We are seeking a highly skilled Industrial Data Scientist with deep expertise in BOPP film manufacturing processes and a strong background in polymer physics. This individual will work at the intersection of manufacturing, process engineering, and advanced data science, developing solutions that improve line efficiency, maximize conversion ratio, reduce waste, enhance demand forecasting, and maintain golden-run performance across plants.

This is a high-impact role requiring strong domain expertise, applied machine learning skills, and the ability to translate research into deployable solutions on real production lines.

Key Responsibilities

Process Data Analysis: Collect, clean, and analyze data from BOPP film lines (extruders, chill rolls, MDO, TDO, corona treaters, winders, slitting lines) to identify efficiency losses, process drifts, and variability affecting throughput and quality.

Predictive Modeling: Build and deploy ML models for:

- Demand forecasting (client-wise, grade-wise)

- Process parameter optimization (temperature, line speed, tension, draw ratio, gauge control)

- Predictive maintenance of bearings, tenter chains, winder motors

Physics-Based & Hybrid Modeling: Use polymer physics, film stretching dynamics, and heat transfer to create hybrid (physics+ML) models for improved accuracy and explainability.

Apply DoE and Response Surface Methodology to identify golden-run operating conditions.

Quality Optimization: Correlate process parameters with film properties (gauge profile, haze, COF, WVTR, OTR) and recommend recipe adjustments to reduce scrap and second-sales.

Early Warning & Anomaly Detection: Design real-time systems to flag process drift, high scrap rates, or quality non-conformance before they cause major losses.

Collaboration & Implementation: Work closely with plant engineers, R&D, and operations teams to translate insights into production-ready SOPs. Integrate ML outputs with SAP PPDS/MES for closed-loop scheduling and optimization.

Visualization & Reporting: Build intuitive dashboards (Power BI, Tableau, Streamlit) to communicate KPIs such as OEE, conversion ratio, waste %, energy consumption, and forecast accuracy.

Key RequirementsEducation

Master’s or PhD in Data Science, Applied Physics, Chemical Engineering, Materials Science, or a related field.

Strong background in polymer physics or film processing mechanics preferred.

Technical Skills

- Advanced proficiency in Python/R, scikit-learn, TensorFlow/PyTorch

- Time-series forecasting (ARIMA, Prophet, LSTM)

- Familiarity with SCADA/MES/IoT historian data systems and SAP PP/PPDS integration

- Understanding of orientation, crystallization, heat transfer, tensile properties

- DOE, multivariate regression, SPC, Six Sigma and response surface methodology

- MLOps practices for model deployment

Soft Skills

- Problem-solving and analytical mindset