On most capital projects, commissioning occupies the final block of the program and absorbs every delay that occurred before it. This is not a planning oversight. It is the predictable consequence of treating commissioning as a phase rather than as a program that runs alongside construction.
The result is consistent: the most technically demanding work on the project is attempted with the least remaining time, by teams under maximum pressure, on systems that are still being completed around them.
Why it ends up compressed
Commissioning sits at the end of the network, so it inherits accumulated slippage from everything upstream. When the completion date is fixed — by a lease, a production commitment, a funding condition, or a regulatory deadline — the only variable left is the duration allocated to commissioning.
It is also, unhelpfully, the activity that looks most compressible to people who have not done it. It is documented as a sequence of tests, and tests appear to be things that can be run faster or in parallel. In practice they cannot, because each test depends on a stable system, and stability is exactly what a project in its final weeks does not have.
Commissioning does not fail because it was planned badly. It fails because it is where all the earlier optimism finally has to be settled.
What running it in parallel means
Interleaving commissioning with construction is a sequencing decision made early, not a heroic effort made late.
Systems are completed and handed over progressively rather than simultaneously, which requires the construction sequence to be planned around system boundaries rather than around trades or areas. Static completion, energization, and functional testing follow each system as it completes.
Documentation is produced as the work proceeds. On projects where commissioning documentation is assembled at the end, the assembly itself becomes a critical-path activity — and it is frequently the reason a facility that is physically finished cannot be accepted.
Where integrated system testing is required, it is scheduled as a distinct activity with its own duration, dependencies, and float, rather than assumed to fall out of the individual tests.
Controls and protection deserve separate treatment
On power and process projects, controls, protection, and integration testing are the highest-risk element of commissioning and the most frequently compressed.
They are last in sequence because they depend on everything else being live. They require specialist resource that is often shared across several projects. And they are the point at which design assumptions meet installed reality, which means they generate defects that must be resolved before the system can be accepted.
Allocating this scope the residue of the program is the single most reliable way to miss an in-service date.
Operational readiness runs alongside
Commissioning delivers a working system. Operational readiness delivers an organization capable of running it, and it is a separate program with its own lead time.
That includes operator recruitment and training, spares and consumables, maintenance regimes, permits and licenses required to operate rather than to build, and the documentation the operating organization will actually rely on.
A facility that commissions successfully into an organization that is not ready has not achieved anything the business case recognizes.
What to ask for early
On any program where the in-service date carries commercial consequence, three artifacts are worth requiring from the outset.
A commissioning strategy that defines system boundaries and the handover sequence. A documentation plan naming who produces what, when, and to whose acceptance. And a resource plan for the specialist commissioning and integration effort, tested against availability rather than assumed.
Produced at the start, these cost very little. Produced at the end, they are produced under conditions that guarantee they will be wrong.
