Build a production tracker
How to build a production tracker with AI
To build a production tracker, describe who it is for and what they do in it, and let Forge64 plan and generate the app. A production system of record: items, versioned multi-level BOMs and routings, work orders released to work centers, operator reporting of quantity/scrap/time per operation, lot and serial genealogy for recall tracing, in-process and final inspections with nonconformance and CAPA, downtime and OEE, and an ingest endpoint for machine counters. The core parts of a first version are plants and sites, shop-floor roles and pin clock-in, items, boms and routings, engineering change control and work orders and operation reporting.
Last checked
What the first version needs
These are the parts a production tracker needs before it is useful. Each one is something the plan covers.
- Plants and sites: Separate plants, lines and shifts under one company with site-scoped data and roles.
- Shop-floor roles and PIN clock-in: Operators on shared tablets act under their own identity without long passwords.
- Items, BOMs and routings: Define what is made, from what, and by which operations.
- Engineering change control: Change a BOM or routing safely with approval and effectivity.
- Work orders and operation reporting: Plan production and capture what actually happened per operation.
- Material inventory: Know what is on hand by item, lot and location without negative stock.
- Inspections, NCR and CAPA: Stop bad material, record why, and drive corrective action to closure.
- Downtime and OEE: Measure availability, performance and quality by machine and shift.
- Production audit trail: Prove who changed a BOM, released an order, overrode a hold or edited a report.
Parts you can add in the first version
- Machine counter ingestion: Accept production counts and states from gateways or PLC bridges safely.
- Supervisor and quality queues: Work lists for holds awaiting disposition, overdue CAPAs and late work orders.
- Traceability and production reports: Batch records, yield, scrap and on-time delivery reports as PDF/CSV.
The data it stores
Each table below becomes part of the Postgres database that belongs to your app.
- sites: work centers, lots and orders are scoped to a site; timezone drives shift boundaries.
- work_centers: kind: machine|cell|manual_station|inspection; status: active|down|maintenance.
- items: kind: raw|component|subassembly|finished|consumable; uom conversions live in a uom table keyed by item.
- boms: status: draft|in_review|released|obsolete; released rows immutable; lines in bom_lines (component_item_id, qty_per, scrap_pct, position) with the version.
- bom_lines: quantities are decimal strings with item UOM precision; multi-level explosion is a bounded recursive query with cycle detection.
- routing_operations: belongs to a versioned routing; run time stored in hundredths of a minute as an integer.
- work_orders: status: planned|released|in_progress|on_hold|completed|closed|cancelled; bom_id and routing_id are frozen at release.
- work_order_operations: status: pending|ready|running|paused|done|skipped; an operation becomes ready when the previous one reaches its transfer quantity.
- operation_reports: insert only; corrections are reversing reports with a reason; kind: start|pause|resume|good|scrap|rework|complete.
Pages and endpoints
| Route | Who can use it | What it does |
|---|---|---|
| /floor | authed | kiosk view of the operator's work center queue with start/stop and quantity entry. |
| /orders | authed | work order list, release and status board. |
| /orders/[id] | authed | operations, materials, reports and batch record for one order. |
| /engineering/boms | authed | BOM and routing editor with versions and engineering change requests. |
| /inventory | authed | stock by item, lot and location with movement history. |
| /quality | authed | inspections, NCR board, holds and CAPA tracking. |
| /trace | authed | forward and backward lot trace with recall report export. |
| /dashboards/oee | authed | OEE, downtime Pareto and scrap by work center and shift. |
| /api/operation-reports | authed | operator report; client report id gives idempotency; offline queue replay supported. |
| /api/lots/[id]/hold | authed | place or release a hold; quality role, audited, signature on release. |
| /api/devices/ingest | paid | signed batch ingestion from gateways using a hashed per-device key; rate limited per device. |
What to leave out of version one
- No machine control, PLC programming, interlocks or safety functions.
- No finite-capacity optimiser or APS in phase 1; manual sequencing with warnings.
- No deleting or editing of released versions, reports, movements or genealogy.
- No per-operator productivity ranking by default.
- No claims of validated or certified status; validation is the operator's responsibility.
- Finite-capacity scheduling: plan for it, build it later.
How Forge64 builds it
Forge64 matches your description to one of 25 reference architectures, writes the plan and data model into the project, generates the app, and runs a quality gate over the code before you see it. Each app gets its own Postgres database and sign-in with no per-user fee. Hosting on Forge Cloud and Forge Payments are rolling out.
This is a reference design for the category, not a promise that every item is generated today. Built-ins such as background jobs, file uploads and search are rolling out; email sent from your app is coming soon, so anything here that needs it is something you add yourself for now.
FAQ
- What does a production tracker need first?
- Plants and sites, shop-floor roles and pin clock-in, items, boms and routings, engineering change control, work orders and operation reporting and material inventory. The plan for a production tracker covers all of these before it generates code.
- What should I leave out of the first version of a production tracker?
- No machine control, PLC programming, interlocks or safety functions. No finite-capacity optimiser or APS in phase 1; manual sequencing with warnings. No deleting or editing of released versions, reports, movements or genealogy.
- Which products define the production tracker category?
- Well-known examples include Tulip, Plex MES, Odoo Manufacturing, MRPeasy and Katana. Use them to decide which features your users expect, then describe your own version.
Build your app with Forge64
One prompt to a running app with auth, database, hosting and payments. Start free, no credit card required.
Start building for free