AI System for Managing Cancelled Appointment Slots

The problem

In daily operations, a last-minute opening may go unused when staff have to search a list by hand. When the cancellation record, appointment duration, resource calendar, waiting list and contact permission are spread across different screens, documents or team records, the opening may not be identified in time.

This fragmented data delays preparation of the contactable candidate list and forces staff to search again for supporting information. Before deciding, the patient services scheduling officer must complete missing fields and verify conflicts one by one.

A common exception is that a procedure may require special preparation or an accompanying person. If this context is not checked, an incomplete or incorrect workflow could suggest the slot to an unsuitable patient, so its output cannot be applied directly.

What we set out to improve

• Number of suggestions that do not meet the eligibility conditions
• Rate of outputs corrected or rejected by the patient services scheduling officer
• Time between the source record and the review output

How the system works

1. 01 Define the review scope and the authorised patient services scheduling officer role
2. 02 Receive the cancellation record, appointment duration, resource calendar, waiting list and contact permission
3. 03 Check source identity, date, version and data freshness
4. 04 Extract the duration and resource conditions for the open slot
5. 05 Filter the list by eligibility and contact permission
6. 06 Check the exception that the procedure may require special preparation or an accompanying person
7. 07 Place low-confidence or conflicting 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 patient services scheduling officer approve, correct, defer or reject the output

Methods we used

• Candidate filtering with deterministic calendar and preparation rules
• Source, format, version and mandatory-field checks for the cancellation record, appointment duration, resource calendar, waiting list and contact permission
• A missing-data warning instead of inference when inputs are absent or conflicting
• Separate assessment of exceptions and the main rule
• Joint retention of the source identity, rule result, contactable candidate list and human decision

Where people stay involved

Data and security

Cancellation records, appointment durations, resource calendars, waiting lists and contact permissions are accessed with the narrowest permissions required for the role. Identity and sensitive fields are masked where possible, and only the necessary section rather than the entire raw file is sent to the model. Access, output and the patient services scheduling officer's decision are logged; production use does not begin until the organisation approves data retention and location.

Who this suits

Frequently asked questions