What Is Accident Investigation and Why Should It Be Carried Out?
Accident investigation is a structured review process that systematically establishes what happened, how it happened and why it happened following a workplace accident, near miss or safety incident. Its core purpose is not to assign blame, but to identify the systemic changes needed to prevent the event from recurring.
This distinction is critically important. Looking for someone to blame puts the organization on the defensive and shuts down the flow of information. A genuine accident investigation, by contrast, is learning-focused: it traces the chain of events backwards, analyses contributing factors across multiple layers, and produces structural solutions that prevent recurrence.
Accident investigation is essential for the following reasons:
Preventing recurrence: Systemic weaknesses must be identified and eliminated so that the same or similar accidents do not occur in the future.
Legal obligation: In Türkiye, the Occupational Health and Safety Law No. 6331 makes the investigation, reporting and notification of workplace accidents to the Ministry of Labour and Social Security (ÇSGB) mandatory. Notification requirements vary according to the type and outcome of the accident.
Insurance and compensation processes: A thorough accident investigation both strengthens the employer’s legal position and enables insurance claims to be managed effectively.
Safety culture: The quality of the investigation is the clearest indicator of an organization’s safety culture. Learning organizations treat accident investigations not as a disciplinary mechanism, but as a tool for safety improvement.
What Is Root Cause Analysis (RCA) and How Is It Applied?
Root Cause Analysis (RCA) is a systematic analytical method that focuses on uncovering the underlying causes of an event or problem rather than its surface symptoms.
A root cause is rarely a single, specific error; more often it is the process, system or organizational weakness that lies at the base of recurring problems. Measures taken without identifying the root cause merely treat the symptom, and the problem re-emerges in a different form at a different time.
The core principles of RCA are as follows:
- Backward reasoning: Starting from the event, the chain of causes is traced backwards
- Multi-layered analysis: Immediate causes, contributing factors and root causes are distinguished from one another
- System focus: The focus is not on individual error, but on the conditions that drive the system towards failure
- Actionability: RCA findings must result in practical corrective and preventive actions
What Is the Difference Between Accident Investigation and Assigning Blame?
These two approaches produce entirely different outcomes from the same event and add entirely different value to the organization.
The blame-seeking approach: When a workplace incident occurs, the most dangerous reaction is to stop at the obvious answer. A worker falls from scaffolding and the investigation concludes: “the worker did not wear a safety harness.” A chemical exposure occurs and the finding is: “the worker did not follow the procedure.” These superficial conclusions appear satisfying but they miss the mark entirely. They blame individuals rather than fixing the system that set those individuals up to fail.
The accident investigation approach: Investigations, in line with international standards under the IMO Casualty Investigation Code, place root cause analysis and learning ahead of blame.
The practical difference can be summarized as follows: blame-seeking asks, “Who made the mistake?” Accident investigation asks, “Why did the system make this mistake possible?” The second question opens the door to organizational learning and genuine safety improvement.
How Is an Accident Investigation Conducted After a Workplace Accident?
A structured accident investigation process consists of five main stages:
Stage 1: Securing the Incident Scene and Initial Information Gathering
The accident area is preserved to prevent the loss of evidence. Witnesses are identified and statements are taken immediately after the event. Photographs, video and physical evidence are documented.
Stage 2: Reconstructing the Chain of Events
A chronological timeline is created. The sequence of events leading to the incident is set out step by step through the questions “what happened, when did it happen, who was involved.”
Stage 3: Cause Analysis
The immediate cause, contributing factors and root cause are distinguished from one another. At this stage, structured analytical tools such as the 5 Whys, the Ishikawa Diagram or Fault Tree Analysis (FTA) are used.
Stage 4: Defining Corrective and Preventive Actions (CAPA)
For each root cause, a CAPA plan is prepared that is actionable and includes an owner and a completion date.
Stage 5: Reporting and Sharing
Findings are shared with all relevant stakeholders, operational units and, within the framework of legal obligations, the relevant authorities. Lessons learned are disseminated throughout the organization.
What Steps Should Be Taken in the First 24 Hours of an Accident Investigation?
