Troubleshooting in Mechanics: A Heuristic Matching Process

Denis Besnard (1) , Lucile Cacitti (2)

1 School of Computing Science [Newcastle]
2 LPC - Laboratoire de psychologie cognitive

This paper deals with expert operators' reasoning processes in troubleshooting. We want to know more about the information that experienced operators use. In a previous study we studied electronics troubleshooting. We found that experts used surface cues in order to implement heuristic rules even if the latter are not relevant to the current fault. We now wish to study the field of mechanics. An experiment was conducted in order to test the hypothesis of a heuristic rule-based level of control responsible for errors among experts. This paper adopts a naturalistic and ergonomic point of view about troubleshooting in mechanics. Our results show that expert mechanics operators' errors rely on heuristics in the troubleshooting process. This strategy relies on an automated matching process between symptoms and procedures. Although this strategy is usually powerful, it is rigid and may lead the operator to not locate the fault of the latter is atypical

Mots clés: Expertise; Mechanics; Rule-based reasoning; Troubleshooting

Domaines: Informatique

Denis Besnard, Lucile Cacitti. Troubleshooting in Mechanics: A Heuristic Matching Process. Cognition, Technology and Work, 2001, 3 (3), pp.Pages 150-160. ⟨10.1007/PL00011530⟩. ⟨hal-00722885⟩