Introduction
If you want a vertical startup, the fastest way to miss it is by “saving time” during Factory Acceptance Testing (FAT).
Across food, beverage, personal care, and household goods, painful commissioning rarely comes from bad machines. It comes from risk deliberately deferred out of FAT and into the plant.
So which FAT shortcuts most reliably predict a rough start-up — and why?
In this article, you’ll learn which compromises are survivable, which almost guarantee pain, and where that risk will surface once the equipment hits your floor.
FAT is the lowest-cost environment to find defects in function, reliability, controls, and human interaction. A shallow FAT doesn’t eliminate problems – it relocates them to commissioning, under production pressure.
Some shortcuts are manageable. The ones below are not.
Why FAT Shortcuts Show Up So Loudly During Commissioning
Commissioning and start-up are inherently high-risk periods. They are the first time the machine is exposed to:
- Real operators and maintainers
- Production materials and real variability
- Full-rate expectations in a real production environment
- Safety, quality, and throughput accountability
When latent issues arrive unresolved, the downstream pattern is predictable:
- Frequent stops and micro-faults
- Logic bypasses and forced modes
- Quality losses and unstable rates
- Heavy OEM dependence long after “handover”
Industry guidance around FAT exists largely because misalignment on scope, criteria, and responsibility creates unbudgeted cost, schedule slip, and finger-pointing between OEMs and CPGs. Shortcuts amplify that misalignment.
If you’ve ever struggled to explain why a rushed FAT caused months of pain, this is the reason: Risk didn’t disappear—it just changed locations.
The FAT Shortcuts That Most Reliably Predict a Non-Vertical Startup
1. “Demo FAT” Instead of an Endurance + Performance FAT
- FAT compressed into a few hours
- One clean cycle and a short speed burst
- Immediate pressure to sign off
- No sustained run at rate
- High-use SKUs (primary runners) are missed, skipped, or under tested
Intermittent issues do not show up in a 30-90 minute demo:
- Sensor noise and marginal timing windows
- Pneumatic variability
- Adhesive and thermal stabilization
- Fastener loosening
- Nuisance faults that only appear after heat-soak or vibration
A demo FAT proves the machine can run. An endurance FAT proves the machine can keep running.
If you’re being pushed to accept a demo FAT, the real question to ask is: “Where will reliability discovery happen instead—and how expensive will that be?”
- “It ran at the OEM” but won’t hold rate on site
- Chronic microstops and fault storms
- Weeks of OEM “tuning” during live production
2. No Clear, Measurable FAT Acceptance Criteria
- Agenda exists, but criteria are vague
- Phrases like “runs well” or “meets expectations”
-
Arguments over:
- Rate at what efficiency?
- Which SKUs?
- Which materials?
- What reject rate is acceptable?
If acceptance criteria aren’t defined, measurable, and agreed upfront, you can’t:
- Objectively separate machine capability from integration or material problems
- Close scope without debate
- Prevent SAT from turning into a requirements negotiation
Ambiguity at FAT almost guarantees conflict at start-up—usually when time and patience are gone.
- Commissioning time consumed by argument, not problem-solving
- Requirements quietly expanding on the plant floor
3. Running “Convenient” Materials Instead of Production-Intent Material
- OEM uses whatever corrugate, cartons, film, or tape is available
- Only best-case material lots are run
- Glue, tape, or film specs differ from production
- Collation or product variability simplified
- High-volume SKU sizes not identified or inadequately tested
Most secondary packaging machines are material-handling disguised as automation.
Corrugate and carton variability—recycled content, warp, scoring quality, humidity exposure—drives:
- Case squareness issues
- Forming and sealing instability
- Random jams that defy “adjustment”
If you don’t test the real operating envelope, you don’t know where the edges are.
You may not be able to replicate all site conditions at the OEM-that’s what the SAT is for. But using non-representative materials at FAT almost guarantees a machine-versus-material blame loop later.
- “The machine is fine; the materials are the problem”
- Endless tuning instead of design correction
4. Skipping Abnormal-Condition Testing and Recovery Behavior
- FAT validates only the normal sequence
-
Little testing of:
- Misfeeds and doubles
- Missing product
- Low material conditions
- Jam detection and restart logic
- Alarm rationalization
Normal operation is not what kills vertical startup. Recovery behavior does.
If operators discover on site that recovery is unsafe, unclear, or requires engineering intervention, they will bypass, wedge, or run manual to keep production moving.
That behavior becomes permanent long before the machine is “fixed.”
