top of page

Focus on First Open Call winners: AI-Driven Maintenance Optimization for Manufacturing



The experiment will be done at BF in Slovenia, EU. Our production is digitalized to a considerable extent as it has MES and SCADA systems as key production applications. The diversity and the amount of data gives us great opportunity to create AI-driven applications.

The goal of the experiment is to develop, explore and introduce Maintenance Copilot, an AI-supported, dynamical workflow maintenance solution that helps maintenance personnel to focus on key issue. It indicates which of the existing maintenance issues may have the most impact on production performance thus helps prioritizing the scheduling of maintenance tasks.

We will focus on one production step – CNC turning, which currently has 4 CNC DMG Mori Seiki NLX machines. The key concept of the solution is to build up the knowledge base about machine maintenance process and exploit it with AI. At the core is the AI learning algorithm that monitors the operation steps of the maintenance personnel and learns which of performed actions lead to successful completion of task. The application user interface will guide the personnel through operation steps for maintenance task and record taken decisions/actions. At the end of maintenance task, the user classifies the detected issues and estimates the condition of key machine components using predefined classifications. User also has possibility to take its own route of maintenance process thus bypassing predefined workflow, but still record performed actions; this is to allow experts to “do the things their own way” while digital solution catches, stores and learns from their expertise. Regardless of the choice, the AI algorithm records the operation and learns to discriminate between good and bad decisions based on classified outcomes. In addition to human input, we feed the AI algorithm with machine parameters from existing digital production systems (MES, SCADA) to estimate “machine health” and tie it to certain event. In turn, we task the AI algorithm to extract and discern data subtleties, i.e. features, and indicate the maintenance personnel on which machine area or element to focus when performing maintenance.



1. Contribution to increased digitalization of the SME

  • Improve the sustainability of processes and products, and their efficiency,

  • Make industrial processes more agile, secure and resilient to future changes,

  • Make manufacturing jobs more attractive for humans, regardless of age, gender or social background, through better human-machine interfaces and more intuitive interaction with digital tools.

    2. Technological impact

  • As it is process- and human-centric it considers particularities of production and maintenance processes of specific SME. Classical approaches must be adapted to specific situations.

  • The constituent technologies are well proven in various solutions. In contrast to classical approaches, there is no need for development of special sensorics and special AI models of production machinery.

      3. Economic impact

  • Reduction of downtimes in BF manufacturing related to breakdowns, with a goal of 10-14% reduction, meaning 60-80 thousand EUR of added value/year.

  • Know-how retention: AI-assisted work orchestrator will help retain key knowledge within the company.

        4. Commercial impact

  • Increase production capacity with existing infrastructure; produce and sell more bolts– in the amount of ~50 thousand EUR/year.

  • Dissemination of executed experiment as branding BF as “tech advanced”.

Blaj Fasteners



  • LinkedIn
  • link-icon-svg-2
bottom of page