The first 24 hours are the critical window that largely determines the quality of an accident investigation. The following steps must be taken before evidence is lost, witness memories fade and the incident scene changes:
First 1–2 hours:
- Providing medical assistance to the injured and securing the area
- Comprehensively documenting the accident area and its surroundings with photographs and video
- Preserving physical evidence (faulty equipment, chemicals, broken components) in place
- Identifying witnesses and taking their initial statements separately — witnesses should be kept apart, as conversation between them risks contaminating their accounts
First 4–8 hours:
- Forming the investigation team and defining roles
- Collecting relevant records such as work permits, maintenance logs, training certificates and procedure documents
- Copying relevant camera recordings (CCTV, VDR)
- Fulfilling legal notification obligations
First 24 hours:
- Preparing a chronological summary report of the event
- Conducting an urgent risk assessment as to whether similar risks exist in other areas or operations
- Completing witness statements in written and signed form
Why Is Root Cause Analysis More Important Than Surface Causes?
The question “why should we dig down to the root cause?” is answered by understanding the true nature of accidents.
The Swiss Cheese Model developed by James Reason envisages an organization’s safety defences as slices of cheese with holes in them. The holes in the Swiss Cheese Model are created by human error, equipment failures, inadequate training, poor communication, insufficient management commitment and the improper reporting of near-miss events. When the holes in the different layers line up, a path is created for an accident or error.
The practical significance of this model is this: the error of a single individual is rarely the true cause of an accident. Every accident is the product of multiple layers of the system failing simultaneously. Focusing only on the immediate cause — “the worker was not careful” — means ignoring the hole in that one slice and failing to address the structural holes in the other layers.
Stopping at the surface cause produces the following consequences:
- The same accident recurs in different contexts
- The organization deceives itself with the rhetoric of “personal responsibility”
- Genuine system weaknesses remain hidden and accumulate, setting the stage for a larger accident
How Is Root Cause Analysis Carried Out Using the 5 Whys Method?
The 5 Whys technique, developed by Sakichi Toyoda and popularized through the Toyota Production System, is elegantly simple: when an event or problem occurs, ask “Why did this happen?”, then ask “Why?” of the answer, and repeat until you reach the root cause. Five iterations is not a rule but a guiding number; some investigations need three, others seven.
A worked example based on an enclosed space accident on board a ship:
Event: A crew member in the engine room lost consciousness.
-
Why 1 — Why did they lose consciousness? There was insufficient oxygen in the space.
-
Why 2 — Why was there insufficient oxygen in the space? The ventilation system was off and entry was made without gas measurement.
-
Why 3 — Why was entry made without gas measurement? The permit-to-work system was not applied.
-
Why 4 — Why was the permit-to-work system not applied? The supervisor either did not know the procedure or skipped it under operational pressure.
-
Why 5 — Why did the supervisor not know the procedure, or why were they able to skip it under pressure? Training on enclosed space procedures was inadequate, and the safety culture allowed work pressure to override safety procedures.
Root Cause: An inadequate training system and an organizational culture that placed work pressure ahead of safety.
Corrective Action: Not simply saying “follow the procedure,” but overhauling the training programme, auditing the permit-to-work system, and reinforcing safety priority as an organizational value.
What Is the Fishbone (Ishikawa) Diagram and How Is It Used?
While the 5 Whys is best suited to events with a linear causal chain, many workplace incidents in complex industrial environments have multiple contributing factors across different domains. The Fishbone Diagram — also known as the Ishikawa or Cause-and-Effect Diagram — becomes indispensable at this point.
The diagram is visualized as a horizontal arrow (the spine) pointing to a “head” that represents the event, with diagonal branches (bones) representing the main categories.
The traditional 6M categories:
- Man: Training, experience, fatigue, communication
- Machine (Equipment): Failure, inadequate maintenance, unsuitable tools
- Method: Lack of procedure, incorrect instruction, non-standard practice
- Material: Defective material, inadequate specification
- Measurement: Calibration issues, incorrect data, measurement error
- Environment: Weather conditions, lighting, noise, temperature
Ishikawa in a maritime application:
When analysing a ship collision:
- Human: Fatigue of the officer of the watch, breakdown of communication on the bridge
- Equipment: Radar and AIS incompatibility, failure to update ECDIS
- Procedure: Failure to apply the SMS procedure on safe passing distance
- Environment: Heavy traffic, poor visibility
- Organization: Inadequate bridge team planning, excessive working hours
For a simpler and more direct problem, the 5 Whys is often perfect because it drills down to the source quickly. If you are dealing with a more complex problem that has many possible contributing factors, the Fishbone Diagram is a great tool for organizing your team’s brainstorming.
Which Analytical Methods Should Be Used to Prevent Workplace Accidents from Recurring?
