Business Forecasting: The Emerging Role of Artificial Intelligence and Machine Learning

Description

Discover the role of machine learning and artificial intelligence in business forecasting from some of the brightest minds in the field In Business Forecasting: The Emerging Role of Artificial Intelligence and Machine Learning accomplished authors Michael Gilliland, Len Tashman, and Udo Sglavo deliver relevant and timely insights from some of the most important and influential authors in the field of forecasting. You'll learn about the role played by machine learning and AI in the forecasting process and discover brand-new research, case studies, and thoughtful discussions covering an array of practical topics. The book offers multiple perspectives on issues like monitoring forecast performance, forecasting process, communication and accountability for forecasts, and the use of big data in forecasting. You will find: Discussions on deep learning in forecasting, including current trends and challenges Explorations of neural network-based forecasting strategies A treatment of the future of artificial intelligence in business forecasting Analyses of forecasting methods, including modeling, selection, and monitoring In addition to the Foreword by renowned researchers Spyros Makridakis and Fotios Petropoulos, the book also includes 16 "opinion/editorial" Afterwords by a diverse range of top academics, consultants, vendors, and industry practitioners, each providing their own unique vision of the issues, current state, and future direction of business forecasting. Perfect for financial controllers, chief financial officers, business analysts, forecast analysts, and demand planners, Business Forecasting will also earn a place in the libraries of other executives and managers who seek a one-stop resource to help them critically assess and improve their own organization's forecasting efforts. Cover Title Page Copyright Page Contents Foreword Preface State of the Art Forecasting in Social Settings: The State of the Art* I. The Facts A Brief History of Forecasting When Predictions Go Wrong Improving Forecasting Accuracy over Time The Importance of Being Uncertain II. What We Know On Explaining the Past versus Predicting the Future On the (Non)existence of a Best Model On the Performance of Machine Learning III. What We Are Not Sure About On the Prediction of Recessions/Booms/Non-stable Environments On the Performances of Humans versus Models On the Value of Explanatory Variables IV. What We Don’t Know On Thin/Fat Tails and Black Swans On Causality On Luck (and Other Factors) versus Skills V. Conclusions Notes References Chapter 1 Artificial Intelligence and Machine Learning in Forecasting 1.1 Deep Learning for Forecasting* Introduction What Is a Neural Network? How Do We Forecast with Neural Nets? Examples of Neural Forecasting Models References 1.2 Deep Learning for Forecasting: Current Trends and Challenges* Applying Neural Nets as Global Forecasting Models Pros and Cons of Neural Forecasting Current Trends and Challenges DL Software for Forecasting References 1.3 Neural Network–Based Forecasting Strategies* Introduction Neural Network Modeling in SAS Visual Forecasting Modeling Strategies Case Study: Ozone Prediction Case Study: Solar Energy Forecasting Best Practices and Other Tips Conclusion Acknowledgments References 1.4 Will Deep and Machine Learning Solve Our Forecasting Problems?* Introduction The Good and the Bad The Problems What about the M4 Competition? Conclusion References 1.5 Forecasting the Impact of Artificial Intelligence: The Emerging and Long-Term Future* Introduction The 10 Major Emerging Trends Supercomputers, High-Speed Networks, and the Cloud Medicine and Genomics Renewable Energy and Energy Storage AVs and Drones Greater Wealth and More Comfort The “Science Fiction” Type of Technological Developments Conclusions References 1.6 Forecasting the Impact of Artificial Intelligence: Another Voice* Forecasting AI’s Computational Power AI’s Impact on Employment Blockchain, IA, and Forecasting Intelligence Augmentation The Long-Term Future AI for Forecasting Summary 1.7 Smarter Supply Chains through AI* Introduction Supply-Chain Challenges Control-System Theory for Supply-Chain Management Applying AI/ML Lessons Learned References 1.8 Continual Learning: The