Seaco Beverages Brought 6 New Formats Live on ASFL Vacuum Sealer OEE
Seaco Beverages Site A brought six formats live on the ASFL vacuum sealer, moving OEE from 62% to 74% in 10 weeks across N=2 lines. Changeover moved from 48 to 23 minutes, FPY from 94.1% to 98.2% on N=52 lots, and energy per pack from 0.18 to 0.15 kWh in the same window. Methods were SMED parallelization, recipe locks for torque and vacuum setpoints, and airflow re-zoning at the sealer tunnel. Evidence anchors include ISO 13849-1 Performance Level d on the safety PLC and SAT record SAT-25-042 referencing GS1 aggregation. Payback modeled at 8.5 months on blended CapEx and OpEx, with MTBF from 3.2 to 7.1 hours and leak rate trending under 250 ppm. Validated during PQ with traceable lots.
| Metric | Baseline | Result | Window / N | Notes (CI/Range) |
|---|---|---|---|---|
| OEE | 62% | 74% | 10 weeks; N=2 lines | ±2.0 pp across weeks |
| Changeover | 48 min | 23 min | 10 weeks; N=36 events | SMED parallelization |
| FPY | 94.1% | 98.2% | N=52 lots | Print-and-verify gating |
| Leak rate | 760 ppm | <250 ppm | N=52 lots | Vacuum window control |
| kWh per pack | 0.18 | 0.15 | 10 weeks; N=2 lines | Air re-zoning |
| MTBF | 3.2 h | 7.1 h | N=114 stops | Spares + PM kit |
| MTTR | 14 min | 14 min | N=114 stops | Unchanged |
| Alarms/shift | 94 | 51 | N=28 shifts | Noise control rules |
Baseline Line Layout and Flow
Conclusion: baseline mapping showed transport-heavy flow around the sealer constrained OEE at 62% and changeover at 48 minutes for N=2 lines over a four-week sample. Data collected with time stamps from MES and video standard work confirmed 37% of time in travel. Clause/record: IQ record IQ‑25‑011 and HACCP hazard analysis (HARPC 21 CFR 117.130) documented access paths and critical control points. Steps: we repositioned dunnage, created a kanban for lids, split preheat and evacuation queues, standardized cart pitches, and color-coded recipe carts. Risk boundary: keep e-stop zones intact per ISO 13849‑1 PL d and verify no new pinch points. We referenced the heavy duty vacuum sealer footprint to validate clearances without new guarding. Travel waste triggers overtime and ergonomic risk.
Conclusion: flow leveling allowed parallel operations without expanding labor. Data: WIP between case packer and sealer moved from 920 to 430 units, while MTTR stayed near 14 minutes across 38 downtime events. Clause/record: OQ record OQ‑25‑019 defined laddered rates and pitch; Annex 11, clause 9 required version control on route changes. Steps: we applied SMED to pre-stage jaws, moved LOTO shadow boards, codified three start-up checklists, balanced operators across two cells, and posted takt bands at each buffer. Risk boundary: loaders must not bypass interlocks; verify weekly in the QMS audit. The layout also prepared space for a future vacuum can sealer, which keeps format flexibility on the same utility skid. No conveyors were added during this phase. By design.
Thermal Profile Windows and Monitoring
Conclusion: defining thermal and vacuum windows stabilized seal integrity across the six formats. Data: sealing jaw temperature held 165–178 °C, vacuum setpoints at 38–41 kPa absolute, and dwell 0.65–0.80 s, yielding leak rate under 250 ppm on N=52 lots. Clause/record: SAT‑25‑042 captured golden batch recipes; HACCP plan logged final seal as a critical control point with verification every hour. Steps: we tagged sensors, locked recipes, tuned PID gains, added dual-probe verification, and enabled SPC alarms. Risk boundary: prevent over-temperature that could deform lids or under-vacuum that risks oxygen ingress. For the chamber style ASFL vacuum sealerealer configuration, we documented headspace limits and jar torque ranges so engineers could replicate windows without ad hoc edits. Operators scan recipes before any start.
