Problems to be Solved and Project Context
Digitalization is driving increased demand for microchips. Shortening the product lifecycle and the high variety of customer specific devices leads to a growing need for high-mix low-volume (HMLV) semiconductor production. Moreover, microchips fulfil increasingly critical safety-related functions, which makes their quality particularly important. Yet, competition is fiercely adding cost pressure.
The project “Autonomous Integrated Scheduling for Semiconductor Industry” (AISSI) proposes to develop, integrate and apply novel AI based approaches in semiconductor manufacturing that builds on European quality-thinking from the automotive sector.. By applying reinforcement learning in the context of autonomous and continuous improvement of integrated production and maintenance scheduling, European semiconductors will gain a competitive advantage in terms of efficiency, cost effectiveness and quality.
Business Relevance and Market Impact
AISSI will expand the capabilities of European semiconductor manufacturers to plan, produce and react to an unprecedented microchip demand. Consequently, AISSI will help to strengthen the market position of the European semiconductor industry globally. AISSI also contributes to the creation of a new generation of digital Small and Medium Enterprises (SMEs), offering large test beds to validate their specific solutions, access to research and potential paths to future large-scale commercialization and growth.
Innovative Aspects and Expected Major Technical Outcomes
The core innovation and outcome of this project is a new fundamental concept and system in semiconductor production and planning, with specific focus on the interaction between manufacturing, maintenance and customer fulfilment requirements. This concept will be built upon a new level of interaction between subject matter expertise (human expert knowledge) and AI, and as such enabling inherent and flexible levels of autonomy, quality and efficiency in the production system.
This new level of interaction will be realized as a semantically and physically inspired system of holistic information processing, which will support and enable a new degree of real-time capability and information consistency during the manufacturing process, and a highly efficient ‘plug-in-capability’ for innovative new AI methods to be deployed.
Major European semiconductor manufacturers, leading researchers in the field of semiconductor manufacturing, data science and AI, and highly innovative SMEs collaborate in this research project to create integrated scheduling solutions that aim to go beyond the state-of-the-art. We aim to create an industry-driven AI eco-system, adhering to the roadmaps of various European smart industry transitions programs.