Master the demand forecasting skills you need to decide what resources to acquire, what products to produce, and where and how to distribute them.
In Demand Forecasting Best Practices you’ll learn how to:
• Lead a demand planning team to improve forecasting quality while reducing workload
• Properly define the objectives, granularity, and horizon of your demand planning process
• Use smart, value-weighted KPIs to track accuracy and bias
• Spot areas of your process where there is room for improvement
• Help planners and stakeholders (sales, marketing, finances) add value in your process
• Identify what kind of data you should be collecting, and how
• Utilize different types of statistical and machine learning models
Demand Forecasting Best Practices teaches you to optimize demand planning to deliver a more effective supply chain. In this unique step-by-step guide, you’ll learn forecasting tools, metrics, and models alongside stakeholder management techniques that work in a live business environment. Follow author Nicolas Vandeput’s original five step framework for demand planning excellence and learn how to tailor it to your own company’s needs. You’ll soon be delivering accurate predictions that are driving major business value.
About the technology
Demand forecasting is vital for the success of any product supply chain. It allows companies to make better decisions about what resources to acquire, what products to produce, and where and how to distribute them. As an effective demand forecaster, you can help your organization avoid overproduction, reduce waste, and optimize inventory levels for a real competitive advantage.
About the book
Demand Forecasting Best Practices is a handbook of techniques for effective demand planning for products of all types. You’ll learn how to optimize your data, metrics, processes, models, and even people to make better decisions and deliver value to your supply chains. Discover pro tips from author Nicolas Vandeput’s global career in supply chain consultancy, and dodge the common mistakes you might not know you’re making. Illustrations, clear explanations, and relevant real-world examples make each concept easy to understand and easy to follow.
About the reader
For anyone who wants to improve their demand planning process, including demand planners, S&OP managers, supply chain leaders, and data scientists.
About the author
Nicolas Vandeput is a supply chain data scientist specializing in demand forecasting and inventory optimization. He founded his consultancy company SupChains in 2016, delivering models and training courses worldwide. He co-founded SKU Science—a demand forecasting platform—in 2018. Passionate about education, Nicolas is an avid learner enjoying teaching at universities. He currently teaches forecasting and inventory optimization to master students in CentraleSupélec, Paris, France.
Author(s): Nicolas Vandeput
Edition: 1
Publisher: Manning
Year: 2023
Language: English
Commentary: Publisher's PDF
Pages: 216
City: Shelter Island, NY
Tags: Machine Learning; Data Analysis; Data Science; Statistics; Forecasting; Demand
