AI System for Preparing Procurement RFQ Packages

The problem

When the internal request, specification, qualification requirements and delivery terms are held in different sources, their fields must be matched manually before a supplier RFQ package draft can be prepared.

The procurement specialist must compare every suggestion in the supplier RFQ package draft with the source record; missing or conflicting evidence must not replace the final decision.

If suppliers excluded from the scope for confidentiality reasons are not referred for separate review, incomplete or confidential information may be shared incorrectly; the record must not proceed without a human decision.

What we set out to improve

• Time taken for the supplier RFQ package draft to reach the procurement specialist for review with its supporting evidence
• Outcomes of records referred for human review because some suppliers are excluded from the scope for confidentiality reasons
• Number of suggestions corrected after identifying the risk of sharing incomplete or confidential information incorrectly

How the system works

1. 01 Retrieve the internal request, specification, qualification requirements and delivery terms from authorised sources using the record identifier
2. 02 Link source, date and version details to the shared work record
3. 03 Validate fields and units
4. 04 Prepare the supplier RFQ package draft with its supporting evidence
5. 05 Check for the exception: some suppliers are excluded from the scope for confidentiality reasons
6. 06 The procurement specialist accepts, amends or rejects the draft
7. 07 Write the approved decision and its source to the audit log

Methods we used

• A source-referenced shared data schema for the internal request, specification, qualification requirements and delivery terms
• A decision structure that separates deterministic rules from AI suggestions
• A review screen showing the supplier RFQ package draft alongside the underlying evidence
• A mandatory human review queue when some suppliers are excluded from the scope for confidentiality reasons
• An event identifier that prevents duplicate processing

Where people stay involved

Data and security

Only the fields required for this decision should be processed from the internal request, specification, qualification requirements and delivery terms. Access to source systems should be authorised separately from the procurement specialist role; personal, commercial and technical secrets should not be passed into the model context unnecessarily. Reads, suggestions and human decisions should be logged separately.

Who this suits

Frequently asked questions