Next Generation of Artificial Intelligence* Introduction Eliminating Our Own Complexities AutoML Continual Learning Continual Learning Augmentation Conclusion Acknowledgment References 1.9 Assisted Demand Planning Using Machine Learning* The Life of a Demand Planner What Is Forecast Value Added? Using Intelligent Automation to Improve a Demand Planner’s FVA Conclusion References 1.10 Maximizing Forecast Value Add through Machine Learning and Behavioral Economics* Introduction Predicting Forecast Value Add Override Size Forecastability of Demand Override Classification Techniques Analysis Behavioral Economics: How to Influence Forecaster Behavior Summary References 1.11 The M4 Forecasting Competition – Takeaways for the Practitioner* M4 Background M4 Results Takeaways for Forecasting Practitioners Criticism of the M4 Looking Ahead to the M5 References Chapter 2 Big Data in Forecasting 2.1 Is Big Data the Silver Bullet for Supply-Chain Forecasting?* The “Big Data” Bubble Forecasting by Item or Consumer Big Data and Causal Forecasting Conclusion References 2.2 How Big Data Could Challenge Planning Processes across the Supply Chain* Introduction “Big Data” Sources and the Potential They Bring The Challenge for Aggregate and Detailed Planning Conclusions References Chapter 3 Forecasting Methods: Modeling, Selection, and Monitoring 3.1 Know Your Time Series* Data Availability Stationarity Forecastability and Scale Key Takeaways Note References 3.2 A Classification of Business Forecasting Problems* Introduction Dimensions of the Classification Strategic Forecasting Tactical Forecasting Operational Forecasting Publicly Available Data Sets Consequences: People, Skills, Methods, and Software References 3.3 Judgmental Model Selection* Forecasting with Judgment Exploring the Performance of Judgmental Model Selection The Behavioral Experiment Why Model-Build Works Better Implications for Software Final Comments References Commentary: A Surprisingly Useful Role for Judgment References Commentary: Algorithmic Aversion and Judgmental Wisdom References Commentary: Model Selection in Forecasting Software Reference Commentary: Exploit Information from the M4 Competition Reference 3.4 A Judgment on Judgment* Patterns That Aren’t There Stories Trump Numbers The Perils of Imagination and Memory Chained to an Anchor A Pleasant Pension Surprise? Judging Lots of Possibilities References 3.5 Could These Recent Findings Improve Your Judgmental Forecasts?* Surprises Competition Combination Conclusions References 3.6 A Primer on Probabilistic Demand Planning* Moving to a Probabilistic Perspective of the Future The Differences between Statistical and Probabilistic Forecasts Planning with Probabilistic Forecasts 3.7 Benefits and Challenges of Corporate Prediction Markets* Introduction: Corporate Prediction Markets Forecast Accuracy Factors Influencing Accuracy Big Data – A Competitive Approach Concluding Thought References 3.8 Get Your CoV On . . .* Getting Your CoV On . . . Discussion on Demand A Side Discussion on MAPE and WMAPE 3.9 Standard Deviation Is Not the Way to Measure Volatility* Takeout 3.10 Monitoring Forecast Models Using Control Charts* Introduction Background Residual Analysis Methodology Illustrative Examples Summary References 3.11 Forecasting the Future of Retail Forecasting* Introduction Your Next Shopping Trip Digital Technologies and Trends Implications for the Retail Industry Implications for Retail Forecasting Conclusion References Commentary References Chapter 4 Forecasting Performance 4.1 Using Error Analysis to Improve Forecast Performance* Preview and Key Points from the Author Key Concepts Error Analysis Drivers of Forecast Quality Lessons Learned Conclusion References 4.2 Guidelines for Selecting a Forecast Metric* Idiosyncracies about the Measurement of Forecast Error Forecast Error Is a Funny Thing Lies, Damn Lies, and Statistics 4.3 The Quest for a Better Forecast Error Metric: Measuring More Than the Average Error* Introduction Point Forecasts vs. Probabilistic Forecasts Uncertainty and Inventory Percentile Error Metrics Conclusions Appendixes References 4.4 Beware of Standard Prediction Intervals from Causal Models* Introduction: Standard Prediction Intervals for a Regression Model Forecast Sources of Error in the Standard Prediction Interval Sources of Error Not Accounted For A Case Study The Derivation of the Inflation Factors Simulation Conclusions Extensions References Chapter 5 Forecasting Process: Communication, Accountability, and S&OP 5.1 Not Storytellers But Reporters* News and Evidence Not Storytellers But Reporters The Duty of Clarity Reference 5.2 Why Is It So Hard to Hold Anyone Accountable for the Sales Forecast?