Demand Forecasting Best Practices
Full quotes from reviewers of Demand Forecasting Best Practices
brief contents
contents
preface
acknowledgments
about this book
How this book is organized: a roadmap
liveBook discussion forum
about the author
about the cover illustration
Part 1—Forecasting demand
1 Demand forecasting excellence
1.1 Why do we forecast demand?
1.2 Five steps to demand planning excellence
1.2.1 Objective. What do you need to forecast?
1.2.2 Data. What data do you need to support your forecasting model and process?
1.2.3 Metrics. How do you evaluate forecasting quality?
1.2.4 Baseline model. How do you create an accurate, automated forecast baseline?
1.2.5 Review Process. How to review the baseline forecast, and who should do it?
Summary
2 Introduction to demand forecasting
2.1 Why do we forecast demand?
2.2 Definitions
2.2.1 Demand, sales, and supply
2.2.2. Supply plan, financial budget, and sales targets
Summary
3 Capturing unconstrained
3.1 Order collection and management
3.2 Shortage-Censoring and Uncollected Orders
3.2.1. Using demand drivers to forecast historical demand
3.3 Substitution and cannibalization
Summary
4 Collaboration: data sharing
4.1 How supply chains distort demand information
4.2 Bullwhip effect
4.2.1 Order forecasting
4.2.2 Order batching
4.2.3 Price fluctuation and promotions
4.2.4 Shortage gaming
4.2.5 Lead time variations
4.3 Collaborative planning
4.3.1 Internal collaboration
4.3.2 External collaboration
4.3.3 Collaborating with your suppliers
Summary
5 Forecasting hierarchies
5.1 The three forecasting dimensions
5.2 Zooming in or out of forecasts
5.3 How do you select the most appropriate aggregation level?
5.3.1 Which aggregation level should you focus on?
5.3.2 What granularity level should you use to create your forecast?
Summary
6 How long should the forecasting horizon be?
6.1 Theory: Inventory optimization, lead times, and review periods
6.2 Reconciling demand forecasting and supply planning
6.3 Looking further ahead
6.3.1 Optimal service level and risks
6.3.2 Collaboration with suppliers
6.4 Going further: Lost sales vs. backorders
6.4.1 Lost sales
6.4.2 Backorders
6.4.3 Hybrid
Summary
7 Should we reconcile forecasts to align supply chains?
7.1 Forecasting granularities requirements
7.2 One number forecast
7.3 Different hierarchies . . . different optimal forecasts
7.3.1 Spot sales and stock clearances
7.3.2 Product life-cycles
7.3.3 Example: top-down vs. bottom up
7.4 One number mindset
Summary
Part 2—Measuringforecasting quality
8 Forecasting metrics
8.1 Accuracy and bias
8.2 Forecast error and bias
8.2.1 Interpreting and scaling the bias
8.2.2 Do it yourself
8.2.3 Insights
8.3 Mean Absolute Error (MAE)
8.3.1 Scaling the Mean Absolute Error
8.3.2 Do it yourself
8.3.3 Insights
8.4 Mean Absolute Percentage Error (MAPE)
8.4.1 Do it yourself
8.4.2 Insights
8.5 Root Mean Square Error (RMSE)
8.5.1 Scaling RMSE
8.5.2 Do it yourself
8.5.3 Insights
8.6 Case study – Part 1
Summary
9 Choosing the bestforecasting KPI
9.1 Extreme demand patterns
9.2 Intermittent demand
9.3 The best forecasting KPI
9.4 Case study – Part 2
Summary
10 What is a good forecast error?
10.1 Benchmarking
10.1.1 Naïve forecasts
10.1.2 Moving average
10.1.3 Seasonal benchmarks
10.2 Why tracking demand coefficient of variation is not recommended
10.2.1 COV and simple demand patterns
10.2.2 COV and realistic demand patterns
Summary
11 Measuring forecasting accuracy
11.1 Forecasting metrics and product portfolios
11.2 Value-weighted KPIs
Summary
Part 3—Data-driven forecasting process
12 Forecast value added
12.1 Comparing your process to a benchmark
12.1.1 Internal benchmarks
12.1.2 Industry (external) benchmarks
12.2 Tracking Forecast Value Added
12.2.1 Process efficacy
12.2.2 Process efficiency
12.2.3 Best practices
12.2.4 How do you get started?
Summary
13 What do you review? ABC XYZ segmentations and other methods
13.1 ABC XYZ segmentations
13.1.1 ABC analysis
13.1.2 ABC XYZ analysis
13.2 Using ABC XYZ for demand forecasting
13.2.1 Products’ importance
13.2.2 Products’ forecastability
13.2.3 ABC XYZ limitations
13.3 Beyond ABC XYZ: Smart multi-criteria classification
Summary
Part 4—Forecasting methods
14 Statistical forecasting
14.1 Time series forecasting
14.1.1 Demand components: Level, trend, and seasonality
14.1.2 Setting up time series models
14.2 Predictive analytics and demand drivers
14.2.1 Demand drivers
14.2.2 Challenges
14.3 Times series forecasting vs. predictive analytics
14.4 How to select a model
14.4.1 The 5-step framework
14.4.2 4-step model creation framework
Summary
15 Machine Learning
15.1 What is machine learning?
15.1.1 How does the machine learn?
15.1.2 Black boxes versus whites boxes
15.2 Main types of learning algorithms
15.2.1 Short history of machine-learning models
15.2.2 Tree-based models
15.2.3 Neural networks
15.3 What should you expect from ML-driven demand forecasting?
15.3.1 Forecasting competitions
15.3.2 Improving the baseline
15.4 How to launch a machine-learning initiative
Summary
16 Judgmental forecasting
16.1 When to use judgmental forecasts?
16.2 Judgmental biases
16.2.1 Cognitive biases
16.2.2 Misalignment of incentives (intentional biases)
16.2.3 Biased forecasting process
16.3 Group forecasts
16.3.1 Wisdom of the crowds
16.3.2 Assumption-based discussions
Summary
17 Now it’s your turn!
Closing words
references
index