- Sensor bypassers become “temporary permanents”
- Tribal-knowledge resets
- OEE never recovers, even after the machine “runs”
5. No Changeover, Wear-Part, or Standard-Work Validation
- Too few SKUs demonstrated
- Change parts present but never swapped
- No timed changeovers
- No verification of repeatability
- No standard work or one-point lessons
Vertical startup isn’t “first SKU at rate.” It’s the operating system being stable:
- Setup
- Changeover
- Preventive Maintenance
- Troubleshooting
If those aren’t proven, you’re vertical for one SKU—and fragile for months.
- Engineering or OEM required for changeovers
- Phantom adjustments and high scrap during format swaps
6. The Wrong People Attend FAT (or Aren’t Empowered)
- Engineers and PMs attend
- Actual operators and maintainers do not
- Or operators attend but never touch the machine
The people who will live with the machine must:
- Run it
- Break it
- Recover it
- Ask uncomfortable questions
Training gaps—not mechanical capability—are often the real bottleneck in ramp-up.
- Slow ramp despite “capable” equipment
- Daily surprises: “We didn’t know it worked that way”
7. Shipping With Open Action Items to Hit a Date
- Large punch list with vague closure
- “We’ll finish it during installation”
- PLC/HMI not frozen
- Drawings not as-built
Anything unresolved at ship almost always gets resolved:
- In shorter windows than necessary
- With limited access
- Under real production pressure
Start-up becomes a construction project—and reliability suffers permanently.
- Workarounds stack up
- Change control erodes
- “Temporary” fixes become permanent liabilities
8. Incomplete Controls and Safety Verification
- Limited I/O checkout
- Weak proof of safety circuits and interlocks
- Minimal alarm rationalization
- Logic forces already in use at FAT
Controls issues consume commissioning time disproportionately because they masquerade as mechanical problems.
Undisciplined logic changes and unverified safety behavior elevate both downtime and incident risk.
- Endless “electrical gremlins”
- Nuisance faults
- Unsafe temporary bypasses
9. No Integration-Minded Testing
- Machine tested as a standalone island
-
No simulation of:
- Starvation
- Blockage
- Accumulation behavior
- Handshakes and rate assumptions remain theoretical with little appreciation for up- and down-stream activity
Vertical startup is a system outcome, not a machine outcome.
Many “machine problems” are actually interface problems that only surface when the line runs as a system.
- Machine runs fine alone, collapses when integrated
- Late controls changes at the worst possible moment
10. Thin Virtual FAT or “Video Sign-Off”
- Short video clips instead of witnessed testing
- Limited camera coverage
- No shared data (rates, faults, rejects)
- No structured checklist
Virtual FAT is not inherently bad. Thin evidence is.
Without structure and data, you can’t verify acceptance—or spot marginal behavior that will matter later.
- Installation surprises
- Poor ergonomics and access
- Missing documentation
A Practical Way to Think About FAT Shortcuts
Use this table as a pre-FAT risk screen before agreeing to scope reductions.
FAT Shortcut
Risk Deferred to Site
Short endurance
→
Reliability / ‘infant mortality’*
Non-production materials
→
Material-handling instability
No failure scripts
→
Unsafe or slow recovery
No changeover testing
→
Human-system fragility
Incomplete controls checkout
→
Commissioning drag + safety risk
No integration testing
→
System-level instability
* Infant mortality is when a machine technically “works,” but starts failing repeatedly in its first weeks or months because it was never run long enough, hard enough, or realistically enough before shipment.
When you see a shortcut, ask: “What risk bucket am I buying—and where will it surface?”
Assumptions and Context to Keep in Mind
A few clarifications:
- Equipment type matters. Case erecting is far more sensitive to corrugate variability than many palletizing applications; shrink systems are sensitive to film and thermal stabilization.
- SAT exists for a reason. You cannot fully replicate plant utilities, upstream variability, or environment at the OEM. Confusing FAT and SAT wastes time.
- Some shortcuts are unavoidable. Product scarcity and schedule pressure happen. The real question is whether you replace lost coverage—or simply eliminate it.
The Bottom Line
If FAT feels rushed, shallow, or overly polite, it’s not being efficient—it’s exporting risk.
The machines that commission cleanly are rarely perfect at FATs. They’re the ones where defects were surfaced early, argued through, and closed before steel hit the floor.
If you want vertical startup, FAT isn’t where you save time. It’s where you decide where the pain will live—at the OEM, or on your line.
