How can an energy owner assess whether an LNG start-up date is realistic?

TLDR: For owner-side commissioning leads and project directors on LNG export projects: a start-up date is realistic when every system on the path to first LNG has completion evidence, an EPC and an operator reviewer who accept it, and a recorded decision. Test it by walking back from first LNG through the commissioning handover, not by reading construction progress.

In April 2025 a carrier brought a cargo of imported LNG into Kitimat, British Columbia, so that LNG Canada could test its equipment before the plant had produced any LNG of its own. The company described offloading that cargo for equipment testing as "critical to the safe start-up and commissioning process in advance of operations," and its first export cargo left the facility on June 30, 2025. Neither event appears on a construction progress curve. The test cargo was a commissioning step the company itself tied to safe start-up, and it came months before the first export.

A first LNG date is a forecast about a chain of handovers: construction to pre-commissioning, the EPC contractor to the operator's commissioning team, and the plant to the regulators and gas suppliers who must agree before hydrocarbons go in. Our view is that the owner should test that chain link by link, working back from first LNG, because the date is only as credible as the weakest link nobody has evidenced.

What makes an LNG start-up date realistic?

A start-up date is realistic when the owner can trace it back through every system that must be commissioned before first LNG and find three things for each one: evidence that the work is complete or credibly planned, a reviewer on each side of the handover who has accepted that evidence, and an explicit decision to accept the date, investigate further or test a different sequence. Overall percent complete only tells you how far construction has progressed.

The Construction Industry Institute describes commissioning and start-up as "the critical last phase of a capital project before the facility is placed in operation to meet commercial objectives." It is also the phase where the schedule changes hands. Construction logic is usually built and owned by the EPC contractor, commissioning is often executed jointly, operator readiness belongs to the owner, and approvals belong to regulators. A realistic date needs all four to agree.

Before testing that agreement, pin down what is being judged. First LNG from a named train, the first export cargo and commercial operation are separate events. Contracts commonly certify mechanical completion and a ready-for-start-up (RFSU) state system by system before hydrocarbons go in, but the definitions vary, so use the terms your contract sets.

Why does the date usually break at the EPC-to-operator handover?

The date tends to break at the handover because that is where separate plans have to meet, and the joins are rarely as well evidenced as the work on either side. The Construction Industry Institute's research on commissioning and start-up lists interface management, integration and a systems focus among the critical success factors for the phase.

In practice, the gaps an owner should look for are specific and checkable:

• Completion reported by area or discipline when commissioning needs it by system, so a system can look finished on paper while one of its components is not.
• Punch items carried forward into commissioning without an agreed view of which ones block energization or the introduction of hydrocarbons.
• A commissioning sequence held in a separate plan or spreadsheet, so its logic never links to the construction activities it depends on.
• Operator readiness, such as staffing, training and approved operating procedures, missing from the logic altogether.
• Regulatory approvals and inspections shown as zero-duration milestones rather than activities with lead times and evidence.
• Assumptions about feed gas, refrigerant and cool-down supply that sit with the owner's commercial team rather than in the schedule.

None of these shows on a progress curve, and each can move first LNG on its own.

How do you walk backwards from first LNG?

You walk backwards from first LNG by listing every system the liquefaction train needs, in reverse order of need, and asking for each one what the schedule says, what the evidence shows and who has accepted the difference. Repeat it for each train, since later trains carry their own sequence.

The length of that list is easy to underestimate. The joint case study by nPlan and LNG Canada describes two LNG trains, a 225,000 cubic meter storage tank, a marine terminal for two carriers at once and feed gas from the Coastal GasLink pipeline. Each is a system, or an outside dependency, with its own handover.

