RCA Timeline: Build It in 45 Minutes

An RCA timeline can either protect the investigation or contaminate it within the first 45 minutes, because sequence errors turn witnesses, photos, alarms, and supervisor decisions into disconnected fragments. This guide shows an EHS manager or incident investigator how to build a defensible timeline before the cause discussion begins.

Why a timeline must precede cause analysis

A timeline must precede cause analysis because sequence is the first control against hindsight bias. HSE publishes accident investigation guidance that treats investigation as a structured process for finding immediate, underlying, and root causes, and that structure depends on knowing what happened before deciding why it happened.

The common failure is to open the RCA meeting with opinions. One supervisor says the operator rushed, another says the permit was weak, and the EHS lead starts writing causes before the team has agreed on the order of events. As Andreza Araujo argues in A Ilusão da Conformidade, investigating to understand is different from hunting for a culprit.

Use this 45-minute workshop when the site has already made the area safe, preserved immediate evidence, and assigned one facilitator. It does not replace a full investigation. It gives the investigation a clean spine so later interviews, technical checks, and corrective actions do not float around a guessed sequence.

Step 1: Define the event window

The event window sets the start and end of the timeline, usually from the last verified normal condition to the first stable post-event condition. In a 45-minute workshop, define a window no broader than 24 hours unless the incident involves long-developing exposure, such as fatigue, maintenance deferral, or repeated alarms.

Most teams start too late. They begin at the injury, spill, dropped load, or equipment stop, which hides the decisions that made the event possible. Across 25+ years leading EHS at multinationals, Andreza Araujo identifies that serious events usually announce themselves before the visible loss.

Write the window on the first line of the worksheet. For example, use "from 06:00 shift handover to 10:45 area release" rather than "forklift contact incident." The broader wording keeps the team from treating impact as the beginning.

Step 2: Lock the evidence sources before discussion

Evidence sources should be listed before anyone debates meaning, because source order prevents the loudest memory from becoming the investigation baseline. In the first 5 minutes, name every available source, including photos, witness statements, equipment logs, permit records, CCTV time stamps, radio calls, maintenance orders, and supervisor notes.

OSHA describes incident investigation as a way to identify root causes and prevent recurrence, not merely to document injury. That prevention goal is weakened when the team blends source types without checking reliability. A photo has location value, a witness has sequence value, and a control-room log has time value, but none of them is complete alone.

This step links directly to photos, witness statements, and equipment logs in RCA, because the timeline needs more than one evidence family. Mark each source as available, pending, or unavailable so the team can see where the sequence is strong and where it is still assumed.

Step 3: Create one line per verified fact

Each timeline line should contain one verified fact, one source, and one time reference. A practical format is time, fact, source, confidence level, and open question, which lets the investigator distinguish a confirmed event from a plausible reconstruction.

The trap is to write conclusions as facts. "Operator ignored the alarm" is not a fact unless evidence proves perception and decision. "Alarm 34 activated at 09:12:16 in the control log" is a fact. "Operator stated at 14:20 that he did not hear the alarm" is another fact. The cause discussion can connect them later.

Use 3 confidence levels: confirmed, probable, and disputed. 3 confidence levels are enough for a first workshop because they keep uncertainty visible without turning the worksheet into a legal file.

Step 4: What changed in the first 15 minutes?

The first 15 minutes after an incident often decide whether evidence remains usable. The workshop should record who secured the area, who moved equipment, who spoke with witnesses, who notified leadership, and which controls were restored before the investigator arrived.

What most RCA summaries miss is that the post-event response can create a second event trail. A supervisor may reset a machine to remove energy, a cleaner may wash away spill direction, or a manager may ask a witness a leading question. Those actions may be well intended, although they still change what the team can know.

Connect this check with incident evidence preservation in the first 24 hours. If evidence was moved for rescue, isolation, or emergency control, record why it moved and who authorized it. Do not treat necessary emergency action as misconduct.

