AI System for Monitoring Medicine Expiry Dates

The problem

In day-to-day operations, batch risks can be overlooked in a long list. When product, batch, expiry date, quantity, demand and location data are scattered across different screens, documents or team records, problems do not become visible in time.

Fragmented data delays preparation of the batch risk list and forces staff to search manually for supporting evidence again. Before making a decision, the hospital pharmacist has to complete missing fields and verify inconsistencies one by one.

A common exception is that a batch may have quarantine, return or special storage status. Incomplete or incorrect automation that overlooks this context could lead to an unsuitable product being allocated or valid stock being destroyed, so the output cannot be applied directly.

What we set out to improve

• Number of batch alerts confirmed by the pharmacist
• Rate of outputs corrected or rejected by the hospital pharmacist
• Time between the source record and the review output

How the system works

1. 01 Define the scope of the review and the authorised hospital pharmacist role
2. 02 Collect the inputs: product, batch, expiry date, quantity, demand and location
3. 03 Check source identity, date, version and data freshness
4. 04 Sort batches by date and location
5. 05 Match demand with potential transfers
6. 06 Check the exception: a batch may have quarantine, return or special storage status
7. 07 Place low-confidence or inconsistent records in a separate human review queue
8. 08 Write the source identity, rule, output and time to the audit trail
9. 09 Have the hospital pharmacist approve, correct, defer or reject the output

Methods we used

• Batch-level checks of dates and stock movements
• Source, format, version and required-field checks for product, batch, expiry date, quantity, demand and location data
• A missing-data warning instead of an estimate when inputs are missing or inconsistent
• Checking exceptions separately from the main rule
• Storing the source identity, rule result, batch risk list and human decision together

Where people stay involved

Data and security

Access to product, batch, expiry date, quantity, demand and location data is limited to the minimum permissions required for each role. Identifiers and sensitive fields are masked where possible, and only the necessary section is sent to the model rather than the entire raw file. Access, outputs and the hospital pharmacist's decision are recorded; the system does not enter production until the organisation has approved the retention period and data location.

Who this suits

Frequently asked questions