Solutions

Meal production planning and ingredient forecasting

Production planning is the join between what customers chose and what the business has to buy. When those two are maintained separately, food cost drifts and nobody can say exactly why.

From a selection to a purchase order

Each step is arithmetic on the previous one. Break the chain and the last number stops being trustworthy.

  1. Selections

    Every customer’s chosen meals for the days inside the production window, frozen at the cut-off.

  2. Aggregate counts

    Portions per recipe, summed. This is what the kitchen cooks to.

  3. Explode the recipes

    Each recipe’s bill of materials multiplied by its count, producing raw ingredient quantities in canonical units.

  4. Deduct what is on hand

    Subtract current stock to get the true requirement rather than the gross one.

  5. Order or flag

    Produce the supplier requirement, and raise a shortfall where the gap cannot be closed in time.

One source, two outputs

The rule that keeps this honest is that production counts and ingredient forecasts must derive from the same selections. If a purchasing sheet is maintained by hand alongside a cook list, the two will diverge on the first mid-week change, and the divergence shows up as either waste or a missing ingredient on a Tuesday morning.

This is also what makes a plan change safe. Add five meals to a customer’s week and the counts rise, the ingredient requirement rises with them, and the supplier order or shortfall updates — without anyone recalculating anything.

Canonical units are not a detail

Recipes get written in whatever unit the person writing them was thinking in: grams, kilograms, pieces, bunches. Purchasing needs one unit per ingredient or the arithmetic is meaningless.

Storing quantities in canonical units and converting at the edges is dull, and it is the difference between a forecast that can be handed to a supplier and one that has to be checked by hand first.

Common questions

What is recipe explosion?
Multiplying each recipe’s bill of materials by how many portions of it are being produced, then summing across every recipe, to get total raw ingredient requirements for a production window.
How far ahead should ingredients be forecast?
As far as the supplier lead time requires, and no further than the cut-off allows you to trust the selections. Forecasting past the point where customers can still change their meals produces confident numbers that are wrong.
What should happen when a shortfall is detected?
It should be raised as an explicit exception rather than silently absorbed. A forecast that quietly rounds a shortage away is worse than no forecast, because the kitchen finds out on the morning it needed the ingredient.

See it running against a real kitchen

The product pages show the actual admin, loaded with a demonstration kitchen — production, packing, routes and the dashboard, exactly as an operator sees them.