Skip to content
1961 Digital TransformX
العربية

Industry

Your cost per part assumes a cycle nobody achieves

Moulding is arithmetic: cycle time, cavities, shot weight, changeover. Every one of those is recorded as a standard somewhere and none of them is measured against reality, so the costing that wins your tenders is a hope rather than a figure.

Book a consultationAll industries

What the day actually looks like

  • The 250-tonne machine runs eleven seconds slower than the standard, and has done since the tool was last repaired.

  • One cavity in an eight-cavity tool has been blocked off for a month. The costing still divides by eight.

  • Regrind from yesterday's purge goes back into the hopper at whatever ratio the operator judges reasonable.

  • Changeover from a thin wall to a thick one takes a full shift, and the schedule was written as though it took an hour.

  • A customer asks which resin lot is in the parts he received in March, and the answer is somewhere in a shift logbook.

Where it usually breaks

  1. Standards nobody re-measures

    Cycle time and shot weight were set when the tool was new. The tool has since been repaired twice and the machine is older. Nothing forces the standard to be revisited, so the routing says forty seconds while the floor runs fifty-one, and the gap becomes a variance that finance attributes to labour.

  2. Cavitation treated as fixed

    Tools lose cavities. A hot runner blocks, a pin breaks, and the tool keeps producing at reduced output because stopping it costs more. But the costing, the schedule and the delivery promise all still assume full cavitation, so the plant quietly runs a shortfall it discovers at dispatch.

  3. Regrind and scrap off the books

    Sprues, purge, startup shots and rejected parts are ground and reused, which is correct practice. What is missing is the accounting: how much regrind exists, at what ratio it was added, and to which lot. So material yield cannot be calculated, and a customer complaint about part strength cannot be traced to a batch.

  4. Changeover is invisible capacity

    Planning treats the machine as available when it is being cleaned, purged, colour-flushed and requalified. Colour sequencing is decided by whoever writes the board, usually light to dark by instinct. The lost hours never appear as a number, so nobody can argue for a schedule that would save them.

  5. Batch traceability by logbook

    Resin lot, masterbatch lot and additive are recorded on paper at the machine, if at all. When a customer or a regulator asks what went into a shipment, the answer is reconstructed from shift sheets and goodwill. In regulated food-contact or automotive work, that answer is not acceptable.

How the work runs

  1. Re-baseline every routing

    We measure actual cycle, actual cavities running and actual shot weight per tool and machine combination, and make those the standard. Then we make the standard something that ages: if the floor runs outside it for a defined period, the routing is flagged for review rather than silently absorbed as variance.

  2. Cost by weight, honestly

    Part cost is built from shot weight and material price, plus machine rate by tonnage, plus the changeover share the part actually caused. Runner and sprue weight is charged where it belongs, and regrind enters at a recovery value you set, not at zero and not at virgin price.

  3. Material genealogy at the machine

    The resin lot, masterbatch lot and regrind lot loaded into the hopper are recorded as a consumption event on the production order. Traceability then runs both ways: from a delivery note back to the lots, and from a suspect lot forward to every customer who received parts made from it.

  4. Schedule the mould, not just the machine

    Planning is done against the tool as a constrained resource with its own maintenance interval and shot counter, alongside machine tonnage and dryer capacity. Changeover time is modelled per transition, including colour, so the sequence the planner chooses has a visible cost.

  5. Scrap with a named cause

    Rejects are booked with a defect reason at the machine: short shot, flash, sink, contamination, colour streak. The reason codes are few enough that an operator will use them and specific enough that they point at a tool, a dryer or a resin lot when the pattern appears.

What changes once it holds

  • Quoted cost per part reflects the cycle the plant actually runs.
  • A blocked cavity changes the plan the day it is blocked.
  • Material yield is calculable, including regrind and runners.
  • A resin lot can be traced to every shipment it entered.
  • Changeover hours appear as capacity, so sequencing can be argued.
  • Scrap has a cause code that points at something fixable.

Weight is the only honest unit here

Moulding costing goes wrong when it is done per piece. A piece is an output; what you actually buy and consume is kilogrammes of resin and masterbatch, and what you actually sell is machine time on a tool. Once the model is built on shot weight, runner weight, regrind ratio and machine tonnage rate, the numbers reconcile to the material you purchased and the hours the machines were switched on. That reconciliation is the control. If the kilogrammes bought, the kilogrammes shipped and the kilogrammes still in the yard do not agree, something is wrong in the plant and you now know it in the same month rather than at stocktake.

Regrind is inventory, not a rounding error

Most plants regrind, and most plants do not account for it. That creates two separate problems. Commercially, the material yield figure is meaningless, so nobody can tell whether a job that ran long consumed more than it should have. Technically, the regrind ratio affects mechanical properties, and a customer who finds a batch of brittle parts will ask what proportion of reprocessed material was in them. Treating regrind as a real stock item with a lot number, a source and a permitted ratio per product answers both questions and costs the operator one entry per changeover.

Planning around tools, dryers and colour

A plant with twenty machines does not have twenty independent capacities. It has machines, tools, dryers, robots and a colour sequence, and any one of them can be the binding constraint on a given day. Scheduling on machine availability alone produces a plan that collapses the moment two orders need the same tool or the same dried resin. Model the tool as a resource with its own shot counter and preventive maintenance point, model the changeover as a transition cost that depends on what ran before, and the plan starts surviving contact with the shift. It also makes the case for a second tool an arithmetic argument rather than a request.

Questions we get asked

Do we need machine monitoring hardware for this?
Not to start. Counting good and rejected parts per shift with a reason code gets you most of the value, and it works on machines of any age. Where a plant already has monitoring, we take the cycle and downtime data from it rather than asking anyone to retype. Adding sensors is a later decision we would rather you make after you can see what the manual figures already show.
How do you cost a part that runs on two different machines?
By the combination, not by the part. The same tool on a larger machine has a different rate, sometimes a different cycle and a different energy cost. We hold a rate per machine class and a routing per tool and machine pair, so the cost reflects where it actually ran. The quoted cost uses the machine you intend to run it on, and the variance tells you when it did not.
Our operators will not record scrap reasons honestly.
They will if the list is short and nothing punishes them for using it. We keep the reason codes to a handful that map to real defects an operator can see at the press, and we do not build reporting that names individuals. The purpose is finding the tool or the dryer that causes the pattern, and operators generally know which one it is before the report does.
Can this handle food-contact and regulated work?
Yes, and the requirement shapes the design. Lot genealogy, declaration of compliance documents held against the resin grade, and a recall path that runs from a lot to the customers who received it are built in rather than added. We do not certify you; that is your certification body. We make sure the records the auditor asks for exist without a week of preparation.
We already have an ERP. Is this a replacement?
Often not. In most plants the finance and purchasing side works and the production side was never configured properly, so the fix is a manufacturing implementation on the system you own plus the governance around it. We say so when that is the case, even though a replacement would be worth more to us. The diagnosis states which of the two we found, with reasons.

What this industry usually needs

Tell us what is not working

Send one line about what is going wrong. On the first call we will tell you whether it is a system problem, a process problem or a governance problem — and which one to fix first. That call is free and it is not a sales meeting.

Book a consultation