Examples
Examples
Try the starting example
- Criteria: name, weight, lower bound, upper bound, benefit or cost
- Cost,1,0,100,cost Quality,1,0,10,benefit
- Vendors: name, raw criterion values in order; NA means missing
- A,0,5 B,50,10 C,25,NA
- Normalised vendor criterion scores (0–100)
- A: 100, 50; B: 50, 100; C: Unranked
- Weighted score of complete vendors
- A: 75; B: 75
- Highest-scoring complete vendor(s)
- A, B
- Missing-data flags
- C: Quality
- Raw values outside supplied bounds
- None
Try another scenario
- Criteria: name, weight, lower bound, upper bound, benefit or cost
- Cost,1,0,100,cost Quality,1,0,10,benefit
- Vendors: name, raw criterion values in order; NA means missing
- A,150,5 B,50,10 C,25,NA
- Normalised vendor criterion scores (0–100)
- A: 0, 50; B: 50, 100; C: Unranked
- Weighted score of complete vendors
- A: 25; B: 75
- Highest-scoring complete vendor(s)
- B
- Missing-data flags
- C: Quality
- Raw values outside supplied bounds
- A: Cost
Calculation method
Calculation method
Normalised vendor criterion scores (0–100)
Benefit: 100×(value−lower)/(upper−lower). Cost: 100 minus that result. Clamp to 0–100 and flag raw values outside bounds.
Weighted score of complete vendors
Normalise entered weights and sum criterion score×weight only for vendors with all required data.
Highest-scoring complete vendor(s)
Retain exact score ties; missing vendors are not ranked.
Missing-data flags
List each vendor’s NA criteria without imputing values.
Raw values outside supplied bounds
Expose clamping rather than silently treating an out-of-bound value as normal.
Symbols and units
- a
- Criteria: name, weight, lower bound, upper bound, benefit or cost
- b
- Vendors: name, raw criterion values in order; NA means missing
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- A bounded raw-metric scorecard under your supplied scales and weights. Missing data is never guessed, and clamped outliers remain flagged. Different bounds or weights can change the ranking.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.
Further reading
UK Government Analysis Function: MCDA scoring ↗
Background reading; this does not imply endorsement or independent review of SumFizz.