A Predictive Decision Approach for Firefighters Using Variable Selection Technique

Firefighters are using empirical approaches to manage their human resources for covering the daily interventions. With data digitization, many data are incremental available and could be treated for prognostic purposes. Prognostics and health management could be an efficient way to provide a help in the decision-making for firefighter interventions. We hence have an access to a data set of the number of interventions of firefighters in the Bourgogne Franche-Comte region of eastern France. The aim of this paper is to propose a variable selection technique for firefighters interventions prediction. For that purpose, we estimate the contribution of each variable on the number of interventions for selecting major variables and show their impact on intervention prediction. The proposed model is applied on a case study, where a efficient variables selection is obtained for prediction.