Excel Solver Model Assumptions and Result Sanity Audit
Check that an optimisation model is sound before acting on it.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Review a Solver model I built. Objective, changing cells and constraints: [MODEL_SETUP]. Input values and where they come from: [INPUTS]. The answer Solver gave: [SOLVER_RESULT]. The decision it supports: [DECISION]. Check that all real limits are included (budget, capacity, minimum or maximum quantities), whole-number needs, units, and that constraints do not contradict each other. Test the answer by hand for feasibility with arithmetic, for example 30 units of A and 10 of B using 100 hours: 30 x 2 + 10 x 4 = 100. Say what should happen if one input changes slightly, and look for unrealistic results at the edges, such as zero of something we must make. Ask for anything missing before drawing conclusions. Provide a table of assumptions (Assumption, Source, Confidence, Effect if wrong) ranked by sensitivity, and three scenarios to test: costs up 10 percent, capacity down 10 percent, and demand lower than planned, with how the answer should respond. Return findings, which constraints might be missing, and three further tests. A model supports a decision but does not make it.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.
See an example result (fictional data)
Invented check. Solver says make 30 units of A and 10 of B using 100 hours: 30 x 2 + 10 x 4 = 100. Profit = 30 x 30 + 10 x 50 = 1,400. Test: if hours fall 10 percent to 90, expect profit to fall; scaling gives about 1,260. Missing constraint noticed: raw material limit not entered. Fix and retest: add a constraint that raw material use cannot exceed the 80 units available. If product A uses 1 unit and B uses 3, then 30 + 30 = 60 units, which fits. Sensitivity: raising hours to 110 should increase profit; if it does not, a constraint other than hours is binding. Record the assumptions in a table beside the model.
Illustrative only, made-up data. Your results will differ and must be reviewed.