The Investigation Process Research Resource Site
A Pro Bono site with hundreds of resources for Investigation Investigators
Home Site Guide FAQs Tutorials Old News Contribute Forums

. . . .. . . . . . . . last updated 8/8/09

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

Research Resources:

Search site for::

Launched Aug 26 1996.


Paper presented at
Surety Assessment Center - Sandia National Laboratories
Symposium on High Consequence Operations Safety
Albuquerque, New Mexico , July 12 - 14, 1994


Ira J. Rimson, P.E., Managing Director
the validata corporation
P.O. Drawer 1315
Springfield, VA 22151
703-978-2944 Fax: 703-323-8417
 Ludwig Benner, Jr., P.E., President
 Ludwig Benner & Associates
 12101 Toreador Lane
 Oakton, VA 22124
 703-758-4800 Fax: 703-758-4800



Uncritical acceptance of poorly executed investigations and unverified investigative reports undermines effective prevention initiatives. Analytical inconsistencies cast doubt on conclusions. Unsubstantiable conclusions foil attempts to correct genuine systemic deficiencies, generate controversy, and waste time and money.

Traditional concepts of mishap etiology fail to account for the complex interactive processes characteristic of high consequence operations. Absent investigation quality controls, mishap definitions are subjective, lacking rigorous scientific proofs. Inadequate analyses of mishap processes result in outputs which cannot achieve effective prevention. Investigation users should demand that analytical logic of output reports and recommendations be documented to a measurable standard.

Multilinear Events Sequencing (MES) technology enables investigators to capture, evaluate, analyze and catalog data in real (investigation) time. It enables investigation managers and users to assure that investigation data are complete or, if not, that the missing elements are identified for more rigorous research. A byproduct of MES technology is its documented inventory of facts, supporting data and the logical analyses by which outcomes are developed, which can be adapted to fulfill widely varying reporting requirements both in the public and private sectors.

This paper proposes MES as the preferred approach both to mishap conceptualization and investigation. By modeling both desired and undesired (mishap) processes, MES methodology enables identification of investigation logic and sufficiency inadequacies, discovery of potentially fatal errors of investigative analyses, and elimination of ineffectual prevention arguments. It establishes specific focal points for monitoring effectiveness of implemented preventive actions.


Problem: Currently, similar High-Risk Catastrophic Events recur continually despite "investigation-based" recommendations ostensibly designed to preclude recurrence. No measures of effectiveness are implemented to verify the effectuality of recommendations.

Need: A methodology which applies the rigor of scientific method and data language to catastrophic event investigation and analysis, in order to establish robust construction and evaluation criteria for recommendations which will achieve future prevention or mitigation.


Multilinear Events Sequencing (MES) rigorously standardizes investigation data acquisition, structural and analytical requirements and validation procedures. These investigation attributes have heretofore been relegated to the exclusive judgment of the individual investigator or support staff. MES invokes a structured methodology to support several fundamental postulates:

(1) All outcomes are the result of a dynamic process comprised of interactions among a matrix of events;

(2) Planned processes are designed to achieve desired outcomes;

(3) Mishaps result when changes during a planned process initiate an unplanned process which ends in an undesired outcome;

(4) The basic structural unit for describing a dynamic process is an "event block", consisting of an "actor" performing an "action" at a time;

(5) A dynamic process links many "actors" performing many "actions" over time;

(6) All "actors" relevant to the process are somewhere, doing something, at all times during the process;

(7) Events occur in logical sequence, usually based on natural laws, which can be verified analytically.

(8) A process must be described completely and accurately before effective action can be taken to change its future attributes; and

(9) If you can't depict a process in a flow chart, you don't understand it!


MES procedures standardize data acquisition, organization and evaluation to achieve verifiable outputs in real (investigation) time. These steps are summarized below.

1 For each event, identify at minimum the actor, the action, and its starting time relative to the mishap process

2 Insert each Event Block onto an array according to its actor and its start time to facilitate its immediate ordering and analysis.

3 Evaluate each pair of temporally contiguous events for precede/follow consistency.

4 Evaluate each preceding event for both necessity and sufficiency to cause the following event(s).

5 Failure of either (or both) of the above tests demonstrates data insufficiency. Each failure represents a discontinuity in the logic flow.

6 Link all events which pass temporal, necessity and sufficiency logic tests to document causal relationship.

7 Generate alternative hypotheses to account for data insufficiencies. .

8 Identify data required to test likely hypotheses, and define search, test, simulation or research plans needed to generate them. Apply similar logic and sufficiency testing to identify the relative probabilities of alternative scenarios and select best fit.

9. Analyze event pairs and sets on the array to identify actor, action and relationship problems which "enabled" the mishap process to proceed unimpeded. Overlay mishap process model on planned action process model to identify points of divergence.

10 Attach verified problem diamonds to specific events or links on the array to document intervention points for recommended corrective actions

11 Analyze each intervention point for alternatives to mitigate the undesired outcome.

12 Integrate recommended corrective actions into future system operation to assess the effectiveness of the recommendation options and their tradeoffs, and establish the most efficacious alternative. Substituting the chosen alternative into the events array will demonstrate monitoring points for future measurements of control effectiveness


MES provides investigators, and the user of the investigation work products, with tools by which they may:

(1) Identify the data needed to describe the mishap process;

(2) Apply scientific data language rigor to mishap process data;

(3) Verify mishap process data relevance to the undesired outcome;

(4) Arrange data to verify their relative chronological accuracy;

(5) Identify the "beginning" of the mishap process; i.e., the point at which the planned process translated into the mishap process;

(6) Identify gaps or uncertainties in the description of the mishap process, and define additional data needed to surmount them;

(7) Identify point(s) at which mishap process progress became irreversible;

(8) Identify and define actors, actions and relationships which contributed to the progress of (or failed to arrest) the mishap process;

(9) Evaluate and discriminate among specific corrective actions devised to address problems demonstrated by current mishap;

(10) Institute quality controls over the complete investigation/analysis/recommendation process to enable quantitative measurement of both predicted and achieved effectiveness of competing recommendations.

(11) Provide validated models of mishap-related event sets for future use by designers, auditors, supervisors, trainers, procedures writers, investigators, litigants, etc.


MES systematically unites disparate elements of the investigation process. It provides a progressive Quality Control capability at each step of the way, from initial data acquisition to final validation of the effectivity of changes instituted to mitigate recurrence. MES may be performed manually, or with assistance of computer programs which simplify data recording and manipulation, logic and hypothesis testing, quality control and data management tasks. In either case the success of the outcome is dependent on the practitioner's willingness to confront traditional analytic approaches and demand verifiable accuracy.


Illustrations are from the following copyrighted references, and have been reproduced here with permission.

Benner, L, and I. J. Rimson, "Quality Management for Accident Investigations". International Society of Air Safety Investigators FORUM: Part 1, V.24, #3 (October 1991); Part 2, V.25, #1 (February 1992)

Benner, L., Four Accident Investigation Games: Instructor's Manual. Ludwig Benner & Associates, Oakton, VA. (c) Ludwig Benner & Associates

Benner, L., 10 MES INVESTIGATION GUIDES, Ludwig Benner & Associates, Oakton, VA (c)1979 by Ludwig Benner & Associates.