to advance the
Search site for::
Launched Aug 26 1996.
|INVESTIGATION THEORY SECTION
A brief introductory essay
Interest in investigation methodologies continues to grow, so a subsection on investigation methodology definitions and evaluations was added March 07
Investigation theory: why bother?
This section of the IPRR addresses the theoretical basis for investigation practices. Widely differing and continuing variations in perceptions of the nature of the accidental phenomenon to be investigated have been reported  and discussed in message traffic at this site. Differences about accident causation models  offer further evidence of the a need to examine the underlying theoretical concepts of accidents, fires, explosions, spills, and other accidental phenomena for which investigations are conducted. Lacking a unifying theoretical foundation, investigation practices have been characterized as aggregations of individual's best efforts, resulting in diverse individualized practices.
Consequences are significant for all users of investigation data, from obvious one for system managers to obscure ones for accident researchers and fire investigators. Current controversy about "probable cause" determination as reported in the recently released Rand study of NTSB needs, also reinforces this need. Challenges to investigation work products like those mounted by Ladkin in Germany and Johnson in Scotland further demonstrate the broad need for something better.
Investigation problems involving the objectives, scope, methodology, output specifications and uses of work products are been reported several times in 1975 and 1980  For example, overt 30 years ago, I personally took classroom polls of individuals involved in investigations to learn the methodology they used. Almost without exception, they were unable to name their methodology, and hardly any realized they had any methodological choices. Further probing disclosed significant differences among their methods, which were summarized and published. (Benner 1980) In 1985 additional research into accident models and investigation methodologies was published, describing the models at 17 government agencies and organizations. Today, the growing list of "investigation methodologies" and my observations of how they evolved further suggest the need for fundamental research into the underlying theory of investigations of all kinds.
Only three major conceptual departures to the ambiguous historical view of accidents as unplanned and undesired events have been advanced - the energy exchange  change+energy based  and accident process concepts , all of which resulted from research initiatives into the fundamental bases for investigations. Each has led to significant advances in the value of investigations. Meanwhile, "refinements" of traditional reconstruction methods by supplementing them with computerized graphics displays of object behaviors, and law-oriented cause ideas with new fads like root cause analysis have occurred. To date, there has been no sound research into the comparative merits of each of these concepts or others which may arise in the future, or of the foundations of investigations. I am aware of only two works which address the issue of criteria for comparison.
My hope is that posting this topic at this site, and using the message board available, will attract individuals who can and will undertake or contribute to the theoretical research needed to bring order-of-magnitude improvements to the state of the art.
Ludwig Benner, Jr.
January 11, 2000
 Benner, L., UNDERSTANDING ACCIDENTS: A CASE FOR PERCEPTIONS AND METHODOLOGIES, Accident Causation, Special Publication SP-461, Society of Automotive Engineers, Inc., Warrendale, PA February, 1980. Examined 200 accident definitions. Describes reasons for investigating, accident models, 5 perceptions of accident phenomenon, 7 investigation processes.
 See Chapter 11 of Manuele, Fred, ON THE PRACTICE OF SAFETY, Van Nostrand Reinhold. New York, 1993 ISBN 0-442-01401-5 for a discussion of causation models.
 See Haddon, W. J. ON THE ESCAPE OF TIGERS: AN ETIOLOGICAL NOTE, Technology Review, May 1970 on the control of energy exchanges to improve vehicle safety.
 Johnson, W. G. SAN 821-2, MORT THE MANAGEMENT OVERSIGHT AND RISK TREE prepared for the U. S. Atomic Energy Commission under Contract No. AT(04-3) 821, February 12, 1973. defined accident as change-led energy exchange.
 Benner, L., ACCIDENT THEORY AND ACCIDENT INVESTIGATION, Proceedings of the Annual Seminar of the Society of Air Investigators, Ottawa, Canada, 1975 existing theories, process charting and application and expectations of theory.