Production choice analysis workshop / 7 MIN READ
Inspect the options hidden by a weighted score
Check what a short list leaves out before presenting it as complete.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-07
A tidy recommendation should not erase a usable alternative merely because the team has adopted a convenient comparison format.
What you will leave with: A complete package comparison and an explanation of the resulting recommendation.
Use this when
- A producer must select one complete delivery package.
- A ranking presentation excludes an option that seems to fit the actual brief.
Not a spending authorization, supplier assessment or universal preference model.
Start with the exhibition, not the score
A fictional gallery commissions a portable exhibition guide. Its opening date is fixed, and the communications team has a limited production allowance. The guide must be usable at the opening rather than merely impressive in a proposal. The producer receives several complete packages and wants a short recommendation that the curator can understand without studying a technical appendix.
With both objectives minimized, one package dominates another if it is no worse in either and strictly better in at least one. The survivors form the nondominated set.
Fictional complete packages have (cost units, days): A(2,10), B(6,7), C(10,2), D(8,9). B dominates D; A, B, C remain. These are separate options, not divisible components.
For the exercise, stipulate that all packages provide the same approved content and required accessibility work. In an actual procurement, that assumption would need checking. A proposal that leaves the accessible edition to the gallery is not the same job as one that includes it. Do that comparison before reducing the offers to the two numbers used here.
Keep the original package descriptions beside the exercise. The price and delivery labels are not permission to invent an offer between the listed packages, or to combine one supplier’s attractive promise with another supplier’s attractive price. The curator needs a choice that could actually be commissioned, not an appealing imaginary blend.
Reconstruct the short list instead of trusting its format
Use score = w × cost + (1−w) × days, with 0<w<1 and lower preferred. Scores are A:10−8w, B:7−w, C:2+8w. B beating or tying A requires w≤3/7; beating or tying C requires w≥5/9. Both cannot hold. At w=1/2, A=C=6 and B=6.5. Kim and de Weck explain this weighted-sum limitation.
Use the displayed arithmetic as a teaching witness. The numeric scales are fixed; no unit conversion is modeled. Do not present this invented score as the curator’s stated preference. Ask the curator to explain the actual work rather than asking them to approve an unfamiliar calculation merely because it has been printed on the recommendation.
Ask the analyst to present the original package sheet next to the shortened recommendation. A reader should be able to follow each disappearance rather than seeing only the final winners. Avoid criticizing a colleague for using a familiar presentation method. The useful conversation concerns what the method answers and what the project still needs to know.
Invite the curator to explain the practical consequence of missing the opening. Is a late guide still useful for the remainder of the exhibition, or would it miss the only occasion when visitors receive a printed copy? The answer belongs in the brief. It should not be concealed in a number that the curator never had an opportunity to interpret.
Return to the actual promise
Under cost≤7 and days≤8, only B qualifies.
The fictional producer now has a practical reason to discuss the middle proposal. Before recommending it commercially, ask what the delivery promise includes: arrival at the gallery, completion at the printer or approval of artwork. Also identify whose response is needed before the delivery clock begins. These are concrete questions about the work, not reasons to quietly alter the teaching inputs.
Preserve the curator’s original opening requirement when asking those questions. If the supplier cannot support it, report that gap directly. Do not rescue a preferred proposal by relabeling a late delivery as an acceptable launch phase unless the curator actually chooses a changed plan.
If the team has no acceptable offer after clarification, return with that result. The next useful discussion might concern the format, the scope or whether an interim guide would serve visitors. Those are new creative choices to explore with the project owner. They should be presented as changes to the plan rather than attributed to the earlier arithmetic.
Write a recommendation that leaves room for scrutiny
The final fictional recommendation names the package, the common scope assumed in the exercise and the supplier questions that remain unresolved. It explains the exhibition need in everyday language. The curator should not have to accept a mathematical label as a substitute for understanding what visitors will receive and when.
Keep an annotated version of the original package sheet with the recommendation. If the opening moves or the guide format changes, start from the actual options and refreshed offers rather than inheriting an unexplained short list. The earlier work remains useful as a record of the former decision, without becoming an instruction for a different project.
Conclude the exercise by asking a second reader to challenge the recommendation from the visitor’s perspective. Can they identify the intended use, the unresolved delivery detail and the assumption about equivalent content? If those facts are hard to find, improve the brief before making the presentation more polished. No real supplier is contacted or selected in this exercise.
Make the reasoning concrete.
A curator receives a two-option recommendation but remembers another complete proposal that appears usable for the opening. The producer must explain the omission before seeking a commercial decision.
Defend the presentation
Opening: Treat the shortened recommendation as the only relevant choice.
Hypothesis: The meeting stays simple.
Risk: The curator cannot inspect the reason an actual offer disappeared.
Reopen the comparison
Opening: Bring the original package sheet and the exhibition brief into the same discussion.
Hypothesis: The recommendation can answer the real visitor-facing need.
Risk: The team may still need supplier clarification or a revised creative plan.
The producer presents the full comparison and resolves the delivery meaning before any purchasing decision.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Visitor use
What must be available when visitors arrive?
Common scope
Which required work is stipulated equal here but needs checking in real offers?
Completeness
Can the curator trace every omission from the original package sheet?
Delivery meaning
Which event starts and ends the promised delivery period?
Changed brief
What new creative choice would be needed if no actual offer works?
Watch for the wrong conclusion.
- Treating an unexplained short list as a complete account of available offers.
- Combining incompatible pieces of separate supplier promises.
- Changing the opening requirement privately to rescue a favored proposal.
Your next action
Rebuild the fictional comparison, then write the delivery clarification the producer needs before recommending a real offer.
Sources & scope
All business examples and numerical fixtures are original and fictional. Sources support the identified mathematical or implementation distinctions. Qualified analysts own model specification and inference. No causal result, model fit, actual customer data or operational change is claimed.
- Kim and de Weck: Adaptive weighted-sum method →Primary research paper · Supports: Supports the weighted-sum limitation.
- discopt: Multi-Objective Optimization →Primary documentation · Supports: Supports the comparison concepts.