CATILaC: ComputeR Aided Techniques for Identifying Latent Conditions

Report Date: 

April 2001

Appendices: 

No

Overview

Introduction to the CATILaC Methodology

By understanding the way that a facility coordinates the work it does, failure events can be placed into a broader organizational context. Once the organizational context is understood, steps can be taken to reduce the possibility of common-cause organizational failures. When this type of analysis is done in the context of a traditional Root Cause Analysis program, substantial insight into the possible causes of operational incidents can be obtained. This software package is designed to guide the user through the process of placing failure events into their organizational context. In doing so, the causes of the events and the human and hardware failures or deficiencies that lead to them will be better understood. Better corrective actions can be developed for all levels of the organization.

The methodology involves both understanding what happened during the course of the event and identifying the hardware failures that contributed to its occurrence. To do this the analyst must identify the sequence of failures that occurred and the causes for each, locate the initiating, or trigger, event, and find the latent failures that became active during the event. Once the event is understood, the human contributions to each of the hardware factors must be identified and analyzed. During the analysis, deficient tasks within work processes are identified. By doing this, the latent conditions that led to the event can be discovered. Figure 1 shows how human contributions are linked to fallible decisions/organizational factors.

CATILaC is focused on hardware failures and the human contributions that cause them rather than on operator actions that contribute to the event. Operations at a nuclear plant, especially post-trigger recovery actions, do not lend themselves to this type of work process analysis. Although it can be done using this software (see discussion of how to include operator contributions in Appendix I), there are other, more complete methods available to do that type of analysis.

Program: 

Type: 

  • TR

RPT. No.: 

2