Demand Forecasting in Supply Chain Using AI Basis Machine Learning
summary
Claims that XGB regressor wins on the Walmart dataset using Python out‑of‑the‑box models and basic parameters, reporting 97.2% test accuracy versus lower test accuracy for linear, ridge, KNN (91.9%), decision tree (93.3%) and random forest (95.6%). Features include store, weekly sales, holiday flag, temperature, fuel price, CPI and unemployment; store and CPI rank highest in feature importance. No hyperparameter tuning is performed; recommendation is to tune XGBoost for further gains.
AI-written summary · about
chapters (11)
- 0:00Introduction and Speaker Background
- 0:19Overview of Forecasting in Retail and Supply Chain
- 0:37Types of Forecasting: Demand and Sales
- 1:36Objectives of the Presentation
- 2:09Introduction to the Walmart Dataset
- 3:21Evaluation of Machine Learning Models
- 3:47Detailed Analysis of Models
- 6:18Correlation Metrics and Feature Importance
- 7:16Why XGBoost Outperforms Other Models
- 7:36Conclusion and Recommendations
- 8:23Closing Remarks
related talks
23mMachine Learning with a Deep Dive into Automated Forecasting System
30mAI in Retail - Advancing Promo Demand Forecasting
20mBusiness Value through XAI
59mPredictability Check: Demand Forecasting
38mOptimizing Retail Forecasting Through AI Models
28mMastering Demand Forecasting: Lessons from Europe's Largest Retailer
54mForecasting E‑Commerce Sales with Gradient Boosting
35mBuilding a successful inventory planning solution with Amazon Forecast