Leader
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum.
Hosting Facilities
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum.
LUMINA
Creation of Insights
Lifespan and Useful life Management with artificial Intelligence and digital twiN Analytics
Challenge: Digital Twin based lifespan analysis tool
LUMINA delivers an AI-powered digital twin platform that predicts the remaining useful life of refurbished refrigerator components in Arçelik’s return centre. By combining survival analysis, machine learning and digital twin visualisation, LUMINA enables quality control staff to identify reusable components, reduce waste and extend product lifespans. The solution will be validated in Arçelik’s refurbishment line, released under the Apache 2.0 open-source licence and contributed to the AI-on-Demand platform. LUMINA supports the European Green Deal by promoting circularity, reducing electronic waste and enabling data-driven decisions across refurbishment operations.
Leader
Hosting Facilities
LUMINA will be tested and validated at Arçelik’s high-TRL playground, specifically the consumer electronics refurbishment line at the return centre, which processes approximately 70,000 returned products annually. The Arçelik facility provides the operational context, quality inspection workflows, domain expertise and the cloud infrastructure for production deployment.
Background
European manufacturers face increasing pressure to reduce waste and improve circularity. Around 20% of materials used in EU manufacturing become waste, with electrical and electronic equipment generating 12 million tonnes annually. Arçelik’s return centre processes around 70,000 products yearly, but sorting and quality decisions follow predefined rules without AI-assisted lifespan prediction. Inspection outcomes are captured manually, which prevents pattern analysis across product categories or seasonal trends. As a result, viable components may be discarded and refurbishment decisions cannot benefit from data-driven insights. LUMINA addresses this gap, in line with the Green Deal.
Solution
LUMINA is a modular, containerised platform built around four components. The Data Management Hub, based on PostgreSQL, centralises product records, service histories, inspection results and model outputs. The AI Lifespan Prediction Engine uses scikit-learn, scikit-survival and MLflow for model development, versioning and inference, packaged as Docker containers for portability across development and production environments. The Data Processing and Analytics Platform, with a Next.js frontend and FastAPI backend, supports data ingestion and provides dashboards for lifespan predictions, product overviews and data quality indicators. The solution is designed to augment human judgement rather than replace it: predictions and confidence levels are displayed to quality control staff, who retain final decisions on reuse, refurbishment or recycling, with all AI recommendations logged for audit purposes. LUMINA will be validated in Arçelik’s refurbishment line cloud environment and released as open-source under the Apache 2.0 licence, with selected assets contributed to the AI-on-Demand platform for reuse across European manufacturing.
Objectives
- Develop and validate a digital twin-based lifespan analysis tool to predict the remaining useful life of refurbished refrigerator components.
- Validate the solution in Arçelik's refurbishment line to identify reusable components that standard sorting criteria typically miss.
- Release the platform as open-source software under the Apache 2.0 license.
- Contribute selected models and assets to the AI-on-Demand platform to support broader ecosystem integration.
- Support Green Deal objectives by promoting circularity, reducing waste, and extending product lifespans.
KPIs achieved
Development
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum. Sed sit amet consequat ex. Pellentesque id odio felis. Etiam tellus dui, maximus non vehicula ut, suscipit vel ex. Maecenas quis lorem ut sem aliquet fermentum. Duis vel nulla hendrerit augue aliquet posuere. In erat arcu, accumsan ut rhoncus in, fringilla at elit. Morbi non gravida ligula. Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum. Sed sit amet consequat ex. Pellentesque id odio felis. Etiam tellus dui, maximus non vehicula ut, suscipit vel ex. Maecenas quis lorem ut sem aliquet fermentum. Duis vel nulla hendrerit augue aliquet posuere. In erat arcu, accumsan ut rhoncus in, fringilla at elit. Morbi non gravida ligula.
Feedback
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum. Sed sit amet consequat ex. Pellentesque id odio felis. Etiam tellus dui, maximus non vehicula ut, suscipit vel ex. Maecenas quis lorem ut sem aliquet fermentum.
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum. Sed sit amet consequat ex. Pellentesque id odio felis. Etiam tellus dui, maximus non vehicula ut, suscipit vel ex. Maecenas quis lorem ut sem aliquet fermentum.
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum. Sed sit amet consequat ex. Pellentesque id odio felis. Etiam tellus dui, maximus non vehicula ut, suscipit vel ex. Maecenas quis lorem ut sem aliquet fermentum.