Where Supply Chain Meets Data Science

Real-world methods for procurement, inventory, and operations professionals who want to go beyond Excel — with reproducible code, realistic datasets, and techniques you can apply today.

What You Will Find Here

Every article on this blog starts with a supply chain problem and solves it with data. No theory without application. No code without context.

Uncover Hidden Patterns — Supplier segmentation with Self-Organizing Maps, hierarchical clustering for material groups, correlation analysis across KPIs. The methods go beyond surface-level reporting to reveal the structures your ERP system collects but never shows you.

Visualize What Matters — Ternary plots for cost driver analysis, correlograms for KPI relationships, process maps from event logs. Each visualization is designed to support a specific decision, not to decorate a slide deck.

Turn Raw Data Into Action — From messy ERP exports to clean, analysis-ready datasets. Data quality assessment, missing value detection, and transformation pipelines that turn transaction logs into strategic insight.

Predict and Optimize — Time series forecasting for demand planning, machine learning for supplier risk scoring, and Demand-Driven MRP for pull-based replenishment. Practical implementations in R and Python that work with the data you already have.

Mine Your Processes — Process mining with bupaR reconstructs actual workflows from timestamps. Find bottlenecks, rework loops, and resource conflicts that process documentation misses — directly from your ERP event data.

Leverage AI for Supply Chain — Large Language Models are changing how we interact with operational data. From automated report generation to intelligent document analysis, this blog explores where AI creates real value in supply chain operations.

Built for Practitioners

This blog is written for supply chain managers, procurement leaders, and operations professionals who want to apply data science to their daily work. Every example uses realistic supply chain datasets — supplier scorecards, inventory records, production logs — not abstract toy problems.

The code is in R and Python. It is reproducible, documented, and ready to adapt to your own data.

Get in Touch

I am available for consulting, speaking engagements, and collaborative projects at the intersection of supply chain management and data science.

Jan-Philipp Grabowski
Freelance Data Scientist & Supply Chain Expert

Email: jan-philipp.grabowski@inphronesys.com
Phone: +49 173 549 1557
LinkedIn: linkedin.com/in/jpgrabowski