Dataset Viewer
Auto-converted to Parquet Duplicate
method
stringclasses
9 values
abbreviation
stringclasses
9 values
origin
stringclasses
9 values
best_for
stringclasses
9 values
key_steps
stringclasses
9 values
output
stringclasses
9 values
strengths
stringclasses
9 values
limitations
stringclasses
9 values
5 Whys
5W
Developed by Sakichi Toyoda; formalised within the Toyota Production System in the 1930s–1950s
Simple equipment failures and process deviations where a single causal chain is plausible
Define the problem clearly; ask 'Why did this happen?' and record the answer; repeat for each answer until the root cause is reached (typically 5 iterations); verify the causal chain; identify corrective action
Root cause statement and single corrective action
Fast to apply; requires no specialist training; promotes iterative thinking; low resource cost
Tends to produce a single causal chain and may miss contributing factors; analyst bias can steer conclusions; less suited to complex or multi-causal events; depth depends heavily on facilitator skill
Fishbone Diagram
Ishikawa / Fishbone
Developed by Kaoru Ishikawa at Kawasaki Heavy Industries; introduced in the 1960s; popularised in quality management literature
Brainstorming contributing causes across multiple categories for manufacturing defects, quality failures, and process incidents
State the problem at the fish head; identify major cause categories (commonly: Man, Machine, Method, Material, Measurement, Environment); brainstorm causes within each category; identify root causes from the diagram
Cause-and-effect diagram (Ishikawa diagram) showing categorised contributing factors
Visual and intuitive; encourages team participation; covers multiple cause categories simultaneously; widely recognised format
Does not show causal relationships between factors; can become cluttered on complex incidents; may encourage confirmation bias toward pre-set categories; does not prioritise causes
Fault Tree Analysis
FTA
Developed at Bell Telephone Laboratories (H.A. Watson) in 1961 for the US Air Force Minuteman ICBM programme; widely adopted in nuclear and aerospace industries
Complex technical systems where catastrophic or high-consequence top events must be analysed for all contributing failure combinations
Define the undesired top event; identify immediate causes using logic gates (AND / OR); decompose each cause to lower-level events; identify basic events; calculate probability if quantitative analysis is required
Fault tree diagram with Boolean logic gates; qualitative minimal cut sets; optional quantitative probability estimates
Rigorous and systematic; handles complex multi-cause logic; supports both qualitative and quantitative analysis; well-suited to safety-critical design review
Time-intensive to build for large systems; requires engineering expertise; tree must be validated to avoid missing branches; retrospective application after incidents can be resource-heavy
Failure Mode and Effects Analysis
FMEA
Developed by the US military (MIL-P-1629) in 1949; widely adopted in aerospace, automotive (AIAG), and medical device sectors
Proactive risk assessment of designed systems or processes to identify and prioritise potential failure modes before an incident occurs
List system components or process steps; identify potential failure modes for each; determine effects of each failure; assign severity, occurrence, and detectability ratings; calculate Risk Priority Number (RPN = S×O×D); prioritise and address high-RPN items
FMEA worksheet with ranked failure modes and recommended actions
Proactive and structured; produces auditable risk register; widely accepted in regulated industries; supports design improvement before deployment
Labour-intensive for complex systems; RPN scoring can be subjective; does not model interactions between failures well; typically single-failure focused rather than combination-failure focused
Bowtie Analysis
Bowtie
Conceptual roots in Fault Tree and Event Tree Analysis; Shell's major-hazard risk management practice formalised the bowtie diagram in the 1970s–1980s; widely adopted in process safety
Major hazard and process safety scenarios where both prevention (threats to top event) and mitigation (consequences after top event) barriers must be visualised
Identify the hazard and central top event (the 'knot'); map threat pathways on the left side; map consequence pathways on the right side; identify barriers (preventive and mitigative) on each pathway; assess barrier effectiveness and degradation factors
Bowtie diagram showing threats, barriers, top event, and consequences on a single visual
Communicates risk controls to non-specialists clearly; covers both prevention and response; supports barrier management programmes; widely used in major hazard industries
Can oversimplify complex causal chains; barrier effectiveness assessment is qualitative unless supplemented with quantitative data; large bowtie diagrams can become difficult to manage; does not replace detailed fault tree or event tree analysis
Incident Cause Analysis Method
ICAM
