Gains:
- Can prepare the input to be given to artificial intelligence for pre-feasibility
- Can inspect the result with a critical eye
- Can make the account available to the customer
Pre-feasibility is a repetitive chain of calculations: from capacity to production, from production to revenue, from revenue to payback time. Artificial intelligence establishes this chain quickly; If the inputs are correct, the output will be consistent.
Inputs you need to give
- Target capacity (tons/24 hours) and daily working hours assumption.
- Target flour type and expected yield.
- Wheat purchasing price and flour/bran selling price (regional).
- Energy unit price and estimated installed power.
- Number of personnel and cost.
- List of investment items (from the unit above).
Requested output
Annual production, annual income, annual operating expenses, gross profit, simple payback period and sensitivity analysis (what happens if wheat price changes ±10%, yield changes ±1 point).
Audit: where did the number come from
Artificial intelligence can make assumptions while doing calculations. Ask the source of each number: "where did you get this energy consumption from?" If it cannot answer, that row is removed from feasibility or replaced with your data.
Don't make up prices and don't let AI do it. If there is an unknown item, write "to be determined"; One made-up number contaminates all subsequent calculations.