In this specialized two-day course, we will discuss a variety of modern trends in Demand Planning and Supply chain forecasting – Machine Learning, AI-driven planning and Big Data analytics. The focus will be on demand modeling with AI-driven planning engines and statistical models, and the process to incorporate market intelligence. We will help you prepare for the role of a winning Data Scientist and a seasoned planning processional. And the two day workshop is the foundation for our Certified Analyst in Demand Planning – CADP program.
This workshop earns a Certificate and eligibility to sit for the CADP certification.
Terminology in Planning – Forecast Horizon, Buckets & Periodicity
Forecast Pass
Demand Management
Data Integration and Cleansing
It is all about the data
The Forecast Problem and Data collection
Define True Demand
Data challenges
- Shipment Vs. Orders
- Gross Demand Vs. Net Demand
Historical shifts in demand
Data filtering
Outliers– Identification and Correction
The process to Identify Outliers
Tolerance band
Methodology for outlier correction
Stat Modeling and AI-driven Planning
Demand Modeling
Key components of demand
Additive Vs. Multiplicative Seasonality in Models
Modeling by decomposition
Introduction to Forecast Modeling
Qualities of a good statistical forecast
Balancing between Model Fit Vs. Model Robustness
Uni-Variate Time Series vs. Multi-Variate methods
Moving Average
AI-driven Planning and Expert Models
Introduction to Planvida
Advanced smoothing models
First Order Exponential Smoothing
Holt Models to accommodate the trend
Holt-Winters Model
Exponential Trend and Dampening
Interaction between components
Modeling special cases of Demand Product Life Cycle & Long-term Planning
Product Lifecycle and trend • Launch Forecasting • Volume effect on line extension
Event Modeling
Baseline vs. Incremental • Illustration of Event Models
Planning for Intermittent Demand
What is Intermittent Demand & what causes it? Strategies to handle intermittent demand & Statistical Models for Intermittent Demand
Higher-order Models
Product Portfolio Management
Impact of Data Volatility on Forecasting
Measuring Volatility
Impact of multiple Extreme Observations on Volatility
SKU Segmentation for demand modeling & inventory strategies
Modeling by exception
ABC analysis - Classification philosophy
Pareto analysis based on dollar usage
Item criticality
Excess, obsolete and Slow-moving Alignment with the product lifecycle
Discontinuance and end of life (EOL)
Process flow for Segmenting SKUs
Example using a three-dimensional matrix; ABC / Volume / Critical / Status; the excess, obsolete impact of Segmentation on Cycle Counting and Inventory Accuracy.
Demand Planning Analytics toolkit
Definition of Demand Forecast Errors
Forecast Accuracy
Forecast Bias vs. Forecast Error
Error and Volatility Reduction
Errors across SKUs vs. Errors across time
Model Diagnostics vs Performance
- MAD
- MAPE vs. MPE
- WAPE
- Root Mean Squared Error
Measuring forecast performance
Forecast Performance Metric
Forecast errors and actionability
Sources of Forecast Error
Definition of Demand Planning Metrics - WAPE & Bias
Types of Bias
SKU Mix Error
Error Analysis for Continuous Improvement
Why S&OP
Fragmented Planning Activities
– Supply chain challenges – Service, costs and inventories
Disparities between the Financial forecast and operational forecasts
Bottom Line challenges from Fragmented Planning
Benefits of a holistic S&OP Design
Core Components of SIOP
Consensus Demand Planning
Rough Cut Planning and Supply Collaboration
Executive Presentation
Demand-Supply Balancing
Modeling by exception
Workshop cost includes two days of interactive learning, breakfast lunch and refreshment breaks. Attendees are responsible for their own accommodation at the Hilton Burlington.
