Accounting for Organisational faults in Task Model Based Systematic Analysis of System Failures and Human Errors

Abstract

The overall dependability of an interactive system is the one of its weakest component which is usually its user interface. The presented approach integrates techniques from the dependable computing field and elements of usercentred design to provide a wider coverage of possible faults. Risk analysis and fault tolerance techniques are used in combination with task analysis and modelling to describe and analyse the impact of system faults on human activities and the impact of human deviation or errors on system performance and overall mission performance. A technique for systematic analysis of human errors, effects and criticality is proposed (HEECA). It is inspired and adapted from the Failure Mode, Effects and Criticality Analysis (FMECA) technique. The key points of the approach are: a) the HEECA technique combining a systematic analysis of the effects of system faults and of human errors, b) a task modelling notation to describe and to assess the impact of system faults and human errors on operators' activities and system performance. These key points are illustrated on an example extracted from a case study of the space domain. It demonstrates the feasibility of this approach as well as its benefits in terms of identifying opportunities for re-designing the system, re-designing the operations and for modifying operators' training. Lastly, a discussion presents the main challenges for also taking into account organisational faults in an integrated way with the proposed approach.