Nenad Milijić

This study examines whether institutional support, artificial intelligence (AI) literacy, and self-efficacy predict students' innovative behavior in higher business education in the context of growing AI use. Drawing on Social Cognitive Theory, we explain innovative behavior as an outcome shaped by the interaction between environmental conditions and personal capabilities. Empirically, the study relies on survey data collected from undergraduate students in North Macedonia to assess the effects of the three predictors on innovative behavior. The findings show that AI literacy emerges as the strongest predictor, while self-efficacy also has a positive and significant effect. In contrast, institutional support is positively related to innovative behavior but does not make a significant independent contribution. These results suggest that students' innovative actions depend more strongly on their competence and confidence than on support alone. The study contributes to the literature on AI in higher education by extending a social cognitive perspective to students' behavioral patterns related to innovation and highlights the importance of strengthening AI literacy and self-efficacy through higher education practice.

The contemporary automotive industry is undergoing an accelerated transformation towards a circular economy, in which repair and recycling processes have become key elements of sustainable development. The repair-oriented processes discussed in this paper are grounded in circular economy R-strategies (repair, reuse, remanufacture, and recycle), emphasizing the extension of component life cycles and resource efficiency. This paper examines the role of artificial intelligence (AI) in optimizing vehicle repair processes, with a particular focus on enhancing techno-economic efficiency and improving traffic safety. The proposed framework is based on the application of AI algorithms, including computer vision and predictive models, which enable precise diagnosis of damage, estimation of remaining component life, and decision-making on optimal treatment of parts. The empirical part of the research is based on a survey conducted among respondents aged 20-25, representative of the digital-native generation of future mobility users. The results indicate a significant correlation between the level of trust in AI diagnostic systems and respondents' willingness to accept circular economy concepts, including the purchase of repaired components and digitally verified vehicles. The study confirms that artificial intelligence is not only a technical tool, but also a key factor in reducing information asymmetry and increasing market confidence. In this way, AI is positioned as a central mechanism that enables synergy between technological innovation, economic sustainability, and traffic safety within modern transport systems.

Occupational injuries remain a significant issue in European countries, affecting both worker safety and broader socio-economic outcomes. This study groups European countries into homogeneous categories based on workplace injury levels from 2014 to 2023. The analysis uses secondary data from Eurostat, including indicators on fatal and non-fatal workplace accidents within aggregated sectors: agriculture, industry and construction (excluding mining), and business economy services. To achieve the research objective, K-means cluster analysis was applied after data standardisation using Z-score normalisation. The results show substantial variability in occupational injury levels across countries, with particularly high dispersion for non-fatal injuries. The clustering identifies two distinct groups of countries, representing relatively safer and less safe working environments. While most countries show consistent classification across both fatal and non-fatal injury models, a significant number display discrepancies, suggesting that these indicators capture different aspects of workplace safety.

This study examines the dynamics of occupational safety in the Republic of Serbia over a ten-year period (2015-2024). Utilizing official data from the Ministry of Labour and the Statistical Office of the Republic of Serbia, the research employs descriptive statistics, Pearson correlation, and Holt's linear trend model to analyze the relationship between employment growth and injury rates. The results indicate a significant deterioration in workplace safety, with the incidence rate nearly doubling from 36.64 in 2015 to 67.03 per 100,000 employees in 2024. A very strong positive correlation (r = 0.894) was found between workforce expansion and injury frequency, suggesting that economic growth has outpaced the development of safety protocols. Furthermore, the Holt model forecasts a continued upward trajectory for the 2025-2027 period. These findings highlight a critical "implementation gap" between legislative modernization and practical safety outcomes, emphasizing the urgent need for a more robust enforcement of occupational health and safety (OSH) standards.

This paper examines the role of collective intelligence and project management through a bibliometric analysis of 289 scientific publications from 2005 to 2025, using Scopus as the database for source extraction. VOSviewer was used to examine trends in publication and citation volumes, co-authorship networks, and the most influential papers in this field. By mapping the relationships between authors, countries, and key references, the study provides valuable insights into the collaborative dynamics and research gaps. In the last two decades, there has been a shift in publications, with the majority being conference papers, followed by journal articles and book chapters. At the same time, the citation trend has shown a consistent increase.The main implications highlight the necessity for more specialized journals on collective intelligence within project management. Additionally, there is a need to promote collaboration between developing and developed countries during the research.

Objectives. Considering that there is no standardized questionnaire for safety climate and resilience assessment, authors usually review a large number of questionnaires from the available literature, which results in a high number of questions distributed to respondents. As the questionnaire length increases, resistance from the respondents increases. Artificial intelligence (AI) tools until now have not been used for item reduction, besides the need for selecting and retaining only the most relevant and informative questions in the questionnaire with adequate accuracy. Methods. AI tools such as multiple linear regression analysis (MLRA) and the multilayer perceptron artificial neural network (MLP ANN) are used in the development of a model able to cluster respondents' ratings and to predict values of organizational resilience based on the respondents' ratings of the specific questions. Results. AI could be used as a valuable tool for item reduction, since the prediction accuracy for MLRA tools is 70.4-71.5% and for the MLP ANN it is 76.4%. Conclusions. This research proves that machine learning algorithms can be used to build predictive models that determine which survey questions are the most predictive for organizational resilience index calculation using safety climate factors.

The goal of the present study was to develop a prediction model of students? expectations and satisfaction with the quality of services provided at public faculties in the Republic of Serbia. The study was conducted on a sample that included 1212 students from 13 higher education institutions of the University of Belgrade, University of Nis and University of Kragujevac. The designed questionnaire contained 34 statements related to assessing the quality of services provided by faculties, thus forming 5 independent variables. The prediction model was created using the Multilayer Perceptron (MLP). The validity of the model was assessed through the Receiver Operating Characteristic (ROC) prediction accuracy curve, which showed a high sensitivity of the model for both categories of the dependent variable. The high value of the Area Under the Curve (AUC) was 0.93 for both classification categories. This is confirmed by the high values of the AUC, which was 0.94 for the Dissatisfaction category and 0.95 for the Satisfaction category.

The research in this paper aims to develop an original model of the influencing factors of the safety climate in industrial processes, as an important element of organizational development. The set of collected data was analyzed by using statistical tool software packages SPSS 21.0 and LISREL 8.80, based on testing a sample of 843 employees of industrial organizations with a predominantly female labore force, in the Republic of Serbia, on the territory of Jablanica District. It was determined that the factors of the working conditions (WC), as well as attitude of employee towards the working conditions (AE), can be indicators of the impact of employee safety. The developed model is a practical, acceptable and applicable solution that can bring positive changes in practice, regarding occupational safety and health (OSH).

During the project realization, disagreements arise and cannot be avoided. Project complexity, various participants, stakeholders' interests, task requirements, and tight project deadlines create an environment for conflict issues. Every individual engaged in the project responds to conflict differently and has a distinct conflict management strategy. Therefore, this research aims to get deeper insights into conflict management styles' differences in specific demographic characteristics. Empirical research was conducted to obtain this study's objective. A total sample of 114 employees from project-based organizations operating in the Republic of Serbia was collected. The findings of the ANOVA test revealed that participants' responses differed in terms of age, work experience, and project type. Findings provide valuable insights by examining differences in conflict management styles and identifying behavioral patterns and variations based on demographic factors.