MANUFACTURING / JAPAN SOURCING

Custom machinery: compare throughput with a defined workpiece and time window

A maximum cycle rate and good output over an agreed trial answer different questions. Define the workpiece, time boundary, stops and quality count before comparing machine capacity claims.

Ask what the quoted rate counts

A buyer compares dedicated assembly machines for a small component. One offer states a maximum of 60 pieces per minute. Before multiplying that by a shift, ask whether it means cycles, attempted pieces or accepted parts, and under which workpiece and feeding conditions.

Vorne distinguishes total, good and reject counts, run and stop time, and ideal cycle time in its production-counter explanation.[1] Those distinctions help frame an enquiry; they do not define your acceptance contract. Agree the output point and whether reworked or repeatedly processed pieces can enter the count.

Throughput needs a part, a clock and a quality count

1Workpiece: identity, variants, input condition and feed
2Time window: start/end and included stops or changeovers
3Output: total attempts, first-pass good parts and rejects

The component, segmented time bar and output bins are schematic. The segment lengths and pictured parts are not trial data or a capacity claim.

Keep the denominator visible

Here is an invented 20-minute observation: the machine runs for 16 minutes and stops for four; it produces 960 attempted pieces, of which 900 pass first inspection and 60 fail. Total output per running minute is 960/16 = 60. First-pass good output over the whole observation is 900/20 = 45 per minute.

Both rates describe this example, but neither may silently replace the other. State whether setup, replenishment, small stops, faults and changeovers are included in the agreed window, how events are recorded, and how unfinished work at the boundaries is treated. The example sets no universal trial length, pass rate or availability target.

Invented observation: same machine, different rate definitions
Record	Example value	Meaning
Observation window	20 minutes	Defined start/end, including the stops in this example
Run / stop time	16 / 4 minutes	Recorded as distinct states
Total attempts	960 pieces	Counted at the agreed output point
First-pass good / reject	900 / 60 pieces	No reworked pieces counted as first-pass good
Total per running minute	60 pieces/minute	960 divided by 16
First-pass good per observation minute	45 pieces/minute	900 divided by 20

Make the actual work and resources part of the claim

Give the supplier representative drawings/parts, variant mix, incoming variation, presentation/feeding, quality criteria and the expected operator tasks. Identify utilities, upstream supply and downstream handling that can constrain the proposed cell. Ask which conditions support the offered rate and which remain assumptions.

JUKI’s contract-business description separates selectable manufacturing stages and production-line engineering.[2] For this enquiry, make the machine boundary explicit: a station, a cell with feeders and inspection, or a wider line. A single station’s fast cycle does not establish the output of a line with other constraints.

Agree a reproducible capacity trial before ordering

Define the workpiece set and quantities, machine/software configuration, staffing, replenishment, test duration, counting rules, stop categories and quality checks. Record the acceptance thresholds and who can approve deviations or a repeat. Keep raw counts, time records and inspection results with the test report.

If factory and installed conditions differ, specify what will be demonstrated at each location and what remains to be confirmed after integration. The packaging-machine FAT/SAT guide covers that division for its specific application. This article defines the capacity question; it does not prescribe one acceptance protocol for every machine.

Questions to send your supplier

Open supplier questions

Product/process: dedicated assembly machine
Workpiece/drawing revision, variants, input condition and feeding:
Machine/cell/line boundary, utilities, staffing and linked equipment:
Quoted rate: cycles, total attempts or first-pass good output?
Output counting point and rework/unfinished-work rules:
Trial start/end, duration and included setup/stops/changeovers:
Quality criteria, inspection and accepted/rejected count:
Please identify conditions behind the rate, assumptions and remaining trials.
Agree thresholds, raw records, decision owner and repeat/deviation rules.
Factory/site demonstration scope, destination and timing:

Save questions ↓

Sources

Vorne — Production counters
Distinctions among total/good/reject counts, run/stop time and ideal cycle time. Our 20-minute example is invented and is not an OEE standard or contractual target.
JUKI — Contract business
Manufacturer describes selectable manufacturing stages and production automation/line engineering. Not confirmation of capacity or support for the hypothetical machine.