How fixed rules and weights change recommendations

Constraints first, preferences second

"I need delivery within five days at most" is a fixed constraint. "Among qualifying offers, I prefer the cheaper one" is a preference. A low price will not save an offer that arrives late.

Prepare comparable offers, non-negotiable constraints, and comparison criteria. For each criterion, define how points are assigned. Set weights before calculating; in this example, they total 100%. The result will be a ranking based on these rules and an explanation of disqualified options.

Three offers, two fixed constraints

We use fictional offers for identical equipment. Prices include shipping and other features are considered equal for this demonstration. These are not live quotes or purchasing recommendations.

Fixed constraints are total price of at most CZK 2,000 and delivery within 5 days of ordering.

• A: CZK 1,800, delivery in 2 days. Meets both constraints.
• B: CZK 1,600, delivery in 4 days. Meets both constraints.
• C: CZK 1,400, delivery in 7 days. Exceeds the delivery limit, so it is disqualified and not scored.

If price or delivery time is missing for an offer, request the data. Until then, it is unverified whether it meets constraints.

Convert price and delivery to points

Currency and days cannot be added directly in this calculation. We use a pre-selected point scale. A higher score means a more favorable outcome.

Price points:

• Up to CZK 1,600 inclusive: 5 points.
• Over CZK 1,600 up to CZK 1,800 inclusive: 4 points.
• Over CZK 1,800 up to CZK 2,000 inclusive: 3 points.

Delivery points:

• Up to 2 days inclusive: 5 points.
• Over 2 days up to 4 days inclusive: 3 points.
• Over 4 days up to 5 days inclusive: 1 point.

Offer A receives 4 points for price and 5 for delivery. Offer B receives 5 points for price and 3 for delivery. This scale is created for demonstration. Its thresholds and point spreads influence the result just like chosen weights.

Weights determine priorities

First, we assign price a weight of 60% and delivery 40%. In calculations, percentages are written as 0.6 and 0.4.

Total score = price points × price weight + delivery points × delivery weight.

• A: 4 × 0.6 + 5 × 0.4 = 4.4.
• B: 5 × 0.6 + 3 × 0.4 = 4.2.

Under these rules, A wins. When we increase price weight to 80% and decrease delivery weight to 20%, the ranking flips:

• A: 4 × 0.8 + 5 × 0.2 = 4.2.
• B: 5 × 0.8 + 3 × 0.2 = 4.6.

Now B wins. Neither the offers nor their points changed; we changed the importance of the criteria. C remains disqualified because the fixed delivery limit still applies. A score is not a probability of a correct decision.

What to check before deciding

• Is data comparable and up to date? For real quotes, record the source and retrieval time.
• Does the selected option satisfy all fixed constraints?
• Do the point scale and weights reflect your true priorities? Test alternative reasonable settings to observe ranking shifts.
• Is any essential criterion missing—such as a feature assumed equal in this example?

In case of a tie, declare a tied result or establish a tie-breaker rule before making a selection. If no offer qualifies, do not automatically relax constraints: search for other options or consciously adjust criteria.

Calculations can be handled by a spreadsheet or routine program. AI can help process inputs, but its data and calculations must be verified. The top score represents the best result according to given rules, not proof of neutrality or an instruction for automatic purchasing. The final action is approved by a human.

Where this workflow is used