Equipment commissioning

The Equipment Commissioning Checklist

12 Jun · 11 min read · by the Phaselo team

An equipment commissioning checklist that lives in a Word document is already out of date the morning installation starts. Commissioning projects fail in predictable places: scope that was never frozen, vendor documentation that arrives three weeks after the equipment, a punch list that grows in a PDF nobody reopens, and a handover that production never actually agreed to. The fix is not a better document. It is to run the checklist as a live project plan, where every line has an owner and a date, progress rolls up, and the day you slip the schedule says so. This guide gives you the five phases, the line items inside each one, and the reason each phase tends to blow up, so you can build a commissioning plan that survives contact with the plant floor.

Why a commissioning checklist needs to be a plan, not a document

A document treats every item as equal and timeless. A commissioning project is neither. The phases happen in order, each task has exactly one person who owns it, and the dates are linked: if mechanical installation finishes two days late, hot commissioning and the performance run move with it, and the production restart date is now at risk. A static checklist cannot show you that. A plan can.

Here is the rule that makes the difference. Turn every checklist line into a task with one owner and one due date, group the tasks into the five phases as work packages, and let progress roll up. When the plant manager asks how commissioning is going, the answer is a number per phase, not a feeling. When a vendor slips, the projected finish date moves on its own and you see it the same day, not at the next steering meeting. That is the whole argument for running this in software built for the work, like Phaselo, instead of a shared spreadsheet.

• A spreadsheet cell with someone's initials is not an assignment anyone gets notified about.
• A checklist with no dates cannot tell you the project is going to land late until it already has.
• A punch list in a PDF is a snapshot. By the afternoon it is wrong, and nobody knows which items are closed.
• Without a recorded baseline, there is no honest answer to the question every commissioning post-mortem asks: when did this start slipping, and why?

Phase 1: Scope and design sign-off

This is the cheapest phase to get right and the most expensive to get wrong. Every hour of rework downstream traces back to a scope that was assumed instead of agreed. The single most common failure: engineering signs off the functional spec and operations never sees it until the equipment is running differently than the line needs.

• Functional specification approved by operations, not just engineering. The people who run the line at 2am sign, or it is not signed.
• Interface points documented in detail: power, compressed air, water, drainage, data, and the upstream and downstream equipment it has to hand off to.
• Long-lead items identified and ordered, with real delivery dates entered as tasks in the plan. A 16-week switchgear lead time decides your whole schedule, so it belongs on the timeline from day one.
• Acceptance criteria agreed in writing: throughput per hour, first-pass quality rate, availability target, changeover time. Numbers, not adjectives.
• Budget approved with a named contingency line, not just the vendor quote. Commissioning discovers work; the budget should expect it.

Why Phase 1 fails

It fails because scope is never frozen. A drawing gets revised, a vendor offers an upgrade, an operations request slips in after sign-off, and now you are commissioning a different machine than the one that was budgeted. The discipline that prevents this is a recorded baseline. The day the scope and budget are agreed, capture them. After that, any change to scope, dates, or budget is a deliberate act with a reason attached, not a quiet edit. That is exactly what baseline governance is for: go-live captures the figures, and re-baselining requires a recorded reason and leaves an audit trail. When finance later asks why commissioning ran 14% over, you point at the baseline and the list of approved changes, not your inbox.

Phase 2: Installation readiness

Readiness is the phase teams skip because the equipment has not arrived yet, so it feels like there is time. There is not. Readiness is everything that has to be true before the truck shows up, and every missing item becomes a crane sitting idle on a day rate while someone runs to find a permit.

• Site survey complete: floor loading checked, access routes measured against the equipment footprint, laydown area assigned.
• Services live and proven at the connection points before the equipment arrives, not promised for the day of.
• Contractor inductions, permits, and isolation approvals arranged ahead of mobilisation, not on the morning of.
• Production schedule adjusted and agreed: the installation window is booked with the line, in writing, weeks ahead.
• Old equipment decommissioning plan written, including isolation, disconnection, removal, and disposal, each as its own task.

Why Phase 2 fails

It fails on the interfaces between trades and companies. The civils contractor finishes the plinth a day late, which pushes the riggers, who belong to a different firm and have another job booked, which means the equipment sits in the laydown area for a week. The fix is one shared timeline everyone can see, where a slip on the plinth visibly pushes the rigging task and flags the collision before it happens. If your readiness tracker is a checklist with no dependencies, you find out about the collision when the riggers do not turn up. For the full version of coordinating mobilisation, permits, and trades around a fixed window, see the walkthrough on planning a maintenance shutdown.

Phase 3: Installation and cold commissioning

Cold commissioning means proving the equipment with the power on but the process off. You are checking that every part does what it should before you put product, heat, pressure, or speed through it. Nothing runs in anger yet. The point is to catch wiring errors, reversed valves, miscalibrated sensors, and failed interlocks while a mistake is cheap and nobody is near a moving hazard.

• Mechanical installation checked against the drawings before power-up: alignment, fasteners torqued, guards fitted, lubrication done.
• Electrical installation tested and tagged; isolations verified; phase rotation confirmed before any motor turns.
• I/O checks: every sensor and actuator proven from the control system, one point at a time, so you know the screen matches the field.
• Safety systems function-tested before anything runs: e-stops, light curtains, guard interlocks, and emergency isolation each proven and recorded.
• Dry or no-load runs: motors bumped for rotation, conveyors run empty, pumps primed and proven on water before product.

