Using a formal language to support natural language in accident reports
- 1 June 1995
- journal article
- research article
- Published by Taylor & Francis in Ergonomics
- Vol. 38 (6) , 1264-1282
- https://doi.org/10.1080/00140139508925187
Abstract
Accident reports written by official bodies, such as the Air Accident Investigation Branch of the United Kingdom's Department of Transport, are produced in response to all major civil aircraft accidents or incidents. There are many statutory, legal and commercial implications that rest on the analysis, conclusions and recommendations that these reports contain. Air accident reports usually follow a standard format of synopsis followed by factual information, including history of flight and the systems involved, followed by analysis and conclusions. Finally, there are safety recommendations aimed at preventing a recurrence of the accident. Natural language is the primary means of communicating all of these findings. In requirements engineering there is an increasing recognition that natural language is not always an adequate means of expressing some of the detailed reasoning associated with the causal analysis of complex systems. Recent work in software engineering has explored the use of formal, mathematically based, techniques to help to gain the required level of clarity and precision. It is argued that accident reports, like requirements documents, could benefit by the use of formal techniques to complement the usual natural language descriptions. In this paper one specific accident report is considered. The limitations of its natural language descriptions are examined and the use of a Petri Net notation to help to elucidate its ambiguities is explored.Keywords
This publication has 6 references indexed in Scilit:
- Introducing the GRASPIN specification language SEGRASJournal of Systems and Software, 1991
- Extending Petri nets for specifying man–machine dialoguesInternational Journal of Man-Machine Studies, 1988
- The use of Petri nets to analyze coherent fault treesIEEE Transactions on Reliability, 1988
- Why a Diagram is (Sometimes) Worth Ten Thousand WordsCognitive Science, 1987
- Timed petri nets: A solution to the minimum-time-reachability problem between two states of a timed-event graphJournal of Systems and Software, 1986
- Petri NetsACM Computing Surveys, 1977