Developed by the Australian mining and resources industry (BHP Billiton group and collaborators) in the 1990s based on James Reason's Swiss Cheese Model and organisational accident theory
Serious workplace incidents and high-potential near-misses in mining, oil and gas, and heavy industry where organisational and systemic factors must be captured alongside immediate causes
Gather facts at the scene; identify absent or failed defences; identify individual and team actions; identify task and environmental conditions; identify organisational factors; formulate recommendations at each level
Structured investigation report with findings mapped to absent defences, individual/team actions, task/environmental conditions, and organisational factors
Explicitly captures systemic and organisational contributors; based on recognised human factors theory; encourages recommendations beyond the immediate cause; structured output aids consistency
Requires trained investigators; more time-intensive than simple RCA methods; can generate large numbers of findings that are difficult to prioritise; less suitable for minor incidents
Tripod Beta
Tripod
Developed by Shell and Leiden University (Professors Wagenaar and Hudson) in the early 1990s; based on cognitive and organisational accident research
Oil and gas, chemical, and major hazard incidents where latent organisational conditions and basic risk factors must be identified alongside immediate causes
Map the incident timeline; identify unsafe acts and immediate causes; trace to preconditions (Tripod factors); identify underlying Basic Risk Factors (BRFs) in eleven categories (e.g. hardware, procedures, training, communication, incompatible goals); develop recommendations
Incident report with causal tree mapped to eleven Basic Risk Factors
Grounded in cognitive and organisational psychology; eleven-factor taxonomy supports systematic scanning; produces comparable data across incidents; widely used in Shell and oil and gas sector
Proprietary framework (licensed tool required for full implementation); eleven-factor taxonomy requires training to apply consistently; cultural fit is strongest in oil and gas environments; less widely adopted outside that sector
TapRoot
TapRoot
Developed by System Improvements Inc. (Mark Paradies and Linda Unger) in the early 1990s; widely used in nuclear, oil and gas, and utilities sectors in the United States
Significant incidents and near-misses where a systematic, guided process is needed to identify root causes and develop actionable corrective actions
Collect evidence and sequence events on a SnapCharT event timeline; identify causal factors; use the TapRoot tree (Root Cause Tree) with yes/no questions to reach root causes across categories; develop corrective actions linked to each root cause
SnapCharT event timeline; Root Cause Tree with identified root causes; corrective action plan
Structured questioning reduces analyst subjectivity; Root Cause Tree covers a broad range of human performance and equipment causes; widely accepted in regulated US industries; produces traceable, auditable output
Proprietary and requires licenced training and software; can be perceived as complex for simple incidents; root cause categories are framework-specific and may not align with other taxonomies; resource-intensive for routine events
Systematic Cause Analysis Technique
SCAT
Developed by the International Loss Control Institute (ILCI) and associated with Frank Bird's Loss Causation Model (1969 updated edition); ILCI later became part of DNV
Workplace accidents and near-misses across general industry where a structured guide-word approach is needed to trace immediate causes to basic causes and management system deficiencies
Describe the incident; identify contact type and immediate causes (substandard acts and conditions); identify basic causes (personal and job factors); identify management system deficiencies using a structured checklist; develop corrective actions
Completed SCAT chart mapping incident to management system root causes
Guide-word structure reduces analyst variance; explicitly connects incident causes to management system gaps; relatively quick to apply after training; aligned to ILCI Loss Causation Model used in many corporate HSE systems
Less granular on human factors than ICAM or Tripod; guide-word lists may not cover highly novel failure modes; output quality depends on analyst familiarity with the model; not as widely cited in academic literature as some other methods

HSE Incident Investigation Methods Reference 2026

Structured reference of 9 well-documented root-cause analysis and incident investigation methods used in health, safety, and environment (HSE) practice.

Details

Citation

SmartQHSE Ltd (2026). HSE Incident Investigation Methods Reference 2026 [dataset]. Zenodo. https://doi.org/10.5281/zenodo.20446335

Disclaimer

Published for reference, training, and research. Consult authoritative sources before operational or compliance decisions. SmartQHSE Ltd accepts no liability for errors or omissions.

Downloads last month
93