Overview

• Department: Maintenance - Engineering - Production.
• Profile: Industrial / Mechanical / Mechatronics students with interest in AI and data analysis.
• Duration indicated: 06 Months (internship period: 4-6 months).

Objectives

• Develop AI-driven solutions to predict failures on Zeta/Omega machines and enable a shift from corrective to proactive maintenance.
• Improve key metrics such as reduction in unplanned downtime and increase MTBF (Mean Time Between Failures).

Data & Inputs

• Collect and preprocess historical data: breakdown records, downtime logs, vibration signals, temperature readings, and other sensor data.
• Validate data quality, handle missing values, and create labeled datasets for supervised learning.

Tasks & Methodology

• Implement and compare AI/ML models (regression, random forest, neural networks) using Python or R.
• Perform feature engineering (time-series features, statistical summaries, frequency-domain features from vibration data).
• Train, validate, and test models; evaluate with appropriate metrics (precision, recall, F1, ROC-AUC, prediction horizon accuracy).
• Propose and design a predictive maintenance schedule based on model outputs and risk thresholds.

Required Skills & Tools

• Maintenance knowledge and familiarity with MTBF concepts and maintenance processes.
• Data analysis & AI/ML experience in Python or R; familiarity with libraries such as scikit-learn, TensorFlow/PyTorch, pandas.
• Experience with time-series data, signal processing (for vibration analysis) is a plus.

Deliverables & Expected Outcomes

• A validated predictive model or ensemble capable of forecasting failures before they occur.
• A concrete predictive maintenance schedule and recommendations for integration into existing maintenance workflows.
• Documentation, source code, and a short deployment/hand-over plan for production use.
• Measurable results: demonstrable reduction in unplanned downtime and documented improvement in MTBF.

Number of Interns & Supervision

• Number of interns: 01.
• Work under the Maintenance / Engineering team with guidance on data access and domain validation.

How to Apply

• To apply for this internship, send your application to G-TN-StagePFE@autoelectric.com including CV, relevant project/coursework, and a short motivation specific to this project.