New Product Forecasting Overview 288What Is a Candidate Product? 292New Product Forecasting Process 293Structured Judgment Analysis 294Structured Process Steps 296Statistical Filter Step 303Model Step 305Forecast Step 308Summary 313Notes 316Chapter 11 Strategic Value Assessment: Assessing the Readiness of Your Demand Forecasting Process 317Strategic Value Assessment Framework 319Strategic Value Assessment Process 321SVA Case Study: XYZ Company 323Summary 351Suggested Reading 352Notes 352Index 355