Plans, sampling, NCRs, and CAPA.
Quality isn’t a binder; it’s a routine. This page defines audit plans, sampling, witness points, and NCR control that keep the installed system aligned with the tested configuration. You’ll verify coatings, wiring, foundations, and controls against evidence standards, proving HVM bollard and crash rated bollard performance before, during, and after SAT. Include one-sentence context that naturally links upward to the parent hubs (this section and the chapter hub). Add SIRA context with a link to SIRA Bollards (UAE) when relevant. Link installation pages only if helpful: What to Expect and Installation Guide.
Important: This is a general guide. For live projects we develop a tailored
Method Statement & Risk Assessment (MS/RA) and align with authority approvals (e.g., SIRA) where in scope.
858.1 Audit plan
Schedule vendor/site audits (848, 611). Assurance for HVM bollard delivery.
An audit plan sets the cadence, scope, and roles for planned reviews across factory and site. Start by mapping milestones from prequalification (848) through pre-works setup (611). Define which activities are witness points and which are hold points, plus how to escalate findings to the RACI owners.
Keep scope aligned to the rating-critical dependencies (421): foundations, reinforcement, embedment, and controls. Tie each audit to required evidence types (photos, measurements, serial capture) so that SAT witnesses (638) can trace what was built to what was tested.
Choose a realistic audit cadence (e.g., weekly site walks, monthly vendor checks). Record risks and owners (846) so issues are closed before concrete or cabling is concealed.
| Aspect | What matters | Where to verify |
|---|---|---|
| Performance | Tested system (bollard + footing) | Crash ratings overview |
| Operations | Duty cycles, fail-state, safety devices & measures | Installation: What to Expect |
858.2 Checklists
Tie to ITP points (714) and rating dependencies (421). Evidence-based.
Build checklists from the ITP (714) so every hold/witness step has a pass/fail criterion and an evidence type. Examples: sleeve plumb tolerance and grout level, rebar cage position, conduit sealing, and controller wiring labels. Include accept/reject bands for DFT, torque values, and clear-gap checks.
Cross-link each item to supporting pages such as Coatings (362) and Datum & alignment (626). Store blank forms under Witness & inspection forms (918) to standardize capture.
858.3 Sampling & witnesses
Hold/witness coverage (638). Audits verify crash rated bollard behavior.
Use risk-based sampling. Critical safety items (e.g., safety relays, interlocks) may require 100% witnessing, while cosmetic items can be sample-checked. For embedded works, set pre-concealment hold points so rebar, depth, and socket embedment are confirmed before pour.
At SAT, follow the agreed witness script (638) and record timestamps, testers’ names, and environmental conditions. Triggered sampling (e.g., after a fault burst) increases coverage where data suggests risk.
858.4 Nonconformance control
Log, disposition, corrective actions (719). NCRs closed fast.
Raise an NCR when work deviates from drawings, specs, or standards. Each NCR needs a clear disposition (rework, use-as-is with concession, or scrap), photos, and responsible owner. Link to Nonconformance & defects (719) for root-cause categories and evidence thresholds.
For systemic issues, open a CAPA and verify its effectiveness over time. Update the ITP and training where procedures caused the error, and use the verification of effectiveness log to confirm closure.
858.5 Supplier QA
Coatings, welding, assembly (362, 315). Factory quality flows to site.
Vendor audits should verify coating systems (362), weld procedures, heat-number traceability, and assembly torques. Confirm test certificates and authenticity statements. Where variants exist, ensure the delivered product matches the family/variant evidence (415).
Document serials for bollards, HPUs, and controllers into the asset register (732) to preserve traceability into operations and maintenance.
858.6 Calibration & tools
Meters/gauges in date (519). Trust the numbers.
Before testing, confirm calibration expiry for torque wrenches, multimeters, pressure gauges, and loop testers. Record instrument IDs in test sheets and capture photos of displays where readings determine pass/fail. Use Power test points (519) and controller diagnostics for repeatable measurements.
Quarantine out-of-date tools and re-test affected work where bad instruments were used, noting any expanded sampling required.
858.7 Records retention
911 naming, 739 archive. Traceable HVM bollard history.
Use the File Index & Naming Rules (911) for consistent filenames (site ID, date, step, reviewer). Store the signed witness forms, photo logs, and redlines in the closeout archive (739) with a clear custodian. Include filename conventions and checksums for integrity.
Where approvals apply (e.g., SIRA in the UAE), ensure submission packs are organized to the authority’s sequence and retained for the specified period.
858.8 Trends & CAPA
Analyze defects; drive CAPA (844). Quality improves.
Aggregate findings into a monthly trend: top defect categories, affected locations, and suppliers. Use a simple Pareto to focus on the few causes driving most defects, then open CAPAs and assign owners. Align with Common pitfalls & lessons (844) so teams learn across projects.
Follow through by updating checklists, toolbox talks, and the ITP. Re-sample after CAPA to confirm improvement; if not effective, escalate and adjust scope or design.
858.9 Independent reviews
Third-party spot checks (538). Confidence for reviewers (717).
Appoint an independent reviewer to sample evidence quality and observe key witness points. This can be a third-party acceptance test (538) observer or an Owner’s Engineer. Their notes should align with the authority submittals pathway (717) and feed into the CAPA loop if gaps are found.
Share a short independent summary with stakeholders to strengthen confidence before handover.
Related
External resources
- BSI: Impact test specifications for Vehicle Security Barriers
- ASTM F2656: Vehicle impact performance of perimeter barriers
- ASIS: Security Risk Assessment Standard
858 Quality Audits & Surveillance — FAQ
What’s the difference between a hold point and a witness point?
A hold point stops work until a reviewer signs off; a witness point allows work to proceed but must be observed and recorded. Both should appear in the ITP with clear pass/fail criteria and evidence to collect.
How many SAT tests should be witnessed?
Witness all SAT safety and performance tests that prove rating-critical dependencies (foundations, clear-gaps, controls). Routine operational cycles can be sample-witnessed, but pre-concealment checks and safety functions are typically 100%.
When do I raise an NCR versus a punch-list item?
Use an NCR for any deviation from drawings, standards, or specifications. Punch-list items are minor defects after substantial completion. If a punch-list item affects safety or rating dependencies, convert it to an NCR.
What proves that a CAPA actually worked?
Define verification of effectiveness up front—e.g., zero repeat defects for two audit cycles, or measurements within tolerance over a set period. Trend results and, if targets aren’t met, escalate and adjust the action plan.