Tag: Forecasting
-
Your Exception Report Is Lying to You: How to Make ‘1% False Alarms’ Actually Mean 1%
A static ±3σ band promised 0.27% false alarms and delivered 3.80%. Here is how to turn your exception threshold from a statistics setting into an honest attention budget.
-
Does Your Forecast Beat a Sticky Note? The Placebo Test for Demand Planning
Your forecast has exactly one job: beat a sticky note that says ‘same as last year.’ Most don’t. Here’s the placebo test that proves it.
-
30 Books Every Forecaster, Demand Planner, and S&OP Lead Should Read — The Inphronesys Bookshelf
An interactive 30-book bookshelf for forecasters, demand planners, S&OP leads, supply chain strategists, and the data scientists who keep them honest. Filter by category and level, build a reading list for your role.
-
The Folly of Forecasting
From the Pythia at Delphi to Google’s GraphCast — humans have always demanded the future. Why some forecasts have gotten dramatically better, why others haven’t, and why AI will not deliver the silver bullet.
-
Why VAR Beat Google’s TimesFM — and How to Build One in R
A peer-reviewed 2025 study put Google’s TimesFM foundation model head-to-head with vector autoregression on real hospital data. Spoiler: the 1980s econometric model won. Here’s what VAR is, why it works for supply chain, and how to build one in R.
-
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.
-
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.