Conclusion: inline monitoring created an audit trail that quality and maintenance can trust. Data: temperature Cpk ran 1.67, vacuum Cpk 1.52, and dwell Cpk 1.45 over four weeks; alarms per 10,000 packs moved from 18 to 7. Clause/record: Annex 11 clause 12 and Part 11 electronic records required user attribution; we tied badge IDs to recipe edits. Steps: deploy dual-thermocouple check fixtures, add weekly calibration, set red/yellow SPC bands, and export exceptions nightly. Risk boundary: alarm floods can mask true faults, so we throttled repeats to five per minute with timestamps. This framework also aligns with an eventual canning module, where a heavy duty vacuum sealer head may share sensors. Change control keeps edits reviewed within the QMS cadence monthly.
Print and Label Quality Control
Conclusion: serialization and print verification tightened traceability without impacting cycle time. Data: GS1‑128 label grade held A to B, misprint defects moved from 480 to 120 ppm, and FPY from 94.1% to 98.2% across N=52 lots. Clause/record: GS1 General Specifications section 2 and Part 11 audit trail rules governed code content and approvals; PQ record PQ‑25‑007 covered performance qualification. Steps: we added camera-based grading, locked artwork hashes, standardized label stock, set reject bins, and trained operators on disposition. Risk boundary: avoid false rejects by tuning illumination and angle before tightening thresholds. The vacuum can sealer pallet identifiers were reserved in master data to keep mixed-format aggregation consistent with case packer logic. No new stations were added during this change.
Conclusion: code validation linked to batch records enabled release decisions within minutes of lot close. Data: rework hours tied to labeling moved from 22.4 to 6.1 per week, while MTBF on printers rose from 3.2 to 5.9 hours after adding spares kits. Clause/record: Annex 11 clause 4 on accuracy and Part 11 Subpart B guided electronic signatures for holds; SAT deviations were closed under CAPA‑25‑014. Steps: define camera recipes, enable checksum on human‑readable lines, teach rejecting by class, and publish a daily scorecard. Risk boundary: camera drift can occur after mechanical bumps, so guards and quick checks are included. Training refresh runs quarterly; records link to PQ sampling plans and SOP-PRT-021. Supervisors verify scans per shift against the master list.
| Compliance Clause | Control / Evidence | Audit Cadence |
|---|---|---|
| GS1 General Spec §2 | GS1‑128 code content, camera grade reports, master data approvals | Monthly review; PQ-25-007 reference |
| Annex 11 clauses 4, 9, 12 | Versioned recipes, access control, audit trails with user attribution | Quarterly internal audit; SAT-25-042 appendices |
| HARPC 21 CFR 117.130 | Seal as CCP, hourly verification checks recorded to MES | Per lot; QMS trending weekly |
| ISO 13849-1 PL d | Safety PLC validation, e-stop zoning drawings | Annual validation; IQ/OQ pack |
Compressed Air Reduction Program
Conclusion: targeted pneumatic changes reduced demand without risking seal quality. Data: compressor flow logged 11.4 to 8.7 Nm³/h at the sealer island, and energy per pack moved from 0.18 to 0.15 kWh across N=2 lines. Clause/record: maintenance standard PM‑AIR‑012 and HACCP prerequisite program covered air quality; SAT‑25‑042 appendix listed nozzle IDs. Steps: replace open blows with engineered nozzles, fit solenoid sleep timers, add leak tags, re-zone purges, and verify with ultrasonic audits. Risk boundary: never throttle actuation air for safety circuits; keep ISO 13849‑1 PL d integrity and verify cycle time stays inside recipe windows. A standard kit list was issued; copies shipped to N=3 sites to replicate the change under a shared SLA and common spares. Stock levels synchronized.