1. Fix the milestone. Agree in writing which event is being forecast: first LNG from a named train, the first export cargo, or commercial operation, and use the contract's own definitions of mechanical completion, ready-for-start-up (RFSU) and handover.
2. Connect the plans. Bring construction, commissioning, operator readiness and external approvals into one logic-linked schedule, or at least link them at every handover point.
3. List the systems in reverse order of need. Start from the liquefaction train and work back, roughly through storage, loading, control and safety systems to the utility systems the process units depend on, and on to each system's mechanical completion and RFSU certificate. Your own system breakdown sets the real order.
4. Compare status with evidence. For each system, set the schedule's reported status beside the turnover dossier, open punch items and test records, and note every gap.
5. Forecast the chain. Run a forecast of the whole path to first LNG and identify which paths are most likely to drive the date, not only the deterministic critical path.
6. Test feasible sequences. For the systems that drive the date, ask the commissioning team which alternative sequences are safe and practical, then compare their effect on the forecast.
7. Record and revisit. Record the evidence, the EPC and operator reviewers and the decision for each system in a handover assurance table, and revisit it at every schedule update.

What goes in a handover assurance table?

A handover assurance table records, for each system dependency on the path to first LNG, the evidence you need, who on each side of the handover is accountable for accepting it, and the decision you reached. Putting a name and a decision against every link turns a general argument about the date into specific questions. The first table lists the evidence to ask for.

The second table names who accepts that evidence on each side of the handover.

Each row then gets one of three decisions, and the decision is what makes the table useful. Accept means both reviewers agree the evidence supports the scheduled date. Investigate means the evidence is incomplete or contradicts the schedule, and someone named will close the gap by a stated update. Test a different sequence means the system is likely to drive first LNG and the commissioning team should propose safe alternatives that can be compared on the forecast.

Two rows need a firmer rule. Flare and relief systems should be accepted or investigated before hydrocarbons are introduced, never carried past that point. Regulatory approvals should first be modeled as activities with durations, and only then accepted or investigated.

How do regulators shape the start-up date in the US and Great Britain?

Regulators shape the date because their approvals arrive in stages, after inspections and filings, and each stage is a dependency with its own lead time. Model them as activities with evidence and owners, not as milestones that will simply appear. The rules differ by market.

How do FERC approvals shape a US start-up date?

In the United States, the Federal Energy Regulatory Commission releases commissioning approvals step by step under the conditions of each project's order, so every release is a separate dependency in the start-up logic. A letter from FERC's Office of Energy Projects to Venture Global Plaquemines LNG on November 14, 2024, granted a request to introduce low-pressure fuel gas to the hot oil heaters, "based on FERC staff's on-site inspection," and stated that the approval did not grant authority "to construct, commission, or introduce hazardous fluid to other project facilities at the LNG terminal." Further authorizations would follow only once compliance with the order's conditions was demonstrated.

The same letter reminded the operator that it must also comply with any applicable Department of Transportation requirements under 49 CFR Part 193, the federal safety standards for LNG facilities enforced by the Pipeline and Hazardous Materials Safety Administration. A US start-up schedule should therefore carry each FERC release and the PHMSA requirements as activities with owners, not as a single approval milestone.

How does COMAH shape a start-up date in Great Britain?

In Great Britain, a new upper-tier site cannot start operation until the competent authority has sent its conclusions on the operator's safety report, so that review belongs in the start-up logic. A site holding 200 tonnes or more of liquefied natural gas is an upper-tier establishment under the Control of Major Accident Hazards (COMAH) Regulations 2015. For a new establishment, the operator must send the safety report within a reasonable period before the start of operation, and regulation 9 states that the operator must not start operation until it has received the competent authority's conclusions.

Outside nuclear sites, that competent authority is the Health and Safety Executive and the appropriate environment agency acting jointly, under regulation 4.

Where does an independent forecast help the handover review?

An independent forecast helps by showing which of the dependencies in your table are most likely to move first LNG, so the joint review spends its time on the few links that matter rather than walking every system with equal weight. That is the job nPlan Insights Pro is built for. Teams upload their schedule and get a forecast for the milestones they care about, such as first LNG, with every activity forecast using what nPlan's AI has learned from more than 780,000 past project schedules. Insights Pro identifies which themes and activities influence the outcome the most, and its Driving Paths view visualizes criticality in the schedule based on those forecasts, showing the critical path and the sub-critical paths most likely to affect the milestone.

