Carl Sandom and Roger S Harvey
The IEE, 2004
Carl Sandom and Roger S Harvey
The IEE, 2004
ISBN 0863413293 (boards)
Buy it from Amazon.com
Buy it from Amazon.co.uk
Human Factors, Ergonomics, HCI, MMI... there is a constellation of terms (the first two essentially synonymous, but from opposite sides of the Atlantic) for some of the ways researchers and increasingly engineers look at the human-facing aspects of systems of all kinds. These have inevitably tended to attract practitioners who prefer to deal with people rather than spokes and wheels, so HF has to labour under the burden of the caricature of the plump academic young lady who breezes into the workshop and starts asking questions about lighting and posture. Sandom mentions some other myths: "that it is about common sense, that it is expensive to include in project plans, and that it is time consuming. None of these could be further from reality..." (page xxvi). If anyone needs an argument in favour of human factors, Stuart Arnold's eloquent Preface should supply it.
To its credit, this book is sensible, practical, and aimed squarely at engineers, as the title suggests. The editors have contributed the introductory (Harvey) and closing (Sandom) chapters, and have invited a capable team of authors to contribute the chapters in the body of the book.
Multi-author books are often difficult to review, and unless carefully edited, they can also be hard to read and impossible to obtain an integrated view from. The editors have however been scrupulous in assembling this book, and the chapters have been carefully edited -- the coverage is uniform, and there are cross-references between chapters as necessary to tie the components into a coherent structure.
The body of the book begins with a chapter by Michael Tainsh on HF integration, within a defence context -- that is where it began, and it is also the author's background, but the lessons are, as he argues, largely generic.
Jan Noyes, Kate Garland, and Daniel Bruneau contribute a chapter on Humans: skills, capabilities, and limitations. This covers the usual ground of sensations, perception, attention, memory and errors, with implications for design. Human information processing is briefly illustrated with examples of visual perception illusions.
Jan Noyes also authors a chapter on the HF toolkit, by which is meant a set of conceptual tools necessary to the HF venture. These include subjective methods such as checklists and focus groups; 'objective' methods such as observation, task analysis, human reliability assessment; and empirical methods like experiment, modelling, and fitting trials. Which to use? That key issue is discussed in a couple of pages.
Leslie Ainsworth covers Task Analysis, with a brief opening section on data collection for that purpose: of course, that task overlaps with requirements gathering; HF is only an aspect of system behaviour.
Erik Hollnagel discusses Automation and Human Work, considering the principles involved. This is a relatively 'chatty' chapter, and it is somewhat over-illustrated with clip-art in places.
Sidney Dekker writes on the theme 'To Engineer is to Err', i.e. where errors come from and whether these are the human's fault.
David Embrey looks at Qualitative and quantitative evaluation of human error in risk assessment, covering human reliability analysis (again), criticality analysis, failure models, quantification, and the choice of techniques.
John Wood writes on Control Room Design, covering classical ergonomic issues like room layout, seating, workstation design, and environmental design -- everything from the principles for wall finishes and lighting to the provision of plants and the relevance of corporate image. It's all mentioned quietly and briefly in a non-controversial tone, just like a well-designed control room.
Robert Macredie and Jane Coughlan discuss Human-Computer Interfaces. They consider different types of interface, and give a detailed example (a railway ticket machine).
Martin Maguire writes about Usability, covering its definition, benefits, role, scope, and types -- both formative (enhancing usability within design) and summative (in terms of user experience), as well as user audit of implemented systems. The chapter ends with several examples of usability projects.
Iain MacLeod describes HF verification and validation. Surprisingly, this chapter quotes Standish Group on the reasons for system failure, in the manner of the introduction to a requirements textbook. It then discusses the MoD CADMID life-cycle, relating HF V&V activities to these. The two V's are almost exclusively used together, only being distinguished in the introduction, and never subsequently referenced separately except briefly in the context of task analysis. One could argue that checking Fitness for Purpose (FfP) was a classical case of Validating that the requirements were right, but the author chooses to use the term Verification for this meaning (nothing wrong with that), while the definition of HF Validation overlaps with that by also mentioning FfP.
Ed Marshall contributes a fascinating chapter about the application of simulators in systems evaluation, again inevitably with a strong military history. Simulation has gradually spread to the nuclear and civil aviation industries: like HF in general, it has been adopted only when its lack has been painfully felt.
Carl Sandom's chapter is not a summary or rounding-up of loose ends, but a straight discussion of Safety assessment and HF. The overlap of course arises because hazard events in a fault tree may be caused by human interactions, whether these are misunderstandings of displayed information or wrong commands (slips).
Safety is something of a theme of the book, being covered from various angles in different chapters. The V&V chapter, for instance, splendidly quotes D. Norman on the 'Three Mile Island nuclear power incident':
"They wanted to see what was the matter with their operators. Could we perhaps train them better? I discovered, much to my surprise, that if they had set out to design a plant to cause errors, they could not have done a better job."
That might have introduced or summarised the book.
This book should prove helpful to practising HF engineers, especially those new to the field; it will also be useful to students, and may be of interest to systems engineers in other fields.
© Ian Alexander 2004