C013 Accidents happen . . .Presentation Outline of this Overview)

PREFACE

This presentation is one of several graphic instructional materials available from Starline Software, Ltd. It is an overview of the first of a three part slide series describing challenges, opportunities and examples for improving the value of investigations. Using these materials, one can learn to investigate accidents, fires, explosions, spills, incidents and most other occurrences more efficiently, more effectively and more consistently

For more information about this and other learning tools, visit the Starline web site at

Produced by

Ludwig Benner & Associates for Starline Software Ltd.

Welcome

Exploring the State of the Art

of Investigations:

Part 1 Investigation Challenges

RECOGNIZED INVESTIGATION CHALLENGES

Whether it is a large mishap. . .

Or a fire

Or someone fell off a ladder...

How should I prepare myself?

Why investigate?

What are my objectives?

Scope of Investigations?

What should I investigate?

How should I investigate?

or by reconstruction?

What investigation method should I use?

Unknowns and

Unknown Unknowns

During an investigation...

How much data are enough?

How do I recognize when I know enough?

Investigation outputs:

What should I deliver?

Quality assurance

How can I tell if my work is OK?

How will my work be used?

By now you are wondering...

Recognized Challenges

Emerging Challenges

Managerial demands

Improve Effectiveness

Improve Efficiency

Improve Consistency

Improve Timeliness

Assure Quality

Improve Credibility

Emerging Challenges

USER-DRIVEN CHALLENGES:

User challenges:

Research origins

Main research data sources

2 kinds of investigation deficiencies

Hidden deficiencies

Investigations Hidden Defects

Symptoms of problems

Next research step:

Problems with current technology

Research disclosed a useful surprise:

-time for a break ...

Conceptual differences

WHAT IS BEING INVESTIGATED?

Concept 1

Concept 2

Concept 3

Reasons example of Stochastic Model

Concept 4

Fault Tree Model

MORT Tree Example

Concept 5

MES-based Event Pairs and Sets for analyses

They cant all be right!

Models drive investigation tasks

Chain of Events Model

Stochastic Model

Branched tree model

Multilinear Model

Finding preferred model

Field applications disclosed

-time for a another break

A false premise

Role of Investigation Models

Research challenges were to...

More research challenges

Investigation Process Criteria

Investigation Process Criteria

Search for Investigation Models

Investigation Process Models

Models in use did not satisfy these criteria:

Models for Investigation Process

Investigation Process Frameworks

Attributes of processes within the LEGAL framework

Processes within SCIENCE framework

Processes with SAFETY framework

Scientific Model most attractive:

Importance of Observations

Observation limits

Understanding as an objective:

Prediction as a Goal

Control as a Goal

Other models

New Investigation Goals

The Research challenge:

Technical Challenges

Investigation data variations

Technical Challenges II

New technology needs

Broader challenges

Output Use Challenges

Functional Integration Challenge

Investigation Process

Quality Control Challenges

Investigation Paradigms

The Opportunity paradigm

The Attitude paradigm.

The Methodology

The Data paradigm.

The Analysis paradigm.

Investigators response?

End of Overview

REFERENCES slide