The forecast then supports the "test a different sequence" decision. Mitigation Strategies, the scenario tester in Insights Pro, lets teams identify, test and communicate the mitigation strategies that give them the best chance of making the date, so each sequence the commissioning team proposes can be compared on the same forecast.

LNG Canada used nPlan during construction of its Kitimat facility. In the joint case study, the project team describes the value in terms of where it directed attention:

"nPlan is part of our toolkit to help us anticipate risk and capture opportunity on the Project. Using data from thousands of other projects helps highlight areas we (as a project team) may have overlooked and allows us to focus our conversations on the areas that could have the biggest impact."

That is the right expectation for a handover review. A forecast only sees what is in the schedule, so it cannot confirm a turnover dossier or decide that a system is safe to energize. The findings go back to the table, where the EPC and operator reviewers test them against site evidence, and commissioning specialists keep responsibility for safety and handover acceptance.

When is this approach not enough?

The walk-back method needs a commissioning sequence to walk, so it suits the period from late construction through start-up rather than the investment decision. Before final investment decision, when commissioning logic is still a set of assumptions, an owner is better served by reviewing schedule maturity and the evidence behind the overall forecast. The contract also changes the owner's role. Where the EPC contractor keeps commissioning responsibility until performance testing, the table becomes the owner's assurance record rather than a joint decision log, but its questions stay the same.

What should you do before the next schedule update?

Before the next update, pick the systems that the current forecast says are most likely to drive first LNG, put them in a handover assurance table with named EPC and operator reviewers, and bring the evidence for each to the next joint review. If the evidence supports the date, you can accept it on the record. If not, you know which links to investigate and which sequences to test while there is still time.

For more on AI in energy capital projects, read nPlan's conversations with Chevron in Forecasting the future: Chevron's big bet on AI for project success, part I and part II, the overview of nPlan for energy projects, and the comparison of traditional risk software and nPlan.

Frequently asked questions

Is first LNG the same as commercial operation?

No. First LNG is the first liquefied product from a train, the first export cargo usually follows it, and commercial operation is a later contractual or commercial milestone. Agree which of these the schedule is forecasting before judging the date.

Should regulatory approvals sit in the start-up schedule?

Yes, as activities with lead times, evidence and owners rather than zero-duration milestones. In the US, FERC staff release approvals to commission and introduce hazardous fluids step by step after inspections. In Great Britain, a new upper-tier COMAH establishment must not start operation until it has the competent authority's conclusions on its safety report.

Who should own the handover assurance table?

The owner should hold the table, with a named EPC reviewer and a named operator reviewer recorded against every system. Where the contract leaves commissioning with the EPC contractor until performance testing, the table becomes the owner's assurance record for contractual conversations rather than a jointly owned decision log.

Can an AI forecast confirm that a start-up date will hold?

No. A forecast shows how likely the date is and which activities and paths most threaten it, based on the schedule you upload. It cannot see a punch list that is missing from the schedule or decide that a system is safe to energize, so the commissioning team and the operator still accept each handover.

Sources

1. LNG Industry: LNG import cargo arrives at LNG Canada facility (April 3, 2025)
2. Shell news release: First cargo leaves LNG Canada (June 30, 2025)
3. Construction Industry Institute: Critical success factors for project commissioning and startup (RS312-1, 2015)
4. FERC Office of Energy Projects letter to Venture Global Plaquemines LNG, Docket CP17-66-000 (November 14, 2024)
5. 49 CFR Part 193: Liquefied natural gas facilities, federal safety standards
6. Control of Major Accident Hazards Regulations 2015, regulation 9
7. Control of Major Accident Hazards Regulations 2015, regulation 4: the competent authority
8. nPlan and LNG Canada: Using AI-led forecasting and risk management to reduce delivery risk during the construction of LNG Canada
9. nPlan Insights Pro
10. nPlan for energy projects
11. nPlan: Forecasting the future, Chevron's big bet on AI for project success, part I
12. nPlan: From gravity hazards to cyclone recovery, Chevron's big bet on AI for project success, part II
13. nPlan: Traditional risk software vs nPlan