Digital Transformation and Machine Learning for Manufacturing Plants
Agile micro-applications that complement the core ERP, and ML algorithms that raise production efficiency.
A digital transformation program prepared for a global chemical manufacturer with multiple plants in Türkiye: micro-SaaS applications that solve real problems on the shop floor, and statistical analysis and machine learning that turn idle production data into predictive insight.
The program starts with web and mobile applications that each handle a single process flawlessly (inventory counts, shipping, quality control, maintenance) without touching the plants’ existing cumbersome systems. In parallel, production, quality and energy data are combined into a single data model, and models for demand forecasting, OEE analysis and anomaly detection are built. Alignment with the plant’s global standards and existing ERP ecosystem is a prerequisite of the design.
Highlights
- Process improvement with micro-SaaS, without replacing the ERP
- ML-based insight from production data
- Architecture suited to multi-plant operations
- Development process aligned with global standards
Technology
- Python
- FastAPI
- Flutter
- scikit-learn
- PostgreSQL
Solving a similar problem?
Next project
Radiation Analysis Simulator →