The Phase That Everyone Underestimates
The commissioning and start-up phase is the period between mechanical completion and steady-state operation. It is short relative to the design and construction phases, it sits at the end of the schedule when contingency is depleted, and it is where the project is judged. A facility that takes six months to reach design throughput will be remembered as a six-month-late project regardless of how punctually it achieved mechanical completion.
The defining problem is that commissioning is not a single phase — it is a tightly sequenced chain of activities across mechanical, electrical, instrument, process, and safety disciplines, each with prerequisites in the others. A weak link anywhere stalls everything downstream. Good sequencing is what turns a chaotic finish into a controlled one.
The Four Stages and Their Boundaries
Commissioning is best understood as four discrete stages with clear handover gates:
Pre-commissioning (cold work, no hydrocarbons). Includes line flushing, drying, leak testing, instrument loop checks, motor solo runs, electrical megger and high-pot tests, and panel function tests. The deliverable is a system that is mechanically complete and electrically energised but contains only inert fluids — typically air, nitrogen, or water.
Cold commissioning (system function with utilities only). Pumps run on water, compressors run on air or nitrogen, control loops are tuned with simulated process signals, and the F&G system is functionally tested. The deliverable is a facility that operates correctly on utilities and is ready to receive hydrocarbons.
Hot commissioning (hydrocarbon introduction, controlled start-up). Inert atmospheres are displaced, vessels are pressurised, and process units come up sequentially. This is where months of paper-checked design meets actual fluid behaviour for the first time.
Performance testing (proving design intent). Throughput is ramped to nameplate, product specs are demonstrated, energy and emissions targets are verified, and the facility is handed over to operations as fit for purpose.
The handover gates between stages exist for a reason. Skipping or compressing them is the single most common cause of late-stage failure.
The Sequence Logic
The sequence within each stage is not arbitrary. It follows three rules:
Utilities before process. Power, instrument air, nitrogen, cooling water, fuel gas, and chemical injection must all be available and stable before any process unit can be commissioned. A surprising number of projects discover at hot commissioning that their nitrogen system was sized for normal operation and cannot support a parallel start-up purge.
Static before dynamic. Vessels, piping, and structural systems are tested before rotating equipment runs against them. Pumps and compressors are commissioned solo before being integrated into loops. Loops are tested before full system runs.
Inboard before outboard. Process units are commissioned roughly in the direction of fluid flow — inlet systems first, then primary separation, then treatment, then export. Each unit needs its upstream feeders ready and its downstream takers able to accept production. Trying to commission downstream before upstream is the classic source of "we can't run because we have nowhere to send the product."
Lay Down the Systems Boundary
The single most powerful tool for sequencing is a clear systems map. The facility is divided into commissioning systems — typically twenty to fifty per project — with clean boundaries that everyone agrees on. Each system has its own pre-commissioning, cold commissioning, hot commissioning, and handover dossier.
Boundaries should follow process logic, not P&ID page boundaries. The flare header is one system. The instrument air distribution is one system. The produced water treatment train is one system. Trying to commission half a system because the other half is not finished produces incoherent test records and disputes at handover.
Once the boundaries are set, the systems are sequenced into a master commissioning plan that drives the entire end-of-project schedule. Construction work-fronts are reorganised to deliver the systems in the order they will be commissioned, not in the order that suits the construction contractor's labour curve. This is a fundamental shift in mindset that brownfield-experienced project managers make instinctively and greenfield-focused managers often resist until it costs them weeks.
Risks That Drive the Schedule
Several factors should be flagged in the commissioning plan because they create non-linear schedule risk:
- First lift of a critical machine. The first time a large compressor or generator runs is when bearing alignment errors, vibration issues, and lube system problems surface. Build a week of float around it.
- First hydrocarbon introduction. Leaks and small failures cluster in the first 72 hours of hydrocarbon service. The commissioning team must be at full strength and the operator must be alongside, not arriving the following Monday.
- First production through new instrumentation. Field instruments calibrated dry behave differently in service. Expect a wave of recalibrations and loop retunes immediately after first product.
- First emergency shutdown. The first time a real ESD trips a fully populated facility surfaces every interaction nobody simulated. Some operators schedule a deliberate ESD test before declaring start-up complete.
Operator Involvement: Not Optional, Not Late
The commissioning team is not a separate population from the operations team — or it should not be. The most successful start-ups are the ones where the operators have been embedded in commissioning since pre-commissioning, learning their facility as it comes alive. By the time the start-up phase begins, they are running the controls and the commissioning team is supporting.
When operators arrive at the end of commissioning expecting a complete and tested facility, two things happen: they distrust what they did not see being built, and they take three to six months to learn the facility's quirks. Both are avoidable.
The Performance Test Window
Performance testing is the formal mechanism by which the facility is judged against design. It demonstrates throughput, product spec, utility consumption, emissions, and availability over a defined period — typically 72 hours of continuous operation at design rate with no off-spec events.
The performance test is also when warranty clocks start, retention payments release, and the project ends. Designing the facility, the commissioning sequence, and the operator readiness to deliver a clean performance test at the first attempt is one of the most valuable things a project manager can do. A failed performance test is not just a delay — it is a renegotiation.
Conclusion
The commissioning and start-up phase is where the project is delivered. The sequence in which systems are commissioned, the discipline of stage boundaries, the integration of operators, and the realistic acknowledgement of where time will be lost are the engineering decisions that determine whether the facility starts up on schedule or limps to nameplate over months. Treat commissioning planning as a first-class engineering activity from FEED onwards, not as something the construction contractor will figure out at the end.
