Tag
inventory planning
Production-planning notes filed under inventory planning.
- 01
The Bullwhip Effect on Your CPG Line: Why a 5% Shelf Wobble Becomes a Whipsawing Schedule
Your inbound order signal swings harder than consumers actually buy. That jumpiness is a structural amplification called the bullwhip effect. Here's how to name it, measure it, and dampen it from the plant floor.
Read post - 02
Croston's Method for Intermittent SKUs: Forecasting the Slow Movers Your Average Keeps Getting Wrong
Standard moving averages and exponential smoothing quietly mis-forecast your sporadic SKUs. Croston's method forecasts demand size and demand interval separately — here's how it works, when to use SBA or TSB, and how to decide which SKUs need it.
Read post - 03
Little's Law for CPG Schedulers: Predict Line Lead Time from WIP and Throughput
The line is running at rate, yet orders keep shipping late. Little's Law explains why: lead time is WIP divided by throughput. Learn to predict flow time, cap queues, and quote promise dates you can keep.
Read post - 04
Tracking Signal: Catch a Biased Forecast Before It Corrupts Your Schedule
A forecast rarely fails loudly — it drifts. Learn how the tracking signal, a cheap MAD-based trip-wire, flags persistent forecast bias so you can fix it before it inflates run sizes or triggers repeat stockouts.
Read post - 05
The Newsvendor Model: How Much to Make for a One-Shot CPG Run
Holiday packs, LTOs, and short-code items give you exactly one shot at the run quantity. The newsvendor model turns overage-vs-underage costs into a defensible batch size using one number: the critical ratio.
Read post - 06
Dynamic Lot-Sizing for Lumpy CPG Demand: When One Fixed Batch Size Stops Working
EOQ and EPQ assume steady demand. Real CPG demand is lumpy — promo spikes, seasonal resets, uneven retailer orders. Here's how lot-for-lot, POQ, Silver-Meal, and Wagner-Whitin size runs when net requirements change every week.
Read post - 07
Economic Production Quantity (EPQ): Sizing Batches When Your Line Makes and Ships at Once
Run long to dilute a painful changeover, or run short to avoid drowning in inventory? EPQ is the math for the sweet spot — built for a line that fills stock gradually while demand keeps pulling it down.
Read post - 08
Cycle Service Level vs. Fill Rate: The Service Metric That Sizes Your Safety Stock
Setting "95% service" means nothing until you know which 95%. Learn how cycle service level and fill rate differ, why they diverge, and how each one sizes your safety stock and run cadence.
Read post - 09
Shelf-Life-Constrained Production Scheduling: Sizing Runs So Perishable SKUs Don't Expire
Classic lot-sizing math quietly assumes product lives forever. In food and beverage it doesn't. Here's how to cap production runs at the shelf-life-feasible quantity so perishable SKUs sell before they expire — without stocking out.
Read post - 10
ABC-XYZ Segmentation for CPG: Which SKUs Deserve Tight Scheduling and Safety Stock
You can't manage 400 SKUs the same way. Learn how ABC-XYZ segmentation ranks SKUs by dollar impact and demand variability, so you spend scheduling attention and safety stock where they actually pay off.
Read post - 11
EPEI (Every Product Every Interval): How Often Should Your Line Make Each SKU?
EPEI is the one number that names your production rotation — how often each SKU comes around. Here's what it means, why smaller is usually better, and how to calculate and shrink it on a real CPG line.
Read post - 12
Heijunka for CPG Lines: Turning Lumpy Demand Into a Schedule the Floor Can Run
Forecasting tells you how much to make. Heijunka tells you in what rhythm and sequence. Here's how production leveling smooths lumpy CPG demand into a repeatable schedule that protects fill rate and your changeover budget.
Read post - 13
Stop firefighting stockouts
A stockout is rarely a surprise — it's a slow signal nobody had time to read. Read the few signals that matter early, and the fixes stay cheap.
Read post - 14
Forecasting that survives contact with the floor
Tie par levels to how product actually moves — with the right window, the right buffer, and current data — and the Tuesday surprises mostly stop.
Read post