For moulders — injection, compression and composites
For moulders, the tool is the asset, not the machine.
Production software for moulding shops: injection, compression and rubber. A press is a machine like any other, and most manufacturing systems model it perfectly well. What they get wrong is the mould — the thing that actually carries the part, wears out, moves between presses, and decides whether next week goes smoothly.
If you process composites rather than call yourself a moulder, this page is still about your floor. The argument below is about tools, cycles and capability, and it holds whether the press closes on a shot or a charge.
What general-purpose systems get wrong
None of these is a moulding problem. They are all modelling problems.
A system designed around machines and work centres can describe a moulding shop, in the way a spreadsheet can describe anything. These are the four places that description stops matching the floor.
The tool is a text field
The maintenance system has an asset record for the press and nothing for the mould. So a worn cavity gets logged against whichever machine it happened to be in, and the tool’s real history is spread across four press records and a toolroom whiteboard.
Service intervals run on the calendar
A tool that ran three shifts a week for a month and one that sat in the rack both come due in March. One is overdue and one is being serviced for no reason, and neither fact is visible until something splits.
The standard cycle is whatever it was in 2019
Every performance figure is measured against it, so an OEE number moves when the estimate is stale rather than when the process changes. Nobody trusts the report, so nobody reads it.
Cost per part is material plus a guess
Energy gets divided out of the site bill, overhead is a percentage somebody set years ago, and a four-cavity tool is costed as though it made one part a cycle.
Tools as first-class assets
A mould with a damaged cavity is not the press's problem.
The tool has its own record, its own history and its own status — in service, in repair, quarantined or retired. Not a text field on a machine, and not a line on a toolroom whiteboard.
- Faults are raised against a machine or a tool, so a worn cavity is recorded as what it is rather than as a fault on whichever press it happened to be in.
- Press maintenance and the toolroom get separate queues, with routing on category, machine, location and severity — a hydraulic leak and a worn cavity are not the same team’s morning.
- Cost rolls up by tool as well as by machine, which is how a mould that has quietly consumed four thousand pounds of repairs this year becomes visible.
- Repair parts are booked at a price snapshot taken when they were used, so last year’s tool repair still shows last year’s price after the supplier puts theirs up.
The full maintenance model, including the escalation ladder and QR fault reporting, is on Maintenance and mould care.
Maintenance on usage, not the calendar
Service intervals that count what the tool actually did.
This is the difference that matters, and it is the one a machine-centric maintenance module cannot express: the count belongs to the tool, and it travels with it.
- Cycles accumulate per mount and per lifetime, and follow the tool across machines: move a mould from one press to another and its count goes with it.
- Usage-based schedules raise their own work order at the threshold you set — “this mould is 4,000 shots from its 250,000 service” — with the due-soon warning arriving in time to plan it rather than discover it.
- Machines carry usage-based and time-based schedules side by side: run hours for the press, cycle count for the tool, months for the things that genuinely are seasonal.
- Service work orders carry a mandatory checklist, including measured values where you need them recorded, so a completed service is evidence rather than a tick.
- Counts come from the machine itself rather than from anyone writing them down — four ways onto a press, including one that needs nothing but a power clamp.
Cycle time and the stale standard
Performance measures your estimate, not your machine.
Ask any moulder why the OEE looks wrong and you get the same answer twice: the shutdown counted as downtime, and the standard cycle is whatever somebody typed in years ago. The second one is quieter and does more damage.
- Raw and capped performance are reported side by side rather than reconciled. Performance above 100% is not a fast press — it is a standard cycle slower than what the tool is really doing, and capping it silently hides exactly that.
- The actual cycle is measured, so the gap between the standard and the floor is a number on screen rather than an argument in a meeting.
- An empty window says “no production in this window” instead of a convenient 0% that drags the weekly average down.
- You can put a shift through it yourself before speaking to anybody: the OEE calculator does the same arithmetic with your figures and stores nothing.