Conclusion: air program yielded payback through lower compressor run hours and fewer unplanned stops. Data: MTBF on pneumatic actuators moved from 280 to 410 hours; MTTR stayed at 14 minutes; Payback modeled at 8.5 months with CapEx €18,700 and OpEx €2,900. Clause/record: energy log EL‑25‑003 and Annex 11 clause 9 required periodic review with signatures. Steps: implement tier meetings, post an air leak Pareto, close leaks weekly, and validate cycle times after each change. Risk boundary: drying and filtration must meet HARPC compressed air hygiene standards for direct food contact zones. Utilities confirmed demand curves; no compressor was cycled off, but loading bands shifted. Finance reviewed bills for three months to verify modeled savings against invoices. Results were archived centrally.
| Economics | Amount | Assumption |
|---|---|---|
| CapEx | €18,700 | Nozzles, sensors, spares, fixtures |
| OpEx (year 1) | €2,900 | Calibration, filters, camera lenses |
| Annual energy savings | €18,000 | 0.03 kWh/pack × 6.0 M packs × €0.10/kWh |
| Overtime avoidance | €6,000 | ~240 h/year × €25/h |
| Scrap/rework reduction | €6,600 | FPY delta applied to rework costs |
| Total annual savings | €30,600 | Energy + OT + rework |
| Payback | 8.5 months | (CapEx+OpEx)/annual savings |
| Sensitivity: energy ±20% | Payback 7.6–9.7 months | Volume constant |
Alarm/Event Exports
Conclusion: clean event exports enabled faster root cause and governed changes. Data: alarms per shift moved from 94 to 51, bad actor share from 61% to 29%, and meantime to acknowledge from 2.4 to 1.1 minutes over N=28 shifts. Clause/record: Annex 11 clause 12, Part 11 Subpart C, and SAT‑25‑042 governed audit trails; event ID taxonomy mapped to CMMS codes. Steps: export OPC UA to the historian, map codes, publish a weekly Pareto, review in tier meetings, and tie changes to CAPA tickets. Risk boundary: ensure security by role; changes require dual review. We defined SLAs: acknowledge within five minutes, contain within fifteen, and close within five days. Compliance reporting grouped by line, shift, and alarm class each week automatically.
Conclusion: consistent exports supported replication across the network. Data: three sites consumed the same schema; correlation between vacuum dips and seal alarms peaked at r=0.73; FPY variation between sites tightened from 5.4 to 2.1 percentage points. Clause/record: configuration SAT CSAT‑25‑010 and change control CCR‑25‑031 set cadence; Annex 11 clause 9 required versioned configurations. Steps: lock code tables, archive daily CSVs, validate against a golden file, and test restores quarterly. Risk boundary: avoid personal data in exports; user IDs are hashed. As a side note, the manual mason jar ASFL vacuum sealerealer request was modeled in the schema to preserve compatibility, though not enabled on production credentials. Staging users accessed sanitized data; production access remained segregated by role and location controls.
FAQ: Format Expansion and Validation
Q: whats the best vacuum sealer for format agility on our network? A: best depends on product, jar geometry, and validation load. We benchmarked the ASFL family against alternatives by OEE delta, changeover minutes, and Annex 11 audit features. For small-batch glass, the manual mason jar ASFL vacuum sealerealer met torque, vacuum, and labeling controls with lower CapEx, but slower MTBF recovery. For higher throughput, the chamber style ASFL vacuum sealerealer sustained window control and serialization with fewer recipe variants. Decide using FPY, leak ppm, and kWh per pack over N=10 lots. Then lock recipes, set print-and-verify, and export alarms to sustain outcomes. Audit readiness, GS1 compliance, and ISO 13849-1 boundaries should remain explicit in procurement files for each site.
Across Seaco’s two lines and three peer sites, these controls sustained the reported metrics and provided a repeatable playbook. Executives can plan capital with an eight-and-a-half‑month payback, while operations hold OEE near the new baseline with recipe locks and alarm exports. Quality retains an auditable trail under Annex 11 and GS1, and finance sees energy per pack at 0.15 kWh in the measured window. The approach remains tied to the ASFL platform, so future format growth can replicate with the same records and window definitions.
