A factory acceptance test proves that equipment meets its specification while it is still at the supplier. A site acceptance test proves that the same equipment works once it is installed, connected and operated in its real surroundings. They sound like two copies of the same exercise, but they answer different questions, and treating them as duplicates is the most expensive mistake a project can make.
This guide to FAT vs SAT comes from the witness’s chair: the people who read the procedure, watch the test and sign the record. It covers what each test can and cannot prove, how to write the procedure, how to avoid retesting everything, and how to handle the punch lists that inevitably follow, with attention to the realities of projects in Indonesia.
- FAT and SAT in one sentence each
- Different questions, different evidence
- Writing the procedure before the test
- Test credit: not repeating what is already proven
- Who does what at the test
- Deviations and punch lists
- Practical points for projects in Indonesia
- Frequently asked questions
FAT and SAT in One Sentence Each
The factory acceptance test is performed by the supplier, before the equipment is moved to its destination, to demonstrate that it meets the contract and works as specified. The site acceptance test is performed after the equipment arrives and is installed, usually with the user’s own people involved. This is how acceptance testing is generally described for industrial equipment.
Between the two lies transport, handling, installation and connection to everything else on site. Each of those steps can introduce faults that no factory test could reveal. That is why both tests exist, and why an owner who skips either one is trading a small cost now for a large uncertainty later.
Different Questions, Different Evidence
A FAT asks whether the equipment, on its own and under controlled conditions, does what the specification says. Typical elements are a completeness check, verification against contract requirements, a functional demonstration and a final inspection. Inputs and loads are often simulated, and that is acceptable as long as everyone knows which parts were simulated.
A SAT asks whether the installed system works with real utilities, real signals, real interfaces and real operators. Things a FAT cannot prove include the quality of site power, the behaviour of field wiring, the alignment on the foundation, the interaction with neighbouring systems and the performance in the actual ambient conditions.
Seen this way, the tests are complementary. The FAT removes the faults that are cheap to find in a workshop with tools and spare parts on hand. The SAT confirms what only the site can show. A project that expects the SAT to find workshop faults will pay site rates for the work.
Writing the Procedure Before the Test
The test procedure should be agreed before anyone travels. It lists the sequence of tests, the setup, the instruments to be used, the acceptance criteria for each step and the record that will be produced. A procedure written on the morning of the test is a negotiation, not a procedure.
Declare simulations openly. If a motor is not coupled to a real load, or a signal is injected rather than generated by a real process, write that down in the procedure, together with what the simulation does and does not prove. Silence invites later claims that the equipment was “fully tested”.
Prerequisites belong in the procedure too: completed assembly, installed software or settings, available utilities, safety arrangements and the documents that the witnesses will want to check on the day. A short readiness confirmation signed by the vendor before the witnesses arrive prevents the familiar scene of a delegation waiting while the equipment is finished.
A pre-FAT dry run by the vendor is a habit worth encouraging. It finds the loose wire or missing document while nobody important is watching, and it makes the witnessed test shorter and calmer. Also confirm that instruments used in the test carry valid calibration, since a perfect result from an uncalibrated gauge proves very little.
Test Credit: Not Repeating What Is Already Proven
A useful concept is test credit. Decide in advance which results from the FAT will be accepted at the SAT without repetition, and which must be repeated because transport or installation could have affected them. This avoids both the waste of repeating everything and the risk of assuming nothing needs rechecking.
For example, a logic test of a control system that has not changed since the FAT can usually be credited. A test of insulation or alignment, which transport and installation can easily affect, normally cannot. The decision depends on the equipment and the risk, and it should be written into the plan.
Keep retest triggers in the plan. If the equipment was modified after the FAT, repaired after transport damage or stored for a long period, the credited results may no longer apply. Naming those triggers up front removes arguments when they occur.
Who Does What at the Test
Roles on test day should be clear to everyone in the room:
- Vendor: prepares the equipment, runs the test, records results and keeps the original data.
- Buyer or owner’s engineer: witnesses, accepts or rejects against the specification, and signs the record.
- Third-party inspector: where appointed, provides an independent witness, verifies instrument calibration, checks the records and reports.
- Operations and maintenance staff: attend where possible, because a FAT is the cheapest training opportunity the project will get.
The vendor operates; the witnesses observe. Witnesses who start operating the equipment blur responsibility, and a later fault becomes harder to attribute. If an operation needs to be repeated differently, record the request and let the vendor carry it out.
Deviations and Punch Lists
Few tests pass cleanly. The deviation list records every item that did not meet the specification or the procedure, with a description, a reference to the requirement, an owner and a target date. The record should be agreed on the spot, before the witnesses leave.
Many projects classify items by when they must be closed. A common convention is one category that must be fixed before shipment, one that must be fixed before commissioning and one that can follow later. The labels vary between contracts, so define them explicitly and avoid assuming that everyone shares the same scheme.
Be wary of the phrase “we will fix it at site”. It can be legitimate when the item truly depends on site conditions. Otherwise it moves a cheap fix into the most expensive location and the tightest schedule. If an open item is released for shipping, record who accepted the risk and when it will be verified.
Practical Points for Projects in Indonesia
Distance shapes both tests. Equipment fabricated in Batam or elsewhere in Indonesia may be tested a short trip from the owner’s team, while equipment built abroad may require travel. Remote attendance by video can supplement a FAT but cannot fully replace a person who can see, hear and touch the equipment, so agree which steps need physical presence.
For SAT, remote locations affect planning. Access, lifting equipment, power, accommodation and the arrival of spares all need to be ready before the test team arrives. A test delayed for a missing consumable costs the same daily rate as one delayed for a failure.
Tropical heat, humidity and dust are real test conditions at many sites. If the equipment is sensitive to them, ask whether the FAT conditions represent the site and whether any climate-related checks belong in the SAT. G.I Quality Services lists factory and site acceptance tests among its services, and independent witnessing of either one works best when the procedure and the credit rules are agreed before anyone arrives.
Frequently Asked Questions
What is the difference between FAT and SAT?
A FAT is carried out at the supplier’s facility before shipment to prove the equipment meets its specification. A SAT is carried out at the installation site after the equipment is installed to prove it works in its real environment and with its real interfaces.
Who performs a factory acceptance test?
The supplier performs it, and the buyer or the buyer’s representatives witness it. An independent inspector may attend to verify records, instrument calibration and compliance with the agreed procedure.
Can a SAT replace a FAT?
Usually not. A fault found at site costs more to correct than one found in the workshop, and some equipment cannot be tested properly once installed. Skipping the FAT shifts risk to the most expensive stage.
Do I have to repeat every FAT test at SAT?
No. Define test credit in the plan so that results unaffected by transport and installation can be accepted, while tests that could be affected are repeated. Retest triggers cover modifications, repairs and long storage.
What is a punch list?
It is the list of open items found during testing or inspection, with owners and target dates. Items are often classified by when they must be closed, such as before shipment or before commissioning.
Can FAT be witnessed remotely?
Parts of it can, if the contract allows and the equipment permits clear viewing. Physical checks such as dimensions, wiring quality and mechanical movement are best witnessed in person, so agree the split before the test.