The argument in full, and how a figure gets flattered on its way to a board pack, is in OEE in plain English.
Changeovers and preheat
A tool change is not a gap in the Gantt.
Scheduling a moulding shop means scheduling the tools and the preheat stations as hard as the presses. A planner that only knows about machines will cheerfully book two jobs that need the same mould, or the same preheater, at the same time.
- The Gantt knows a mould change needs a preheat station, and treats that station as the finite resource it is.
- Tools are scheduled as resources in their own right, so two jobs cannot quietly claim the same mould.
- Draw how a part is made in the process designer and the drawing runs the job — the designer says how, the Gantt says when.
- Versions carry effectivity dates, so a works order raised last month still resolves the process that was in force last month, and a two-year-old job stays reproducible after two process changes.
More on scheduling and the process designer: Process designer, scheduling and tool changes.
Capability on special characteristics
Cp and Cpk with the assumptions named on screen.
Tier supply brings PPAP and IATF-style regimes with it, and a capability figure is only worth as much as the estimator behind it. Most systems print one number and leave you to guess which.
- Cp and Cpk are reported alongside Pp and Ppk, with the sigma estimator named on screen for each — within-subgroup or overall. Two honest numbers rather than one convenient one.
- Control limits are frozen from an audited baseline study, not recomputed on page load. A process drifting out of control cannot quietly redraw its own limits to make itself look in control.
- A Cpk watchlist ranks characteristics worst-first, so the weekly review starts where it should.
- Specifications go through a review and approval workflow in which the approver cannot be the author — the segregation an IATF 16949 audit expects to see enforced rather than promised.
- Every measurement is stamped with the specification revision in force when it was taken, so a two-year-old Cpk is still reproducible after two spec changes.
The quality module in full: Quality, RFT and SPC.
What a moulded part actually costs
Cavities multiply parts, not energy.
Cost per part on a moulded part is a composite — energy plus material, labour and overhead, tied to the works order. Energy alone is the figure competitors publish, and on a multi-cavity tool it is not even the interesting half.
- A four-cavity tool makes four parts a shot, and the cost per part reflects that rather than dividing a cycle’s energy by one.
- Material is valued at what it actually cost when it moved — the price on the movement, then the lot it came from, then standard cost — and each line names which basis it used, so a standard-cost guess can never dress up as a measurement.
- Labour comes from hours booked at the job at the rate in force when they were booked: setup at the setup rate, running at the production rate.
- Overhead pools are defined in your own terms — factory rent and rates, supervision, moulding-shop burden — each spreading by the driver that suits it, so every job can show which £0.31 of burden landed on it and why.
- Energy, material, labour and overhead are shown separately and summed. Nobody has to take the total on trust.
The costing model in detail: energy monitoring software and your cost per part. If your site holds a Climate Change Agreement, the same kWh-per-kilogram number is the one you are assessed on: Climate Change Agreements in plain English.
Process coverage, stated plainly
Injection, compression and rubber.
Nothing above is injection-specific. A tool that carries its own maintenance history and accumulates its own usage count is the same idea whether the press closes on a shot or a charge, and whether the variable you watch is barrel temperature or cure time and platen temperature. Where this page says “shots” it means the unit your floor counts in — cycles, shots or pressings.
If your material is a composite and your identity is a composites processor rather than a moulder, the modelling is unchanged: the tool is still the asset, the count still follows it between presses, and capability is still reported with its estimator named.
The honest limit: SmartFactory has one live production deployment in moulding. That is one, not a customer list, and we would rather say so than imply a base we do not have. What the software does is described above and on the module pages, and every claim is demonstrable in a trial on your own floor.
Have you connected one of ours?
It is the only technical question a moulding shop actually asks, and it deserves an engineer rather than a brochure. Tell us what is on your floor — the presses, their age, what the controllers are, and whether anything already talks to anything. You will get a straight answer about what connects today and what would need a clamp.