FAT vs SAT is a question of place and proof. A factory acceptance test (FAT) proves the control system works as specified on a test bench before it leaves the workshop. A site acceptance test (SAT) proves the same system works in the real plant, with real field devices, power and process. FAT finds defects cheaply; SAT confirms the installation.
What you receive: send your functional description, I/O list or test specification and we return a written proposal within one working day. Request a quotation
By Eng. Junaid Ali, KSA Industrial Automation Company. Published October 2026.
What does FAT stand for, and what does it prove?
FAT stands for factory acceptance test. It checks the delivered system against the contract specification in a controlled place, using simulated signals instead of live plant equipment. This post covers the functional side: PLC logic, HMI screens, alarms, SCADA links and communications. The wiring and insulation tests on the panel hardware are a separate workshop activity done by the panel builder.
The scope is wider than software alone, since the operator interface, the historian and any alarm routing to phones or email can all be proved on the bench. A functional FAT normally proves:
- Every input and output reads and drives correctly, using simulators or signal injectors.
- Each interlock, sequence and failure mode behaves as the functional description says.
- Alarms appear with the right text, priority and acknowledgement behaviour.
- Communications to other systems exchange the agreed data.
- The documents delivered match the build.
What does SAT stand for, and what does it prove?
SAT stands for site acceptance test. It repeats the critical tests after installation, now with real instruments, valves, motors, power quality and plant noise. It also proves what FAT cannot: that field wiring is correct, that devices are scaled properly, and that the system works with the process.
SAT is usually shorter than FAT in logic testing and longer in physical checks. Plan it around the plant: instruments need to be installed, calibrated and powered, and operations staff must be free to witness. When these are not ready, the SAT slips, so the dependencies belong in the project schedule from the start. A loop check, where each instrument is verified from field to screen, is part of this stage. For the method, see loop checking and instrumentation.
What is the difference in FAT vs SAT, side by side?
| Point | FAT | SAT |
|---|---|---|
| Location | Workshop or integrator's test bench | Customer plant |
| Signals | Simulated or injected | Real field devices and process |
| Main purpose | Prove the design and software | Prove installation and operation |
| Typical findings | Logic errors, missing alarms, wrong tag text | Wiring faults, wrong scaling, noise, device failures |
| Cost of a fix | Low, before shipment | Higher, with plant waiting |
| Witnessed by | Customer representative and integrator | Customer operations, maintenance, integrator |
| Sign-off result | Release for shipment | Handover to operations |
Who signs a FAT or a SAT?
The customer's representative signs, alongside the integrator's engineer. A signature means the line item met its written pass criterion, not that the project is finished. Many Saudi projects also require the EPC contractor or a consultant to witness. Agree this in the inspection and test plan (ITP) before testing starts, because a test nobody is authorised to accept has to be repeated. The ITP also states the hold points where work stops until a witness signs, and the rule for what happens to a failed line: fix, re-test, and record both attempts.
Worked example: how a punch list moves from FAT to SAT
This is an example to show the method; real lists vary. A pump control system has 120 test lines in the FAT. During the test, the team records findings in three categories: A (stops release), B (fix before SAT) and C (cosmetic, fix when convenient).
| Finding | Category | Action |
|---|---|---|
| Duty pump does not swap on fault | A | Fix and re-test before shipment |
| Alarm text truncated on HMI screen 4 | B | Fix, verify at SAT |
| Tag name spelled differently on trend | C | Close during SAT |
Say the FAT has 4 category A, 7 B and 9 C findings. All 4 A items are closed and retested before the release, so 16 items remain. At SAT, those 16 are the first checks. A further 6 new items appear from site conditions, including a mis-scaled level transmitter, so the final SAT punch list has 22 entries, all with names and closure dates. The number of open items at handover should be zero for A and B.
How should you plan both tests?
- Write the test procedure from the functional description, not from the finished program.
- Agree the pass criteria and the witness list in the ITP.
- Run a dry FAT internally first, so the witnessed test is a confirmation.
- Ship with the signed FAT report and the open punch list.
- At site, repeat only what installation can change, then add the process-dependent tests.
- Close the SAT with signed sheets, the final backups and as-built documents.
KSA Industrial Automation Company supports FAT and SAT on projects for Saudi Aramco, SABIC and the Royal Commission for Jubail and Yanbu. If you want the on-site execution handled for you, our control system commissioning service plans and runs it, and the SCADA integration page lists what we test on the supervisory layer.
Common questions about FAT vs SAT
What does FAT mean in the pharmaceutical industry?
In pharmaceutical plants FAT also stands for factory acceptance test, and it often includes extra documentation such as validation protocols and traceable calibration records. The principle is the same: prove the equipment before it ships, then confirm it again on site.
Can a SAT replace a FAT?
You can skip a FAT, but you move the discovery of logic faults to site, where plant time and delays cost more. We recommend a FAT for any system with more than a handful of signals, and always for systems with interlocks that protect equipment.
Who writes the FAT and SAT procedures?
The integrator drafts them from the functional description and I/O list, and the customer or consultant reviews and approves them. Writing them before the software is finished keeps the tests independent of the code.
How long does a FAT or SAT take?
Duration follows the number of signals, the complexity of sequences and the number of interfaces to test. A small skid may need a day, while a multi-area system can need weeks. The test plan states the days, the witnesses and the re-test allowance.
What is a factory acceptance test report?
It is the signed record of the FAT: the test list with pass or fail for each line, the punch list with categories, the document revision tested and the names of those who witnessed. It becomes part of the handover file.
Need FAT vs SAT procedures written, or a test witnessed? Send your functional description or I/O list and receive a quotation within one working day. Email: inquiry@ksaindustrialautomation.com | Request a quotation