In this paper, we derived competences from previously developed competence models, ensuring the effective use of advanced technologies in future factories to improve the sustainability of their business models and strategies. Based on the analysis of the Hogan competence model and competence models for sustainability and leadership, we compiled a selection of competences for digitalisation, automation, robotics, artificial intelligence, and soft competences such as emotional intelligence and cultural literacy. We also included competences required for sustainability, corporate social responsibility, and circular economy. The selected competences formed the core for the conceptual development of a decision support tool for the individualised selection of training for employees. The concept was tested in customised training to improve employees’ skills and motivation for lifelong learning at the selected industrial partner. The developed assessment algorithm was used to monitor the progress of individual employees’ skills development before and after their training participation. The results of the assessment help human resource departments make decisions for selecting the most effective and optimal training for employees to improve their sustainability-related competences. Such a systematic approach can improve and evaluate competences that companies need to transition to a circular economy.

Neurosymbolic AI (NeSy) aims to integrate the statistical strengths of neural networks with the interpretability and structure of symbolic reasoning. However, current NeSy frameworks like DeepProbLog enforce a fixed flow where symbolic reasoning always follows neural processing. This restricts their ability to model complex dependencies, especially in irregular data structures such as graphs. In this work, we introduce DeepGraphLog, a novel NeSy framework that extends ProbLog with Graph Neural Predicates. DeepGraphLog enables multi-layer neural-symbolic reasoning, allowing neural and symbolic components to be layered in arbitrary order. In contrast to DeepProbLog, which cannot handle symbolic reasoning via neural methods, DeepGraphLog treats symbolic representations as graphs, enabling them to be processed by Graph Neural Networks (GNNs). We showcase the capabilities of DeepGraphLog on tasks in planning, knowledge graph completion with distant supervision, and GNN expressivity. Our results demonstrate that DeepGraphLog effectively captures complex relational dependencies, overcoming key limitations of existing NeSy systems. By broadening the applicability of neurosymbolic AI to graph-structured domains, DeepGraphLog offers a more expressive and flexible framework for neural-symbolic integration.