Capacity Planning Prompts: Scenarios Without Spreadsheet Hell

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-operations-capacity-planning-prompts-adults

1. A capacity planning model contains 200 cells. How many assumptions are likely embedded in this model?

   1. Zero assumptions, because formulas calculate deterministically
   2. Approximately 200 assumptions, most of which cannot be recalled on demand
   3. Exactly 10 assumptions that the planner actively remembers
   4. Exactly one assumption per formula

2. In a scenario-sweep prompt, which of the following should NOT be requested as an output for each scenario?

   1. Headcount needed
   2. Top 3 assumptions driving the result
   3. Employee satisfaction scores
   4. Ending inventory levels

3. What should be flagged during scenario analysis as a warning sign?

   1. Outputs that match historical data closely
   2. Minor variations in capex estimates
   3. Large jumps in output resulting from small input changes
   4. Consistent assumptions across all scenarios

4. Why should capacity planning outputs be rounded to match input precision?

   1. Because rounded numbers are easier to graph
   2. Because stakeholders prefer round numbers
   3. Because an LLM making up the last three digits creates false certainty that misleads decision-makers
   4. Because the lesson requires it as a formatting rule

5. What is the primary benefit of running multiple scenarios rather than a single baseline forecast?

   1. It eliminates the need for human judgment
   2. It produces a single definitive number for planning
   3. It surfaces which assumptions are most critical by comparing results across different conditions
   4. It reduces the complexity of the underlying model

6. When running scenario analysis, what should be done with assumptions that cannot be verified from available data?

   1. Replace them with industry average values
   2. Flag them as unverifiable so they can be explicitly debated
   3. Ignore them to keep the analysis clean
   4. Assume they are correct and proceed

7. The lesson emphasizes 'freshness over sophistication.' What does this mean in practice?

   1. Avoid sophisticated modeling to keep things understandable
   2. Prioritize getting any answer quickly over getting the right answer
   3. Re-run scenarios whenever key inputs are updated, rather than relying on outdated analysis
   4. Use simple spreadsheets rather than complex AI models

8. A capacity planning prompt asks for scenarios including baseline, +20% demand, -20% demand, a key supplier delay, and a product line discontinuation. What is the purpose of including both demand changes and disruption scenarios?

   1. To meet a requirement of five scenarios minimum
   2. To test both demand-side uncertainty and supply-side risk in the same analysis
   3. To confuse the model into providing more detailed outputs
   4. To ensure the final number is an average of all possibilities

9. What does the lesson identify as the fundamental nature of capacity planning?

   1. A mathematical optimization problem
   2. An assumption-surfacing problem
   3. A cost reduction initiative
   4. A demand forecasting exercise

10. If a capacity model shows that changing the lead time estimate by one day causes inventory requirements to change by 15%, what should a planner conclude?

    1. The model is broken and should be rebuilt
    2. Lead time is not an important factor
    3. The model needs more decimal places of precision
    4. The model correctly identifies lead time as a high-sensitivity input requiring careful estimation

11. A manager asks an AI to run capacity scenarios and receives an answer of '14,562 units needed, 3.27 FTE required, and $47,293 in capex.' Why might this response be problematic?

    1. The false precision of the decimal places and exact figures suggests the model is fabricating accuracy
    2. AI cannot handle capacity planning calculations
    3. The manager should have asked for ranges instead
    4. The numbers are too small to be useful

12. Why should scenario-sweep results be re-run when key inputs are updated rather than relying on the original sophisticated analysis?

    1. Original scenarios become invalid after any data change
    2. Sophisticated analyses are always wrong
    3. The lesson requires at least three updates per year
    4. Fresh analysis with updated inputs is more actionable than sophisticated analysis based on stale data

13. What does forcing a model to rank input sensitivity for every scenario accomplish?

    1. It identifies which assumptions most significantly drive outcomes under different conditions
    2. It automatically corrects erroneous inputs
    3. It reduces the need for human oversight
    4. It makes the model more complex

14. When a scenario sweep reveals that two different scenarios produce nearly identical results, what does this likely indicate?

    1. The model is highly accurate
    2. The scenarios may not be capturing the factors that truly drive variability
    3. The AI is malfunctioning
    4. The planning team should use the simpler scenario

15. In the context of capacity planning, what makes assumptions dangerous when left unexamined?

    1. They are always wrong
    2. They become locked in the spreadsheet and cannot be debated or updated
    3. They work correctly without examination
    4. They are visible to all stakeholders by default