PlatformQuality & SPC
Know — quality, RFT & SPC
Quality management: capability to show a customer, not a claim.
Quality software is easy to sell and hard to get right, because the difficult part is statistical rather than visual. Inspections start with a scan and end in charts your auditor recognises — with the statistics done properly, the assumptions named on screen, and every measurement tied to the specification revision that was in force when it was taken.
Inspections
A scan starts the check, and the data lands clean.
Quality data goes wrong at the point of entry, not in the report. So the system starts where the part is, with a scanner rather than a keyboard, and refuses a bad scan there instead of carrying it forward.
- Inspection sessions by customer, part, inspector, shift and cell — built around a barcode scanner, because typing part numbers is how bad data gets in.
- Right First Time, rework or scrap dispositions with configurable rework destinations and multi-select defect types.
- Batch validation happens on the scan: a mis-scan or an already-consumed batch is refused at the point of scanning, not discovered in next month’s report.
- Inspection requirements are defined per product type with sampling plans, so “did we actually take the samples we said we would” is a question with an answer.
Right first time
Find where the scrap actually lives.
- RFT against target, broken down by operator, shift and part — the three cuts that turn “quality is down” into a conversation with a next step.
- Defect Pareto and per-part defect heatmaps, so the dominant defect and the machine it belongs to arrive together.
- Weekly trend lines that show when quality changed, which is usually a better question than by how much.
- Scrap events tie back to the job, machine and shift that produced them — the same events that pull down the Quality term in your OEE.
SPC
Control charts that behave the way the textbook says they should.
SPC is easy to draw and easy to get wrong, and a chart that looks right is the easiest thing in the world for software to produce. The failure modes are always the same: limits recomputed from the data being judged, capability quoted without saying how sigma was estimated, and a chart drawn from six points.
Limits frozen from baseline study BS-114 (25 subgroups, n = 5) — not recomputed on page load
Both estimators are named on screen, and every measurement carries the specification revision in force — so a two-year-old Cpk still reproduces after two spec changes.
- X̄-R charts for subgroups and I-MR charts for individuals, using the standard AIAG constants — the same tables your quality engineer already has in a folder.
- 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.
- Cp and Cpk 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, and what each index actually measures if the distinction is new.
- Nelson rules 1 to 5 flag the patterns that matter, marked on the chart at the point they occurred rather than as a note underneath it.
- Minimum-data guards throughout: below the accepted subgroup count the system says the study is provisional instead of drawing you a confident-looking chart from thin air.
- A Cpk watchlist ranks characteristics worst-first, so the weekly review starts where it should.
Specifications
An approval trail that satisfies the auditor without a filing cabinet.
- 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, not promised.
- Every change is versioned with who wrote it, who approved it and when, and the trail is immutable.
- Specs overdue for review demote themselves automatically rather than sitting approved for four years because nobody had a reminder.
- Every measurement is stamped with the specification revision in force at the time it was taken — so a two-year-old Cpk is still reproducible after two spec changes.
- Labels, inspections and specifications point at each other: a label traces to the check that consumed it, and that check traces to the spec it was judged against.
- Full lot and serial genealogy connects every inspection to the material it came from — recall answers in one screen, not one weekend.
- The same discipline governs how parts are made: a released process is a versioned, effectivity-dated document with its own approval trail, so the routing a job ran against is as reproducible as the spec it was judged against.
This is quality management software for a production floor, and it says where it stops. SmartFactory covers in-process inspection, RFT analytics, variable control charts and capability. Gauge studies, PPAP and FMEA documentation remain in the systems that own them.
See your factory live in thirty minutes.
Book a demo and we’ll walk you through a running factory — or register for a free trial and get a simulated factory of your own to play with, the same day. No hardware, no card, no IT project.