Punch list management: tracked items, not a PDF

Cold commissioning is where the punch list is born, and the punch list is where most commissioning projects quietly lose control. A list of defects exported to a spreadsheet is a snapshot that is wrong by the afternoon. Treat each punch item as a tracked task instead: it has an owner, a category (mechanical, electrical, controls, safety, documentation), a severity, and a status that moves from open to closed in one place. Now the question is answerable at any moment: 12 items open, 3 of them safety-critical and blocking hot commissioning, owned by two contractors, oldest is four days old. That is the difference between a punch list you manage and one that manages you.

Tie the gate to the data. You do not start Phase 4 with any safety-critical punch item open. Because the items are tasks rolling up to the phase, that gate is visible to everyone, not a judgement call made in a corridor.

Phase 4: Hot commissioning and performance runs

Hot commissioning is the live version: real product, real process conditions, vendor support on site. Cold commissioning proved the equipment can move. Hot commissioning proves it can do the job at the rate, quality, and reliability you agreed in Phase 1. This is where the acceptance criteria stop being a document and start being a test you either pass or fail.

Cold versus hot commissioning, plainly

Cold commissioning is power on, process off: you verify the machine without putting it under real load. Hot commissioning is power on, process on: you run actual product under actual conditions and measure the result. Keep them as separate phases with a hard gate between them. The gate is simple: no safety-critical punch item open, all safety functions proven, and a signed authority to run product. Skipping straight to hot because the schedule is tight is the single most expensive shortcut in commissioning, because now you are diagnosing faults with product, heat, and a moving line in play.

Acceptance criteria and the performance test

The performance run is a formal test against the numbers from Phase 1, not a vibe check that the machine seems happy. Define it before you start: how long the run is, what counts as a pass, who witnesses it, and what the equipment has to demonstrate.

• Throughput: hold the agreed units per hour across a sustained run, for example a continuous 8-hour shift, not a lucky 10 minutes.
• Quality: first-pass yield at or above target across the run, with rejects recorded and categorised.
• Availability and reliability: a reliability run (often 72 hours, sometimes a full week for critical assets) with unplanned stops logged and counted against the agreed limit.
• Settings captured: every recipe, setpoint, and parameter that produced the passing run recorded as the as-commissioned baseline.
• Deviations tracked as items: every gap between specified and actual performance is a task with an owner and a status, closed before sign-off or formally accepted with a reason.

Also in this phase: operator training delivered on the running equipment and signed off by the operators themselves, and the critical spares physically on the shelf and confirmed, not just on a purchase order. An asset you cannot fix at 2am on a Sunday is not commissioned, whatever the certificate says.

Why Phase 4 fails

It fails when the performance test is vague, so a marginal result gets argued into a pass and the shortfall becomes production's problem for the next five years. It also fails when the vendor demobilises before the reliability run is complete, because the moment support leaves site, fault-finding slows to a crawl. Hold final payment milestones against the signed performance test, and keep the dependency on the plan so the reliability run finishes before the vendor's last day, not after.

Phase 5: Handover production actually signs

Handover is where commissioning projects look finished but are not. The equipment runs, the project team moves on, and six weeks later maintenance discovers there is no greasing schedule, the as-built drawings never arrived, and nobody recorded the warranty claim contact. A handover is not an email saying it is done. It is a checklist of artefacts, each owned and dated, ending in a signature from the people who will live with the equipment.

• As-built documentation received and filed where maintenance can actually find it, with the drawings reflecting what was installed, not what was first drawn.
• Maintenance plan loaded: preventive schedules, procedures, and spares linked to the asset in the CMMS.
• Warranty terms recorded with start and end dates and a named claim contact, so a failure in month seven is a claim, not a write-off.
• Spares and consumables confirmed on the shelf with part numbers and reorder points set.
• Formal acceptance signed by production: the operators and maintainers who own the equipment from here, not the project sponsor who hands it over.
• Lessons-learned review held while the team still remembers, with actions captured as tasks rather than good intentions.

Why Phase 5 fails

It fails because handover has no owner and no acceptance gate, so it dribbles out as the team drifts to the next job. Make production's sign-off a task with a name and a date on it, with the documentation items as its dependencies. Acceptance is not granted until the artefacts exist. That single rule turns handover from a hopeful email into a closed gate.

How baseline and slip tracking keep commissioning honest

The reason commissioning schedules slip silently is that nothing recomputes the finish date for you. A checklist does not know that a two-day delay in cold commissioning has pushed the performance run into the vendor's demobilisation week. A real plan does. Build the five phases on a critical-path timeline and three things happen automatically: you get a projected finish date that updates as tasks move, you see which tasks are on the critical path (the ones where a one-day slip slips the whole project), and you see total float on everything else (how much a task can move before it starts to bite).

Pair that with a recorded baseline and you get the honest version of progress. The baseline is the plan you committed to at sign-off. Slip is the gap between baseline and projected finish, visible the day it opens, not at the post-mortem. Proactive overdue and slip alerts mean the owner of a late task hears about it in-app and by email without anyone chasing. When a re-baseline is genuinely needed, because scope grew or a vendor delivery moved, it is recorded with a reason and left in the audit trail, so the schedule has a memory. See how that timeline and critical-path engine work on the Gantt view. This is the same engine that makes Phaselo a credible alternative to Jira for operations teams, which was built for software sprints rather than phased site works.

Run your commissioning checklist as a live project

Copy the five phases above into a plan, not a document. Each phase is a work package, each checklist line is a task with one owner and one due date, punch items and deviations are tracked tasks with a status, and progress rolls up so you always know where commissioning stands. Phaselo gives operations teams exactly that structure: a tree for the work breakdown, a Gantt for the critical path and slip, a board for status from the floor, a report for the money, and people views so everyone sees their tasks. It is $8 per user per month with a 14-day free trial and no credit card, and you can see the pricing in full. Start a free trial and paste this checklist into your next commissioning project.