Step 5: Add decision points, not only physical events

Decision points belong on the RCA timeline because incidents are shaped by approvals, pauses, assumptions, and handovers as much as by physical contact. Add at least 5 decision points when the event involves permit-to-work, contractor work, maintenance troubleshooting, simultaneous operations, or production pressure.

In more than 250 cultural transformation projects, Andreza Araujo observes that organizations often document the mechanical event better than the managerial decision path. That gap matters because corrective actions aimed only at equipment rarely touch weak escalation, unclear authority, or normalized exceptions.

Ask the group where the work could have been stopped, escalated, redesigned, or delayed. Then write the decision actually taken and the information available at that time. This protects the team from judging yesterday's decision with today's knowledge.

Step 6: How do you mark uncertainty without weakening the RCA?

Uncertainty strengthens the RCA when it is named clearly. Every disputed or missing line should carry an open question, owner, and due time, so the timeline shows what is known at 45 minutes and what must be verified before final causal analysis.

The weak version hides uncertainty to make the report look complete. That creates a false sense of precision and can push the team toward a neat but wrong story. Andreza Araujo's position in Sorte ou Capacidade is useful here: an accident is not bad luck, and relying on a convenient story is another way of counting on luck after the event.

Use direct labels such as "time disputed," "source missing," "sequence unclear," and "decision owner unknown." Assign one person to each label. A 24-hour verification deadline is usually enough for logs and photos, while laboratory tests or engineering reviews may need longer.

Step 7: Separate immediate cause from latent condition

The timeline should distinguish the immediate contact event from the latent conditions that made it credible. James Reason's Swiss cheese model helps here because it frames the event as failed layers rather than as one person's final act.

Unsafe behavior may appear in the sequence, but stopping there treats the symptom and leaves the system intact. In Andreza Araujo's grounding for incident investigation, the question is why before who, because an honest mistake can reveal a flawed process whose weakness will repeat if the investigation stops at the person closest to harm.

Mark each line as physical event, decision, control status, environmental condition, communication, or recovery action. Those labels make it easier to see whether the timeline is dominated by operator actions while management controls, design barriers, and supervision decisions remain invisible.

Step 8: Test the sequence with a contradiction review

A contradiction review checks whether 2 or more sources disagree about time, order, location, or action. Spend 8 minutes looking for conflicts before the team accepts the timeline, because one contradiction can change the direction of the RCA.

ILO describes occupational safety and health management systems through policy, planning, implementation, evaluation, and improvement. Investigation belongs in that evaluation loop, and the loop fails when teams accept contradictions because the first story sounds plausible.

Common contradictions include badge access before reported arrival, CCTV time stamps that differ from witness memory, maintenance logs that show a prior fault, and permit times that do not match actual work start. 8 minutes of contradiction review can prevent days of analysis built on the wrong order.

Step 9: Freeze version 1 and assign verification

Version 1 should be frozen at the end of 45 minutes with a date, owner, unresolved questions, and the next verification deadline. The point is not to declare the timeline final, but to stop the first story from changing silently as memories, pressure, and politics enter the process.

This is where the timeline supports a first RCA plan for a new incident investigator. A disciplined version 1 gives the investigator a working map, while still making room for interviews, technical inspection, and document review.

Every hour without a frozen version 1 increases the chance that witness memory, equipment restart, and leadership pressure will replace evidence with a cleaner narrative.

Comparison: event log vs RCA timeline

An event log records what was reported. An RCA timeline tests the order, source, confidence, and decision context of what happened, which is why it is more useful before causal analysis.

The difference is investigation quality. A log helps memory, but a timeline disciplines judgment before the team decides what failed and what must change.

The timeline is the investigation's first control

An RCA timeline is the investigation's first control because it prevents the team from converting fragments into causes before the sequence is stable.

Build version 1 in 45 minutes, then verify it within 24 hours where evidence is available. If your organization needs to strengthen incident investigation, Andreza Araujo's books, Safety School content, and ACS Global Ventures consulting can help rebuild the routine around facts, decision points, control restoration, and learning that survives pressure.