Human reliability

A great deal of progress has been made toward improving and evaluating the reliability of hardware systems however, the place where systems most frequently fail is in the interface of humans with the system. Human reliability is generally much lower and more difficult to control than hardware reliability. 

If a stimulus is provided to a piece of hardware, the same response is obtained from that piece of hardware day after day, year after year, until it fails. Even when hardware fails, the failure modes, rates, frequencies, and effects can be predicted with a reasonable degree of accuracy. 

System Safely for the 21 Century The Updated and Revised Edition of System Safety 2000. by Richard A. Stephans 

Studies of military personnel indicate that the best peacetime leader or soldier may not be the best in combat. Some seem to function relatively well under the extremely high stress of combat while others who are good performers in a peacetime or administrative environment are very poor performers under high stress. 

The ability to perform when ill or fatigued varies widely. Some individuals performances degrade gradually over time as fatigue or illness sets in. Others may perform at a constant level up to a given point in time, when performance falls off very, very rapidly. 

---

Human error probabilities can also be estimated using methodologies and techniques originally developed in the nuclear industry. A number of different models are available (Swain, Comparative Evaluation of Methods for Human Reliability Analysis, GRS Project RS 688, 1988). This estimation process should be done with great care, as many factors can affect the reliability of the estimates. Methodologies using expert opinion to obtain failure rate and probability estimates have also been used where there is sparse or inappropriate data. 

Swain A. and H. Guttman 1983. Handbook of human reliability analysis with emphasis on nuclear power plant applications (NUREG/CR-1278), Nuclear Regulatory Commission, Washington, DC. 

Frequency Phase 3 Use Branch Point Estimates to Develop a Ere-quency Estimate for the Accident Scenarios. The analysis team may choose to assign frequency values for initiating events and probability values for the branch points of the event trees without drawing fault tree models. These estimates are based on discussions with operating personnel, review of industrial equipment failure databases, and review of human reliability studies. This allows the team to provide initial estimates of scenario frequency and avoids the effort of the detailed analysis (Frequency Phase 4). In many cases, characterizing a few dominant accident scenarios in a layer of protection analysis will provide adequate frequency information. 

B. J. M. Ale, The Implementation of an External Safety Policy in the Netherlands, International Conference on Elazard Identification and Risk Analysis, Human factors and Human Reliability in Process Safety, January 15-17, 1992, Orlando, PL, 173-183, American Institute of Chemical Engineers, New York, NY, 1992. 

Swain, A. D., and H. E. Guttmann (1983). Handbook of Human Reliability Analysis With Emphasis on Nuclear Power Plant Applications. NUREG/ CR-1278. Washington, DC United States Nuclear Regulatory Commission. 

Appendix HI, of WASH-1400 presents a database from 52 references that were used in the study. It includes raw data, notes on test and maintenance time and frequency, human-reliability estimates, aircraft-crash probabilities, frequency of initiating events, and information on common-cause failures. Using this information, it assesses the range for each failure rate. 

Incorporating human reliability is a function of education, training, ergonomics, stress, and physical condition. Incorporating this accurately into PSA is difficult. References for doing this are Gertmann (1994) and Dougherty (1988) as well as the many technical reports. 

The PRA procedures guide, NUREG/ CR-23(X), partitions human reliability analysis (HRA) into four phases (Figure 4.5-1). The familiarization phase, evaluates a sequence of events to identify human actions that directly affect critical process components. From plant visits and review, this part of HRA identifies plant-specific factors that affect human performance such as good or bad procedures used in the. sequence under consideration. The familiarization phase notes items overlooked during systems evaluation. 

HRMS Human Reliability Management System Kirwan, 1992... 

STAHR Socio-Technical Approach to assessing Human Reliability Phillips et al.. 1983... 

Human reliability [lata NJUREG/CR-1278 was supplemented by judgment of system analysts and plant personnel. Human error probabilities were developed from NUREG/CR-12 8, human action time windows from system analysis and some recovery limes from analysis of plant specific experience. Data sources were WASH-1400 HEPs,Fullwood and Gilbert assessment ot I S power reactor Bxp., NUREG/ CR -127K. and selected acro ptice li.it.j... 

Human reliability data Developed by a team of psychologists led by A. Swain. Used WASH-1400 data. Data from NUREG/ CR- i 278 w.i used. 

Bell, B. J. and A. D. Swain, Procedure for Conducting a Human- Reliability Analysis for Nuclear Power Plants, SNL, May 1983,... 

Comer, M. K. et al, enerating Human Reliability Estimates using Expert Judgment. SNL, November 1984. 

J. R. Fragola, 1988, Human Reliability Analysis A Systems Engii ith Nuclear Power Plant Applications, Wiley New York, NY. 

Human Reliability AnMysis Where Shouldst Thou Turn, Rel. Eng, 83-299. 

Fragola, J.R. and E.P. Collins, 1985, Human Reliability Data Framework Development and Application, ANS Trans. 49 p 137 (quoted information is in a supplement). 

Gertman, D. L., and H, S. Blackman, 1994, Human Reliability and Safety Analysis Data HancMjook, Wiley, New York, NY. 

Kirwan, B.. 1992., Human Error Identification in Human Reliability Assessment, A Ergonomics, 23, pp 299-318 and pp 371-381. 

Luckas, W. J. et al., A Human Reliability Analysis for the ATWS Accident Sequence at the Peach Bottom Atomic Power Station, BNL Technical Report A3272, May 1986. 

Phillips, L. D., P. Humphreys, and D, E. Embrey, 1983, A Socio-Technical Approach to A Human Reliability (STAHR), TR 83-4, July,... 

Swain, A. D., 1989, Comparative Evaluation of Methods for Human Reliability Analysis, GRS-71, Gesellschaft fur Reaktorsicherheit (GRS) mbH, Garchin Koln, Germany. 

This book brings together a wide range of tools and techniques used by human factors and human reliability specialists, which have proved to be useful in the context of human performance problems in the CPI. Although many human factors practitioners will be familiar with these methods, this book is intended to provide ready access to both simple and advanced techniques in a single source. Where possible, uses of the techniques in a CPI context are illustrated by means of case studies. 

Despite the lack of interest in human factors issues in the CPI in the past, the situation is now changing. In 1985, Trevor Kletz published his landmark book on human error in the CPI An Engineer s View of Human Error (revised in 1991). Several other books by the same author e.g., Kletz (1994b) have also addressed the issue of human factors in case studies. Two other publications have also been concerned specifically with human factors in the process industry Lorenzo (1990) was commissioned by the Chemical Manufacturers Association in the USA, and Mill (1992), published by the U.K. Institution of Chemical Engineers. In 1992, CCPS and other organizations sponsored a conference on Human Factors and Human Reliability in Process Safety (CCPS, 1992c). This was further evidence of the growing interest in the topic within the CPI. 

---