Beyond the 5 Whys and Ishikawa, a range of RCA tools is available to suit different levels of complexity:
Fault Tree Analysis (FTA): Using a logical, top-down diagram, it shows which combination of sub-events can bring about an undesired event (an accident). It is a powerful tool for visualizing multiple failure modes in complex systems.
Bowtie Analysis: It visualizes, in the shape of a bowtie, the pathways running from a hazard or root cause to possible events (left side) and from those events to their consequences (right side). It is used to understand where and how risk barriers function.
HFACS (Human Factors Analysis and Classification System): This system, which classifies human error across four layers, is widely applied particularly in high-risk sectors such as maritime and aviation. Unsafe acts, preconditions, unsafe supervision and organizational influences are analysed separately.
Barrier Analysis: It examines why the barriers that should have prevented the accident failed to work. For each barrier, the questions “was the barrier present, was it adequate, was it applied?” are asked.
Why Are Accident Investigations Critically Important in the Maritime Sector?
Shipping is a critical piece of global infrastructure, carrying roughly eighty percent of world trade by volume. This scale magnifies the consequences of accidents proportionately.
According to the most recent report from EMSA (the European Maritime Safety Agency), the data are striking in the extreme:
Between 2015 and 2024, 64.5 percent of casualty events were attributed to human action, while 50.5 percent of contributing factors were associated with human behaviour. Taken together, 78.8 percent of the marine casualties and incidents investigated are linked to the human element.
Between 2015 and 2024, EU Member States reported a total of 26,751 marine casualties and incidents, an annual average of 2,675. In 2024, 2,659 marine casualties and incidents were reported.
These data point to two critical conclusions:
First: The fact that the human element features to such a high degree demonstrates that the “human error” narrative represents a superficial analysis. In reality, human error is a symptom of inadequate training, lack of procedures, fatigue, unsuitable equipment and organizational pressure.
Second: By analysing around a thousand reports in 2024, EMSA focused on specific areas within recurring hazards such as management, enclosed space entry, working at height, collision, grounding and fire. These investigations place root cause analysis and learning ahead of blame.
The IMO Casualty Investigation Code is the international standard for the investigation of marine casualties. This code requires investigations to be conducted with a safety focus, findings to be shared with the IMO, and lessons learned to be disseminated to the global maritime community.
How Is Root Cause Analysis Carried Out for Accidents Occurring on Ships?
The shipboard environment carries additional complexities for accident investigation owing to factors unique to it, such as isolation, 24-hour operation, multinational crews, high dependence on automation and dynamic physical conditions.
Special requirements of maritime accident investigation:
Analysis of VDR (Voyage Data Recorder) data: Bridge conversations, navigation data, engine telemetry and alarm logs are an invaluable source of evidence relating to the period before the event.
SMS (Safety Management System) documentation: Safety management system procedures under the ISM Code, risk assessments and safety drill records are reviewed.
Working hours and fatigue analysis: Actual working hours are compared against STCW rest hour requirements; fatigue is a common contributing factor in collisions, groundings and enclosed space accidents.
Special protocol for enclosed space incidents: The enclosed space entry procedure, gas measurement records, permit-to-work documents and the condition of rescue equipment are examined separately.
The role of shore management: Maritime accidents are often not solely ship-based; operational pressure, maintenance budget constraints, port schedule impositions and communication from the shore team must all be investigated as contributing factors.
Why Should Near-Miss Events Be Investigated?
A near miss is an event that ends without physical damage or injury, but in which an adverse outcome was averted by a hair’s breadth. Many organizations dismiss such events as trivial; this is a serious safety mistake.
According to the safety triangle model developed by H.W. Heinrich, behind every serious accident lie numerous minor incidents and near misses. The logic underlying this approach is that the same system weakness, under different conditions, results at times in a near miss and at other times in a serious injury.
Proactive safety management: the combination of these theories encourages organizations to incorporate minor incidents and near misses into their reporting system, because they are indicators of potential major accidents.
The value of near-miss reporting manifests itself in the following dimensions:
Early warning: A near miss reveals where the system’s layers of defence are eroding. This information provides an opportunity to intervene at low cost, before a serious accident occurs.
Barometer of safety culture: Active near-miss reporting is a sign of a strong safety culture, one in which personnel feel comfortable reporting safety concerns without fear of punishment.
Quantitative analysis: Near-miss data accumulated over time reveal patterns on the basis of specific equipment, operations or locations, and lay the groundwork for proactive risk management.
What Information Should an Accident Investigation Report Contain?