* Responsibility versus Accountability Are You Forecasting or Sales Planning? Are Responsibilities and Overall Accountability Well Defined? Appropriate Metrics for Sales Planning Process Monitoring Changes to the Game Who’s “On First” in Your Company? The Checklist Conclusion 5.3 Communicating the Forecast: Providing Decision Makers with Insights* Asking Decision Makers What They Need Storyboard Structure Risk Management Conclusion References 5.4 An S&OP Communication Plan: The Final Step in Support of Company Strategy* Introduction Communications to Support Business Strategy S&OP and Strategy Execution The S&OP Communication Plan Summary References 5.5 Communicating Forecasts to the C-Suite: A Six-Step Survival Guide* Six Tips for Explaining the Forecast to Execs Introduction Articulate What the CFO Needs to Believe to Use the Forecast Accountants and Statisticians Think Differently about Data Don’t Talk about Complex Diagnostic Statistics Expect a Skew in Consensus Forecasts Explain Sensitivities to Changes in Independent Variables as Thumb Rules, Not Coefficients Your Forecast Will Be Wrong – Be Ready to Explain If Error Looks More Likely to the Upside or Downside References 5.6 How to Identify and Communicate Downturns in Your Business* Part 1: Forecasting Heroes Catch Turning Points Ways to Forecast a Turning Point Part 2: The Best and Worst Forecasting Year 5.7 Common S&OP Change Management Pitfalls to Avoid* 5.8 Five Steps to Lean Demand Planning* 5.9 The Move to Defensive Business Forecasting* The Limits of Forecast Accuracy The Naïve Model The Next Stage of Forecasting Advancement Defensive Business Forecasting Opportunities for Improvement Afterwords Observations from a Career Practitioner: Keys to Forecasting Success Introduction Characteristics of Companies Successful at Forecasting Concluding Thoughts Demand Planning as a Career REFERENCE How Did We Get Demand Planning So Wrong? Business Forecasting: Issues, Current State, and Future Direction Statistical Algorithms, Judgment and Forecasting Software Systems Good Judgment: Can Software Help? Machine Learning and Judgment REFERENCES The for Forecasting References The Future of Forecasting Is Artificial Intelligence Combined with Human Forecasters The Impact of AI on Forecasting Algorithms The Impact of Humans on AI Algorithms for Forecasting Technological Advances Are Enabling More Effective AI/Human Forecasting Emerging Evidence of the Positive Impact on Forecasting Outcomes Resulting from the Integration of AI Models and Human Forecasters References Quantile Forecasting with Ensembles and Combinations Forecasting Using Possible Futures Quantile Forecasting Evaluating Quantile Forecasts Ensemble Forecasting Combination Forecasting Conclusions Supplements References Managing Demand for New Products Align Supply with Demand Fix Problems That Are Fixable Look for Additional Opportunities Develop an Exit Strategy Strategy Implementation Solving for the Irrational: Why Behavioral Economics Is the Next Big Idea in Demand Planning Business Forecasting in Developing Countries Data and Method Key Findings Suggestions to Close the Gap References Do the Principles of Analytics Apply to Forecasting? Principle 1: Analytics Follows the Data; Analytics Everywhere Principle 2: Analytics Is More Than Algorithms Principle 3: Democratization of Analytics; Analytics for Everyone Conclusion Groupthink on the Topic of AI/ML for Forecasting Everyone Agrees! Let Us Agree Not to Worry about History AI/ML as Secret Sauce The Larger Problem The Benefits of AI/ML to Forecasting Conclusion Taking Demand Planning Skills to the Next Level How to Create Your First ML-Driven Forecast? Where to Start? Notes References Unlock the Potential of Business Forecasting Building a Demand Plan Story for S&OP: The Business Value of Analytics A Demand Plan Story: The Setting A Demand Plan Story: The Actors A Demand Plan Story: The Plot Conclusion About the Editors Index EULA