AI System for Comparing Food Recipes with Batch Records

The problem

In day-to-day operations, the recipe version and production record can diverge. When the approved recipe, weighing record, ingredient lot, production order and revision note are scattered across sensors, documents, field records or team systems, issues do not become visible in time.

This fragmented data delays preparation of the recipe-to-batch variance report and forces teams to search for supporting evidence again by hand. Before making a decision, the food engineer and production manager must complete missing records and verify each conflict individually.

A common exception is that an approved trial or recipe revision may differ from the standard. Without this context check, incomplete or incorrect automation could block a valid batch or allow a non-conforming batch to proceed, so no physical or operational action can be applied directly.

What we set out to improve

• Number of records the specialist confirms as genuine production discrepancies
• Proportion of outputs corrected or rejected by the food engineer and production manager
• Time between the source event and the inspection output

How the system works

1. 01 Define the review scope and the authorised food engineer and production manager roles
2. 02 Retrieve inputs: approved recipe, weighing record, ingredient lot, production order and revision note
3. 03 Check source identity, date, calibration or version, and data recency
4. 04 Identify the applicable recipe version
5. 05 Compare weighing and lot consumption with the recipe
6. 06 Check for an exception: an approved trial or recipe revision may differ from the standard
7. 07 Route low-confidence or conflicting records to a separate specialist queue
8. 08 Write the source identity, rule, output and timestamp to the audit trail
9. 09 Have the food engineer and production manager approve, correct, defer or reject the output

Methods we used

• Version-controlled reconciliation of recipes and weighing records
• Source, format, timing and required-field checks for the approved recipe, weighing record, ingredient lot, production order and revision note
• A missing-data warning instead of an estimate when inputs are incomplete or conflicting
• Separate evaluation of exceptions and the main rule
• Combined retention of the source, rule result, recipe-to-batch variance report and human decision

Where people stay involved

Data and security

Access to the approved recipe, weighing record, ingredient lot, production order and revision note is limited to the minimum permissions required for the task. Commercial, location and employee information is masked where possible, and only the necessary subset of the raw dataset is passed to the model. Access, outputs and the food engineer's and production manager's decisions are logged; the system is not released to production until the organisation approves the retention period and data location.

Who this suits

Frequently asked questions