A comprehensive accident investigation report should include the following sections:
Executive Summary: A brief description of the event, key findings and critical recommendations.
Incident Description: The date, time and place of the accident, the persons and equipment involved, and the chronological flow of events.
Investigation Process: The composition of the investigation team, the evidence collected, the interviews conducted and the analytical methods used.
Immediate Causes: The unsafe acts or conditions that directly triggered the accident.
Contributing Factors: The secondary factors that made the immediate cause possible or aggravated it.
Root Causes: The system, process or organizational weaknesses that lie at the base of the contributing factors.
Corrective and Preventive Actions (CAPA): For each root cause, the recommended action, the responsible person and the completion date.
Lessons Learned: Transferable conclusions for other operations or locations exposed to similar risks.
Appendices: Photographs, diagrams, witness statements, and samples of relevant procedures and records.
Who Should Be on an Accident Investigation Team?
An accident investigation team should be built on a balance of impartiality, technical competence and an understanding of the organizational context.
Core team members:
Investigation leader: A person who directs the process as a whole, is independent, and is independent of the line manager responsible for the accident area.
Technical expert(s): One or more people with technical expertise in the area where the accident occurred. In a ship accident, an expert in machinery, navigation or deck operations; in a shipyard accident, an expert in the relevant technical discipline.
OHS specialist / Safety officer: Experienced in legal requirements, risk assessment methodology and CAPA processes.
Employee representative: A union representative or safety representative, who brings the worker’s perspective and the reality of day-to-day practice into the investigation.
Management representative: A senior representative who understands the organizational context, resource decisions and policy framework.
When is an independent external investigator needed?
An independent professional accident investigator should be brought in for serious accidents, events affecting multiple parties, insurance disputes, or situations where there are questions over the ability of internal resources to conduct an impartial investigation.
Is Human Error Really the Root Cause of Accidents?
This is one of the most debated questions in the maritime and occupational safety sectors. The data appear to support both “yes” and “no”; however, a deeper analysis reveals how complex the issue is.
Human action appears as the cause of 55 percent of marine casualties, followed by system/equipment failure at 27 percent.
Yet this figure is misleading on its own. The “human error” label usually describes an individual’s act and conceals the system conditions behind it.
The Swiss Cheese Model shows that accidents result from multiple layers of defence failing simultaneously; it emphasizes concurrent factors such as management decisions and design errors, and does not reduce events to a single chain of events.
According to the current perspective adopted by experienced accident investigators:
- Human error is an outcome; it is rarely a cause in and of itself
- The same person operates the same system in different ways under different conditions
- Instead of asking “Why did the human make an error?”, one should ask “Why did the system lead the human into error?”
- Fatigue, inadequate tools, poor lighting, unclear procedures and performance pressure are the systematic precursors of human error
How Should Corrective and Preventive Actions (CAPA) Be Defined?
Root Cause Analysis and Corrective and Preventive Action work together: they provide a systematic way to investigate problems methodically, identify causes and demonstrably prevent recurrence.
The five characteristics of effective CAPA are as follows:
1. Directly addressing the root cause Each CAPA must be clearly mapped to the identified root cause or contributing factor. A generic recommendation such as “train the workers” does not resolve why the training gap arose in the first place.
2. A hierarchy-of-controls approach CAPAs should be prioritized according to the hierarchy of controls:
- Elimination (removing the hazard) — the strongest control
- Substitution (replacing with a less hazardous alternative)
- Engineering controls (physical barriers)
- Administrative controls (procedures, training, warnings)
- Personal protective equipment — the weakest layer, not sufficient on its own
3. Being defined in SMART format Specific, Measurable, Achievable, Realistic, Time-bound.
4. Assignment of responsibility and resources For each action, a clear owner, completion date and required resources must be defined.
5. Verification of effectiveness After a CAPA has been implemented, whether the defined action has actually worked must be monitored through a verification mechanism.
What Are the Most Common Mistakes in Accident Investigations?
1. Reaching a conclusion too early Accidents that appear as simple as they first look are closed without digging deep enough. The conclusion “the worker was careless” should be the starting point of a genuine investigation, not the end point.
2. Allowing evidence to be contaminated Failing to preserve the accident scene in time — removing equipment before it is photographed, allowing witnesses to talk to one another — leads to irretrievable losses of information.
3. Blaming only the individual Holding a single person responsible while ignoring organizational contributing factors conceals systemic problems and leaves the organization exposed to a recurrence of the same accident.
