Paperback - 272 pages
Published by: Ashgate Publishing Company
Publication Date: December 1997
This book should be compulsory reading for all engineers (whether maintenance engineers, design engineers, production engineers, mining engineers, chemical engineers or whatever) working in heavy and hazardous industries.
The author, Professor James Reason, has worked extensively for the last 25 years in the area of human error and the way people and organisational processes contribute to the breakdown of complex, well-defended technologies such as commercial aviation, nuclear power generation, process plants, railways, marine operations, financial services and healthcare institutions. His research work has been carried out in collaboration with a variety of organisations including Shell, British Railways, British Airways, Singapore Airlines, and the Bureau of Air Safety Investigation in Australia. His background and experience is impeccable, and it shows in this book.
Although written by an academic, the book is aimed at (in Reason's words) "real" people, and particularly someone with a technical background - and is ideal for those with an engineering background. In particular, it is aimed at those who are responsible for ensuring that adequate defences are in place to prevent "organizational accidents" - those incidents which, in Reason's words are "those comparatively rare, but often catastrophic events that occur within complex modern technologies, such as nuclear power plants, commercial aviation, the petrochemical industry, chemical process plants, marine and rail transport, banks and stadiums".
The primary theme of the book is to make the point that, in most organisations today, there is a belief amongst technical managers that the main threat to the integrity of their assets is posed by the behavioural and motivational shortcomings of those at the "sharp end" - Production Operators and Maintainers. For them, the often-repeated statistic is that human errors are associated with 85-90% of all events. However Reason argues that the majority of these events are actually caused by the conditions under which people work - the "latent" conditions - and that it is easier to change these latent conditions than to manage the minds of individual workers - in short, Reason argues, most human performance problems are technical, rather than psychological - as Reason is a PhD in psychology, this is an interesting finding. His thoughts are also aligned with a number of approaches to Root Cause Analysis in use today, including the Apollo method, and, in particular, the Proact method.
The book introduces a number of Reason's concepts and models for the effective management of organizational risk, including his now famous "Swiss Cheese" model, which considers the multiple defences that organisations have in place to prevent organizational accidents (these could be 'hard' defences, such as physical barriers, interlocks, alarms, personal protective equipment, etc., or 'soft' defences, such as regulatory surveillance, licensing, training, permit-to-work systems, and other personal oversight) as being 'layers' of defence - each with its own built-in weaknesses, or "holes". On most (almost all) occasions, incidents are prevented from becoming organizational accidents because they hit, and are captured, by one of these layers of defence. However, when all the holes in the defences line up, then there is potential for an organizational accident. (Incidentally, this is one of the biggest problems with the RCM approach to failure management, because it typically analyses only one failure mode at a time, and assumes that all other defences are operating properly - it also typically focuses only on the 'hard' defences, not the 'soft' defences - but I digress). The key thing to understand from this is that BIG accidents do not require BIG errors in order for them to occur - in fact some of the worst accidents have occured because of the unhappy alignment of a number of relatively small, and individually unimportant acts, omissions and/or circumstances.
Throughout the book there are a large number of case studies and examples discussed (including Piper Alpha, the Nakina Derailment of 1992, the Battle of Agincourt in 1415 (!), Three Mile Island, Chernobyl, multiple aviation accidents (including United Flight 232), Flixborough, Bhopal, Clapham Junction and others) which all superbly illustrate the points being made, and bring Reason's concepts into sharp focus.
In addition, Reason's arguments are backed up in many many instances by hard, factual, quantitative research, giving the book an authority that is hard to argue with. Of particular interest to Maintenance people will be Chapter 5 of the book, entitled "Maintenance can Seriously Damage your System". Those who are familiar with RCM, and the work of Nowlan and Heap will recognise that, across industries, over 50% of equipment failures demonstrate a higher than normal probability of failure in the early part of their life. This chapter documents research that confirms this finding in a different manner - for example, compilation of research from the Institute of Nuclear Power Operations (INPO) in Atlanta and the Central Research Institute for the Electrical Power Industry (CRIEPI) in Tokyo, indicated that 60% of Nuclear Power Plant "significant event reports" were associated with maintenance, testing and calibration activities. In addition, a Boeing study indicated that of 276 inflight engine shutdown events on Boeing aircraft, various forms of faulty installation accounted for over 70% of all contributing factors. And these statistics occur despite the existence of rigorous repair and maintenance procedures. All Maintenance personnel should read this chapter, and be scared - very scared.
But this book does not just describe theories, concepts, problems and make us very scared. In the later part of the book, Reason moves on to suggesting some approaches which may assist in reducing the probability of organizational accidents. These include error management tools such as Tripod-Delta (used by Shell worldwide), Human Error Assessment and Reduction Technique (HEART - developed by Jeremy Williams), and Maintenance Error Decision Aid (MEDA - used in the Aviation industry).
The final chapters in the book describe the Regulator's unhappy lot, and make some highly salient points regarding the creation of a "Safety Culture" within an organisation.
As mentioned at the start of this review, this book is a "Must Read" for anyone working in hazardous industries.