MRP-Lite Make/Buy Suggester: Explode Demand, Net Stock, Plan Orders

The problem this kills

Every week, somebody on your team rebuilds the same monster spreadsheet. They paste in next month's demand, line up the bills of material, subtract what's on hand, subtract what's already on order, and then squint at the result trying to figure out: what do we need to make, what do we need to buy, and by when do we have to start so it actually arrives in time?

It is the most error-prone job in the plant. One stale on-hand number, one BOM that changed last quarter, one open PO that got missed, and you either over-order parts you don't need or — worse — discover a shortage on the day the line is supposed to run. The math is genuinely hard: a single finished good explodes into dozens of components, each with its own lead time, each needing to be back-scheduled from the date you actually need it. Do that across a few hundred SKUs by hand and mistakes are guaranteed.

This is exactly the calculation a real ERP's MRP module does — and exactly the calculation most small and mid-size manufacturers don't have, because the ERP is too expensive, half-implemented, or its MRP was never turned on. So it stays in the spreadsheet, and it stays fragile.

What you'll build

A simple internal web app for your planners and buyers. You import five things — finished-goods demand (what's needed and when), your bills of material (what each product is made of), your on-hand stock, your open purchase orders and work orders, and your lead times per item. You don't need a working ERP connection; CSV or a Google Sheet is enough.

The tool does the heavy lifting. It explodes the demand down through every level of your BOMs, nets the gross requirement against what you already have on hand and what's already on order, and produces a suggested planned order for everything that's short — marked clearly as make or buy, with a release date back-scheduled from the need date using that item's lead time, and respecting your lot-sizing rules. Shortages and negative balances are shown in red so nothing hides.

Then the part that matters most: the planner reviews the suggested planned orders, approves the specific ones to convert, and only then does the tool export a clean planned-order CSV in the exact columns your purchasing or production system expects. The AI suggests; a human decides; nothing leaves the building until a person signs off.

What's inside the Implementation Plan

The downloadable plan is a step-by-step file you paste into an AI coding agent. It opens by interviewing you about your business — how you plan today and who does it, what systems and spreadsheets and ERP you're juggling, how your SKUs and item codes and BOMs are actually named and structured, your typical and peak volumes, your real lot-sizing and lead-time rules, and the messy exceptions (phantom assemblies, substitute parts, safety stock, scrap allowances, multi-level BOMs). It reflects a short tailored spec back to you and waits for your thumbs-up before it builds anything — so the tool matches your plant, not a generic template.

From there it walks the agent through the data model, the five imports, the BOM explosion engine, the netting logic, the lead-time back-scheduling and lot sizing, the shortage flagging, the planner approval gate, and the planned-order CSV export. Every step ends with a ready-to-copy prompt. There's a full "No API yet?" path that uses Google Sheet / CSV imports as the data source and produces a clean planned-order CSV — so you can build and run the whole thing this weekend no matter what's on your ERP roadmap.

The governance it includes (this is the point)

This tool drives what your business buys and builds, so the controls aren't optional. The plan builds in login so only your team can use it, row-level security so you only ever see your own organization's data, a complete audit trail of who imported what, who changed which assumption, and who approved which planned orders, a hard human-approval gate so no requisition or work order is ever exported until the planner reviews and approves the specific suggestions, and duplicate guards (keyed on item + need date + source) so the same requirement can't be planned twice when you re-run or re-import.

Who it's for

Planners, buyers, and production schedulers at small and mid-size manufacturers who are doing material-requirements planning in a spreadsheet because their ERP has no MRP, or no ERP at all. If you redo the same explode-and-net spreadsheet every week and live in fear of a missed shortage, this is for you. You don't need to be a developer — you need to know how your plant plans, and the plan handles the rest.

You've got this — paste the first prompt and let the agent interview you.