Job : stage
Date : April 1, 2026
Contact Mail : liseth.pasaguayo@emse.fr , tel : 0444444444
Stage Master M2 “Probabilistic modeling for the analysis ..." - Saint Etienne
“Probabilistic modeling for the analysis of events related to the industrial process of knitting machines and knitting production defects”
Planned start date of internship: April 1, 2026
5-month internship with internship allowance (€4.50/hour).
75% of public transportation costs covered (conditions apply),
Staff lounge (sports and cultural activities, employee benefits for leisure and social events),
1 day off per month (subject to the supervisor's approval)
Location: Saint-Étienne
Supervisors: Liseth PASAGUAYO, Anis HOAYEK, and Mireille BATTON-HUBERT, EMSE/LIMOS
The work will focus on three main areas:
1. Probabilistic modeling of machine events:
This will involve modeling events occurring during production as stochastic processes, taking into account their frequency, duration, and sequentiality. Markov chains will be used in particular to represent transitions between different machine operating states and characterize their normal dynamics.
2. Detection of anomalies in event sequences:
Based on the probabilistic models constructed, anomaly detection methods will be developed to identify atypical machine behavior (rare sequences, unusual transitions, excessive durations). Anomaly detection will be considered as a tool for highlighting operating conditions that may increase the risk of faults, rather than simply identifying outliers.
3. Causal analysis and impact on production defects:
Finally, the link between detected anomalies, machine events, and defects observed during quality control will be studied using causal graphs and appropriate statistical models. This analysis will make it possible to estimate the impact of certain events or abnormal conditions on the probability of defects occurring, and to identify interpretable risk factors for improving industrial processes, as well as the root causes of anomalies.
• Second year of a master's degree and/or third year of engineering school,
• Training in applied and/or fundamental mathematics (statistics and probability),
• Strong skills in R and/or Python programming,
• Additional skills in machine learning would be highly appreciated.