From design errors to design opportunities using a machine learning approach
PAKM'06 Proceedings of the 6th international conference on Practical Aspects of Knowledge Management
| Hi-index | 0.00 |
As current engineered systems (e.g. aviation systems) have been equipped with automated and computer-based artefacts, human-system interaction (e.g. human computer interaction) has been an important issue. Design errors that are attributable to human-system interaction failures are not pure engineering design issues, but a multidisciplinary subject with related other areas such as management, psychology, physiology or ergonomics. To identify such design errors (called design-induced errors) in accident reports is important for designing more reliable systems. However, the lack of precise definitions of the concept of design-induced error and the diversity of expression of such failures make it difficult to retrieve relevant documents from accident reports. This paper describes how an ontology and annotation scheme can help to overcome such limitations. Engineering designers can be assisted by the developed ontology and annotation scheme to reason on the issues of design induced error.