Advanced AI Architectures for Integrated and Enhanced Manufacturing Operations (AIMOps)

Purpose and goal

AIMOps project aims to design, develop, and deploy advanced AI architectures to enable predictive and prescriptive decision making across manufacturing operations by promoting synergy between them, leading to improved system-level performance. Goals include creating scalable AI for multimodal data from production, maintenance, and quality domains, building and validating robust prototypes, deploying them by applying MLOps and a long-term lifecycle perspective.

Expected effects and result

The expected results include the architectural design of AI models, prototype development and deployment, and knowledge dissemination materials. These results will enable industrial partners to make proactive shop-floor decisions, leading to higher productivity and quality, reduced costs and downtime, and enhanced operational performance. This will also strengthen Sweden’s competitiveness in industrial AI and foster innovation.

Planned approach and implementation

The project will integrate data from all shop-floor operations and apply advanced AI models to capture complex links between process parameters, machine health and product quality. Both simple and advanced models will be tested on industrial use cases to balance complexity, cost, and predictive accuracy. Successful models will be deployed using MLOps frameworks, with dashboards and user interfaces enabling actionable insights.