CE39 - Sécurité globale, résilience et gestion de crise, cybersécurité 2024

Minimizing the risk of attentional bias in emergency call centers – LePetitCamion

Submission summary

"You do send a small truck, right? This phrase, which is harmless and apparently uninformative, is uttered several times by a caller to a CTA-CODIS alert processing center. It is ignored by the operator, although it is repeated several times and pronounced with a particular emphasis. When the emergency services arrive on the scene with their standard response vehicle, they discover that the narrowness of the street prevents them from reaching the scene of the disaster... This example illustrates the "tunnel effect" which corresponds to "a drop in attention at a given moment, implying that the operator fixes on an element while forgetting his environment". Although not very frequent, this phenomenon can occur in the CTA/CODIS, with about 360.000 calls per year (SDIS31 values), with the associated stress and fatigue. Every second counts and operators must handle callers for whom the situation is exceptional. They have to orient the exchange in order to focus the conversation on the priority elements necessary for the qualification of the disaster and the distribution of the help. This project aims to reassure operators by providing them with a tool capable of verifying the consistency between the conversation and the information transcribed on the call-taking form. The tool will have to be able to detect missing information, the presence of unusual elements and to check the mental flexibility and the attention of the operator. Far from the will to dehumanize urgent calls, the objective of this project is to improve the quality and safety of the operator's work by placing her or him at the center of the process. The founding principle of the project is to use the possibilities of AI in terms of voice and speech analysis to draw the operator's attention to potential weaknesses, the operator always remaining the only judge of the situation. Indeed, ergonomic, psycho-sociological and more generally acceptability aspects occupy an important place during this project.

Project coordination

Jean-François Bonastre (INRIA DGDS)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partnership

INRIA DGDS
LIA Laboratoire Informatique d'Avignon
IRIT Institut de Recherche en Informatique de Toulouse
SDIS SCE DEPARTEMENTAL INCENDIE ET SECOURS - Haute Garonne
AnthroPi

Help of the ANR 720,147 euros
Beginning and duration of the scientific project: December 2024 - 48 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter