How to run a pre-startup safety review in 30 days

OSHA's Process Safety Management standard treats pre-startup safety review as a required gate before highly hazardous chemicals enter a new or significantly modified process. This guide shows senior EHS leaders how to run that gate in 30 days without turning PSSR into a signature chase.

Why PSSR fails when it becomes a final checklist

A pre-startup safety review is not a ceremonial walkdown at the end of a capital project. Under OSHA 29 CFR 1910.119(i), the employer must confirm that construction and equipment meet design specifications, safety systems are ready, procedures are in place, and training has been completed before startup.

The weak version of PSSR happens when the team waits until commissioning pressure is already unbearable. At that point, open actions become negotiation items, the operations manager wants production, maintenance wants access, contractors want demobilization, and the EHS manager is asked to bless a condition that should have been challenged two weeks earlier.

Co-host Andreza Araujo's work in Safety Culture: From Theory to Practice is useful here because it separates declared culture from operated culture. A company may declare that startup readiness is nonnegotiable, although the real test is whether leaders protect the PSSR gate when schedule pressure arrives.

Step 1. Define the startup boundary before the review begins

The first PSSR decision is the boundary of what will be started, energized, pressurized, heated, loaded, or chemically introduced. A boundary that is too vague allows risk to move through interfaces, which is why the review should name equipment, utilities, bypasses, temporary hoses, software changes, contractor scope, and operating modes.

Most checklists fail at this point because they ask whether the project is complete rather than asking which exact operating envelope will exist at the first introduction of hazardous material. What matters is not the construction package on paper, but the physical and organizational condition in which the startup crew will work.

For a 30-day process, the EHS manager should issue a one-page PSSR charter by day 3. It should name the process unit, the startup condition, excluded work, responsible owner, required approvers, and any temporary control that will exist only during startup.

Step 2. Connect PSSR to management of change instead of duplicating it

PSSR should confirm that the change is ready for startup, while management of change should prove that the change was understood before approval. The two processes overlap, but they answer different questions, and treating them as the same meeting weakens both.

OSHA's PSM structure places MOC and PSSR in separate elements because change approval and startup authorization create different risks. MOC asks whether the proposed change is acceptable, while PSSR asks whether the built and trained reality matches the approved basis before exposure begins.

By day 5, the PSSR leader should obtain the closed MOC package and mark every assumption that must be field-verified. If process safety information, procedures, relief device data, interlock logic, or training records were changed during MOC, each item needs a PSSR evidence owner.

Step 3. Build the evidence register before the field walkdown

A PSSR evidence register is the single list of proof required before startup, not a folder of documents that nobody reads. For a covered process, that register usually includes design basis, redlined P&IDs, relief system documentation, operating procedures, alarm setpoints, interlock proof tests, emergency response arrangements, mechanical completion records, and training evidence.

Across real conversations on Headline Podcast, the recurring leadership pattern is that safety fails less from absent slogans than from weak verification. Dr. Megan Tranter and Andreza often return to the same practical question, who can show the evidence when the room becomes uncomfortable?

By day 8, assign an owner and due date to each evidence line. A useful threshold is visible in the register itself: 100% of startup-critical evidence should have an owner before the first PSSR walkdown, because unowned proof becomes an open action by default.

Step 4. Run the process hazard review with startup conditions in mind

The process hazard review must be checked against actual startup conditions, because startup rarely matches stable operation. Lines are being flushed, valves may be in temporary positions, contractors remain nearby, alarms may be newly configured, and operators are still learning how the modified system behaves.

If the project already used HAZOP, Bow-Tie, or LOPA, the PSSR should not repeat the study mechanically. It should ask whether the safeguards assumed in that study are installed, tested, understood, and available in the startup mode that will actually occur.

For smaller modifications, a focused What-If analysis may be enough, although the PSSR still needs written proof that the team considered abnormal startup, loss of utilities, wrong valve lineup, temporary bypass, delayed contractor removal, and emergency response.

Step 5. Verify field condition with operations, maintenance, and contractors together

The PSSR walkdown should be a joint verification by operations, maintenance, engineering, EHS, and affected contractors. A single-discipline walkdown misses interface risk, especially when one group assumes another group owns a valve status, blind removal, instrument calibration, insulation gap, or temporary scaffold.

As co-host Andreza Araujo has observed across more than 250 cultural transformation projects, the dangerous gap is often between departments rather than inside one department. PSSR should expose that gap before startup, not document it after the first upset.

By day 15, run the walkdown with the evidence register in hand. Photograph exceptions, mark each finding as startup-critical or post-startup corrective, and refuse vague labels such as monitor, review later, or acceptable for now when the finding affects containment, isolation, alarm response, or emergency access.

Step 6. Separate startup-critical actions from improvement actions

A PSSR action list needs two categories: startup blockers and post-startup improvements. Without that separation, leaders either delay startup for minor issues or accept serious gaps because the whole list feels impossible to close.

The decision rule should be written before the meeting. Startup-critical actions include anything that affects process containment, safe operating limits, emergency shutdown, relief protection, operator response, isolation, fire protection, rescue, or legally required training under OSHA PSM.

This is where many organizations hide risk inside optimism. A risk matrix can help prioritize, but risk matrix blind spots become dangerous when the team scores a poorly understood startup condition as low probability simply because the plant has not started yet.

Step 7. Train for the first abnormal hour, not only normal operation

PSSR training should prepare operators and supervisors for the first abnormal hour of startup. Procedure reading is necessary, although the higher-value test is whether the crew can explain safe limits, alarm response, stop criteria, communication channels, and escalation triggers without improvising under pressure.

OSHA PSM includes training as a separate element, and PSSR depends on that element being complete before startup. In a 30-day PSSR plan, training evidence should be reviewed by day 20, not collected the night before introduction of hazardous material.

A practical drill is to ask the startup supervisor three questions in the field: what condition stops startup, who has authority to stop it, and which instrument or field sign proves the condition exists. 3 answered questions reveal more than a signed attendance sheet because they test decision readiness rather than presence in a classroom.

Step 8. Hold the authorization meeting as a risk decision

The final authorization meeting should be a risk decision with named accountabilities, not a document routing exercise. The meeting confirms closed blockers, accepts or rejects residual conditions, names the startup authority, and records which leader owns any post-startup action.

On Headline Podcast, the phrase real conversations matters because some safety decisions require visible disagreement before they become reliable. If engineering, operations, maintenance, and EHS cannot disagree in the PSSR meeting, the disagreement will return during startup as informal workarounds.

By day 28, the authorizing leader should sign only after hearing the evidence owner for each blocker. By day 30, the startup brief should be issued to all affected workers and contractors, with stop criteria stated in plain language.

PSSR in 30 days compared with checklist-only PSSR

Each week a startup proceeds without this evidence discipline, the organization teaches people that schedule pressure can outrank process safety, which becomes harder to correct after the first successful but lucky startup.

Conclusion

Pre-startup safety review works when leaders treat it as a startup authorization system whose evidence, field verification, training, and decision rights are visible before exposure begins.

Headline Podcast exists for this kind of leadership conversation, where safety and operational pressure meet in real decisions. For more discussions on leadership, culture, and safer work, listen at Headline Podcast.