Bid Scoring

Overview#

Procure's bid scoring engine evaluates submitted bids against the tender's weighted evaluation criteria. The AI assesses each bid across multiple dimensions, produces per-criterion scores, identifies strengths and weaknesses, checks compliance, and generates a ranked recommendation.

Submitting Bids#

Bids are submitted against a specific tender:

curl -X POST http://localhost:8014/api/v1/tenders/{tender_id}/bids \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $TOKEN" \
  -d '{
    "supplier_id": "sup-001",
    "amount": 6500000,
    "proposal": "Our team brings 15 years of GCC government cloud migration experience. We propose a phased 4-month migration using our sovereign cloud platform, with dedicated Arabic-speaking project management and 24/7 support throughout the transition.",
    "delivery_days": 120,
    "qualifications": [
      "ISO 27001:2022 certified",
      "SOC 2 Type II compliant",
      "3 completed GCC government cloud migrations",
      "Native Arabic project management team"
    ]
  }'

Bid Fields#

Evaluating a Single Bid#

Run AI evaluation on an individual bid:

curl -X POST http://localhost:8014/api/v1/tenders/{tender_id}/bids/{bid_id}/evaluate \
  -H "Authorization: Bearer $TOKEN"

Response:

{
  "bid_id": "bid-001",
  "scores": {
    "technical_approach": 88.5,
    "team_qualifications": 92.0,
    "price": 75.0,
    "timeline": 85.0,
    "local_content": 90.0
  },
  "total_score": 86.2,
  "strengths": [
    "Strong GCC government migration track record",
    "Native Arabic project management capability",
    "Comprehensive sovereign cloud infrastructure"
  ],
  "weaknesses": [
    "Bid amount is at the upper end of the budget range",
    "Proposed timeline of 120 days is shorter than typical for this scope"
  ],
  "compliance": true,
  "value_assessment": "Strong technical proposal with demonstrated GCC experience. Price is competitive given the qualifications. Recommended for shortlisting."
}

Evaluation Output#

Bulk Evaluation#

Evaluate all bids for a tender simultaneously and get a ranked comparison:

curl -X POST http://localhost:8014/api/v1/tenders/{tender_id}/evaluate-all \
  -H "Authorization: Bearer $TOKEN"

Response:

{
  "tender_id": "tnd-a1b2c3",
  "evaluations": [
    {
      "bid_id": "bid-001",
      "scores": {"technical_approach": 88.5, "price": 75.0, ...},
      "total_score": 86.2,
      "strengths": [...],
      "weaknesses": [...],
      "compliance": true,
      "value_assessment": "..."
    },
    {
      "bid_id": "bid-002",
      "scores": {"technical_approach": 72.0, "price": 92.0, ...},
      "total_score": 79.8,
      "strengths": [...],
      "weaknesses": [...],
      "compliance": true,
      "value_assessment": "..."
    }
  ],
  "ranking": [
    {"bid_id": "bid-001", "rank": 1, "total_score": 86.2},
    {"bid_id": "bid-002", "rank": 2, "total_score": 79.8}
  ],
  "recommendation": "Bid from Supplier A (bid-001) scores highest across technical approach and team qualifications. While the price is higher, the technical superiority and sovereign cloud capability justify the premium. Recommend awarding to Supplier A."
}

Scoring Methodology#

The AI evaluates each criterion by analyzing the bid's proposal text, amount, delivery timeline, and qualifications against the tender's requirements:

• Technical scores are based on how well the proposal addresses the tender's technical requirements
• Price scores compare the bid amount against the budget range and other bids
• Timeline scores evaluate delivery feasibility relative to the tender's timeline expectations
• Qualification scores match certifications and experience against stated requirements
• Compliance is a binary check against mandatory requirements

All scores are on a 0-100 scale, and the total_score is the weighted sum based on the tender's evaluation criteria weights.