Where SCM meets data science and AI.
Real-world methods for supply chain and operations management professionals who want to go beyond Excel — with reproducible code, realistic datasets, and techniques you can apply today.
Newest Entries
-
AI Model Distillation, Explained: The Compression Trick Behind the $589 Billion Monday
A single Chinese app knocked $589 billion off Nvidia in one day. The trick…
-
I Raced Amazon’s Forecasting AI Against a 66-Year-Old Formula. The Formula Won.
I put Amazon’s zero-shot forecasting AI against five classical models on 148 real retail…
-
The 20 Most Influential People in Forecasting (And What to Learn From Each)
A TIME 100-style guide to the academics, ML engineers, and supply chain voices who shaped modern forecasting — and the one resource to start with for each.
-
When the Algorithm Is Wrong and the Expert Is Right
Statistical models don’t know about your supplier’s factory fire, your competitor’s clearance sale, or the regulation that just changed. Here’s when expert judgment beats the algorithm — and the biases that make it worse.
-
The M5 Lesson: Why Simple Still Beats Fancy in Supply Chain Forecasting
The 2020 M5 competition taught a lesson the forecasting world keeps forgetting: on real supply chain data, simple models win more often than you’d think. Here’s what the Walmart SKU benchmark actually showed — and why it matters for today’s Time Series Foundation Model hype.
-
I Ran 6 Models on Real Demand Data — Here’s How I Picked the Winner
Six forecasting models, one real demand series, one honest horse race. Here’s the model that won — and the metric that made the choice unambiguous.
-
Is Your Forecast Any Good? The Forecaster’s Toolbox
Four acronyms decide whether you trust a forecast: MAE, MAPE, RMSE, MASE. Here is when each one lies to you — and the one benchmark that catches them all.
-
Design for AI: What 40 Years of DFMA Teaches Us About Working With LLMs
Forty years ago, Boothroyd & Dewhurst showed that manufacturing cost was a design problem. Today, bad prompts and failing AI pilots are the same kind of mistake. Here are 7 principles for designing work that LLMs can actually execute.
-
Three Equations From the Navy in 1957: Why Holt-Winters Still Runs Your Forecast Engine
Holt-Winters wasn’t born in a statistics lecture — it was written for the U.S. Navy in 1957 to solve an inventory problem. Nearly seven decades later, three recursive equations still beat most of the software sitting on top of your ERP.
-
Anthropic Just Killed One of Its Own Prompting Tricks — Here’s What That Means
Six years ago prompting was a happy accident inside a GPT-3 paper. Today it’s the single skill separating AI winners from losers. Here’s the complete history — what Anthropic, OpenAI and Google actually published, and what still matters in 2026.