4. Leaving the CAPA superficial Symbolic measures such as “all workers were reminded” or “the procedure was updated” do not deliver a genuine system change.
5. Failing to disseminate lessons learned Once the accident investigation is complete, the findings turn into a file that remains only at the incident site. No systematic sharing and organizational learning mechanism is established.
6. Delaying witness statements Human memory undergoes extremely rapid transformation. A statement taken after 24 hours cannot substitute for one taken within a few hours.
7. Lack of independence Having the investigation conducted by the unit or managers directly responsible for the accident seriously undermines the impartiality of the findings.
How Long Should an Accident Investigation Take to Complete?
The duration of an accident investigation varies considerably according to the complexity of the event, the number of parties involved and the depth of the technical analysis required.
A general guideline framework:
| Accident Type | Target Completion Time |
|---|---|
| Near miss / minor incident | 5–10 working days |
| Lost-time injury accident | 15–30 days |
| Serious accident / multiple injuries | 30–60 days |
| Very serious accident / fatality | 60–180 days (or longer) |
Factors affecting the timeline:
- Requirements for expert analysis or laboratory examination
- Multi-party events and the involvement of multiple authorities (port, insurer, judiciary)
- Flag state investigation requirements in maritime accidents
- The coordination complexity of events with an international dimension
The risks of closing too quickly: Investigations rushed under pressure often lead to inadequate root cause analysis and superficial CAPA. Time pressure reinforces the tendency to favour the “easiest” explanation.
How Do Accident Investigation and Root Cause Analysis Benefit Businesses?
Accident investigation and RCA are not a cost item, but an investment with a calculable return.
Direct benefits:
- Reduction in injuries and workforce losses through the prevention of recurring accidents
- Long-term reduction in insurance premiums (as the accident record improves)
- Avoidance of administrative fines and legal sanctions
- Strong documentation for P&I and H&M insurance claims
Indirect benefits:
- Increased staff engagement and job satisfaction as safety culture strengthens
- Efficiency gains through the improvement of operational processes
- Reinforcement of customer and partner trust
- Sustaining a favourable picture in Port State Control (PSC) inspections and insurance audits
Value specific to the maritime sector: For a shipyard, ship operator or maritime service company, a comprehensive accident investigation capability is a directly decisive factor in ISM Code compliance, PSC inspection results and charter negotiations.
What Should Be Considered When Procuring Professional Accident Investigation Services?
Where internal capacity falls short or independence is critical, procuring external accident investigation services is the most appropriate solution. The following criteria should be considered when selecting the right service provider:
Methodological competence: A team that effectively applies structured analytical tools such as the 5 Whys, Ishikawa, FTA, Bowtie or HFACS, and can select the methodology according to the complexity of the event.
Sector experience: Accident investigation experience specific to the maritime sector; knowledge of the ISM Code, STCW, SOLAS and the IMO Casualty Investigation Code.
Independence and impartiality: The absence of any conflict of interest between the investigator and the parties concerned. This is critical both for the credibility of the findings and for the admissibility of the report in legal proceedings.
Reporting standards: A report format of a standard acceptable to insurers (P&I, H&M), classification societies and legal authorities.
CAPA follow-up capability: A service model that does not stop at producing a report but supports the implementation process and verifies effectiveness.
Confidentiality and legal assurance: A clear protocol on how information gathered during accident investigations will be used. In investigations running in parallel with legal proceedings, the legal status of the documents should also be assessed.
Orionis Shipping Accident Investigation and Survey Services
At Orionis Shipping, we provide accident investigation and root cause analysis services through our expert team, which has an in-depth understanding of the unique operational realities of the maritime sector. From ship accidents to shipyard incidents, from near-miss analysis to comprehensive safety survey reports, we produce ISM Code-compliant, independent and actionable findings.
To direct your safety investments correctly, prevent recurring incidents and hold a strong position in insurance and Port State Control inspections, take advantage of our professional accident investigation service.
Contact us to learn more about our services and the accident investigation process.
This content is for informational purposes. For your occupational safety legislative obligations, refer to the Occupational Health and Safety Law No. 6331 and the related regulations, as well as the current provisions of the Ministry of Labour and Social Security.
Tags: accident investigation, root cause analysis, 5 whys method, ishikawa diagram, CAPA corrective preventive action, workplace accident investigation, maritime accident analysis, near miss, ship casualty investigation, ISM code safety, EMSA casualty report, human error analysis, Swiss cheese model, occupational safety survey
