Why human error modeling has failed to help systems development

Abstract

Until the 1980s, human reliability analysis focused upon individual erroneous actions. More recently, attention has shifted to the managerial and organizational contexts that create the latent conditions for such failures. Unfortunately, these developments have had little impact upon many industries. The problems of technology transfer are less due to commercial neglect than to the failure of human-factors' research to seriously consider the problems of systems development. For example, most error-modeling techniques are poorly documented. In consequence, errors are likely to be made when designers apply error modeling techniques. There are further ironies. Many of these techniques depend entirely upon the skill and intuition of human factors' experts. The lack of professional accreditation procedures prevents companies from assessing the quality both of those experts and their advice. Until these practical problems are addressed, increasingly esoteric models of human and organizational failure will be of little practical benefit. Over the last 5 years, a number of researchers have become increasingly concerned to support technology transfer between human error modeling and constructive systems development [1]. As a result, workshops were staged in Glasgow (1997), Seattle (1998) and Liège (1999). This special edition presents a collection of papers from these meetings.