Leader
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum.
Hosting Facilities
Mauris et nulla tellus. Nulla consectetur placerat nunc placerat bibendum.
eTwin-Cycle
Creation of Insights
AN ADAPTIVE DIGITAL TWIN FOR AUTONOMOUS CIRCULAR MANUFACTURING
Challenge: Product-production Digital Twins
eTwin-Cycle is an adaptive digital twin solution for autonomous circular manufacturing in injection moulding.
The project combines AI, real-time machine data, 3D quality inspection, and reinforcement learning to optimise process parameters cycle-by-cycle and compensate for recycled material variability.
Demonstrated in industrial conditions at ELVEZ, the system targets ≥30% recycled plastic content, ≤5% scrap rate, and >10% reduction in specific energy consumption, enabling more sustainable and efficient automotive plastics production.
Leader
iThermAI
Hosting Facilities
The project will be hosted at ELVEZ’s industrial injection moulding facility equipped with injection molding machines and a ZIVID One+ 3D scanner and a thermo camera. The production environment includes defined control plans, quality procedures, operator training systems, and 1000 lux inspection workplaces for decorative automotive components, enabling real industrial validation of the solution.
Background
European injection moulding manufacturers face increasing pressure to use recycled plastics while maintaining strict automotive quality standards.
However, recycled materials introduce variability that often causes dimensional and decorative defects, higher scrap rates, unstable production, and increased energy consumption. Existing processes rely heavily on manual parameter adjustments and operator experience.
eTwin-Cycle addresses this challenge by enabling AI-driven, adaptive process optimization for stable, energy efficient, and circular manufacturing.
Solution
eTwin-Cycle is an adaptive digital twin solution designed to optimize injection moulding processes for circular manufacturing. The system combines real-time machine data, OCR-based parameter acquisition from legacy controllers, 3D quality inspection, and AI-driven process optimization. At the core of the solution is a reinforcement learning (RL) agent that continuously adjusts critical process parameters such as temperature, pressure, injection speed, cooling time, and dosing conditions on a cycle-by-cycle basis. This enables the system to compensate for variability in recycled plastic materials and maintain stable production quality. The solution will be validated on Krauss Maffei injection moulding machines producing decorative automotive parts with strict dimensional and visual quality requirements. eTwin-Cycle supports operators through real-time monitoring, process recommendations, and future-ready autonomous control capabilities. By reducing scrap, stabilizing production, and lowering specific energy consumption, the project enables manufacturers to confidently increase recycled plastic content while improving sustainability, operational efficiency, and competitiveness.
Objectives
- Develop an adaptive digital twin for injection moulding process optimisation.
- Enable autonomous cycle-by-cycle adjustment of machine parameters using AI and reinforcement learning.
- Increase recycled plastic content in automotive parts to at least 30% without quality loss.
- Reduce scrap rate from 20% to below 5% through stable and intelligent process control.
- Lower specific energy consumption by more than 10% through process optimisation and waste reduction.
- Demonstrate the solution in a real industrial environment at ELVEZ on injection molding machine.
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.