The Investigation Process Research Resources site
A Pro Bono site with hundreds of resources for Investigation Investigators
Home Page Site Guide FAQs Older News Site Inputs


to advance the
State-of-the-Art of
investigations, through
investigation process

Research Resources:

Search site for::

Launched Aug 26 1996.


WELCOME to the IPRR site.

Check the Site Guide (above) for an quick overview. L Benner

  Updated through . . . . January 2017

Everyone involved in any way with the investigation of accidents or incident for safety purposes should be aware of several issues with present practices:
  1. The interchangeability of investigation and analysis in academic and practitioner dialogues is confusing, and masks the conflation of three very different functions involved in present safety investigations practices and studies, namely: 1) the reconstruction of what happened from source data created during an occurrence to create a documented description of what happened; 2) the analyses of that description to satisfy various user needs, and 3) use of the descriptions, analyses and other input data to prepare and implement responses to what was reported. Each function requires a different knowledge base, skill and tool set (KST) and data, but most present practices demand they be performed by individuals untrained in the specific KST sets, to the detriment of the respective functional outputs and conflated reports.

  2. Over two dozen investigation methods are in use, each one producing different outputs. Investigation practices including 3CA, 5 WHYS, AEB, APPOLO, ASSET, ATHEANA, CAS-HEAR, FRAM, HPEP, HPES, HPIP, MORT, MTO, PPS, PRCAP, PROSPER, PSA, RASP, RCA, SOL, STEP, TRIPOD, TOP-SET, MES/STEP, CAST, ACCI-MAP, SHELL, SCAT, ISIM, Why/Because, TapRooTŪ, FACS, HF-compatible RCA and perhaps others. This rsesults in a cacophony of investigation data bases that include mixtures of facts, implied allegations, subjective judgments and characterizations, unobservable attributes, semantic ambiguity and inconsistency, unverifiable or unreproducible content, and other incompatibilities. The consequences for safety research and report utility are largely unrecognized.

  3. A US governmental agency recently imposed a new regulatory requirement for the reporting of accident/incident requiring use of one of the investigation methods without any comparative technical analysis to support its selection, which is likely to stifle promising investigation process research interest or supporting resources to improve a demonstrably inferior practice.

  4. The Wikipedia article on accident investigation/accident analysis is based on antiquated concepts and practices, and should be treated circumspectly. (1/2017)

  • The References List section has been significantly updated with numerous additions from my papers, annotated to indicate the papers' main messages. New works have been added to the on-line section of my archives at my wb site. (2/16)

  • A new paper What do STAMP-based Analysts Expect from Safety Investigations?describes variations in accident investigation reports and other input data challenges that STAMP/CST analysts need to recognize in selecting candidate accidents for analyses, and options for addressing these challenges. It can be viewed at Procedia Engineering, Volume 128, p 93-102 (11/2015

  • A new paper arguing that some legal constructs be purged from safety investigations with the title Is It Time To Purge Legal Constructs From Safety Investigations is posted at the web site, with some comments. (11/15)
  • An article ISASI: 50 Years of Investigations describing the most noteworthy aviation accident investigations since 1964, and the role of investigations can be read in July-September 2014 ISASI Forum, p10-23 (Aug 2015)

  • Thought-provoking Alaska newspaper article for investigators and researchers in the Alaska Dispatch News, 2014 NTSB: 'Inadequate safety management' contributed to fatal trooper helicopter crash, November 5, 2014. Last viewed at (Nov 6 2014)

  • A new paper proposing a shift from a "prevention centric" to a "user centric" accident investigation paradigm is posted in the presentation section at Other proposed changes include replacement of accident causation models with an input/output accident process model, standardization of investigation input data, and standardization of investigation input data integration into an explanatory description of what happened, which could then be analyzed for whatever purposes desired by the analyst. A video and slide presentation are available at that site.(11/2013)

  • An interesting report by European Commission Joint Research Centre -IE (2011) Comparative Analysis of Nuclear Event Investigation Methods, Tools and Techniques, Interim Technical Report, EUR 24757 EN compares around 20 different accident investigation methodologies and methods. (8/2013)

  • See previous News posted here at News Archives .


    Material posted here is not necessarily endorsed by this site. The page coding is copyrighted, and is now available free under the terms of a GNU General Public License. View Copyright

    If you use or download any material from this site, please reference this site as the source, and honor copyrights.

    NOTE: This site does not engage in, support or report on "ACCIDENT CAUSATION RESEARCH" )



    View Investigator's Top 10 Axioms