Serbian Journal of Management

Human resources as a strategic enabler of circular economy adoption: a bibliometric and thematic analysis

2026-06-09T12:27:49+02:00

This paper presents a comprehensive bibliometric and thematic analysis of 313 peer-reviewed articles indexed in Scopus, mapping the intersection between circular economy (CE) and human resources (HR). The study confirms the central role of HR in enabling the adoption and diffusion of CE practices across industries and regions. Our findings highlight that the success of CE initiatives depends not only on technological advancements and structural changes but critically on the preparedness, engagement, and well-being of the workforce. Key HR practices-including skill development, leadership cultivation, digital and virtual training, and inclusive talent management-are shown to be vital for equipping employees with the capabilities needed to implement circular strategies. The analysis reveals the growing importance of integrating digital technologies, such as Industry 4.0 tools, with human capital development, while also emphasizing the need for supportive working conditions, including safe, fair, and inclusive workplaces, to stimulate sustainable behaviors. The study identifies 15 thematic clusters and highlights emerging topics such as digital circular economy, circular entrepreneurship, and decent work. This paper contributes to understanding the strategic role of HRM in circular transitions and suggests directions for future research on contextual factors, digital readiness, and HR’s role in advancing circular business models globally.

Factors Influencing ERP system usage in manufacturing and service industries: a comparative study using PLS-SEM and fsQCA

2026-06-09T12:27:49+02:00

Enterprise Resource Planning (ERP) systems represent an important technological infrastructure for enhancing operational efficiency and strategic decision making in various industries. This study examines the key determinants influencing the adoption of ERP systems, with a particular focus on organizations operating in the manufacturing and service sectors in the Western Balkans region. In order to empirically evaluate the proposed conceptual framework, Partial Least Squares Structural Equation Modeling (PLS-SEM) and Fuzzy Set Qualitative Comparative Analysis (fsQCA) were employed. In this way, both linear relationships and configuration pathways underlying ERP implementation behavior were captured. The results of the PLS-SEM analysis show that constructs such as work compatibility, perceived usefulness and external factors are statistically significant predictors of users’ attitudes towards using ERP systems. In addition, a significant relationship was found between attitude and actual use of ERP systems, with user experience serving as a moderating variable. Notably, the multi-group analysis revealed no significant differences in intentions, attitudes or preferences between users from the manufacturing and service sectors. The results of the fsQCA study also show that work compatibility and perceived usefulness are the core present conditions necessary for effective ERP adoption. This study contributes to the existing literature by integrating multiple analytical approaches to explore the complexity of ERP usage. Additionally, it provides practical insights for enterprise managers, ERP vendors and policymakers.

Saturated linear unit as an universal symmetric activation function for deep learning

2026-06-09T13:10:43+02:00

There is a number of symmetric activation functions used in artificial neural networks for deep learning. In this paper, we propose a universal activation function based on the Saturated Linear Unit (SaLU) that can be used instead of any known symmetric activation function. It is not necessary for classification tasks to have an exact calculation of the probability of detected classes. The classification decision is made based on the highest probability for the input values. We propose, as a proof of concept, that the two most commonly used hyperbolic tangent and algebraic sigmoid activation functions can be effectively replaced by SaLU by choosing a single parameter. Moreover, the theoretical step function can also be replaced by SaLU for a wider transition range. All derivations use symbolic processing. Also shown is a visualization of the range of inputs that result in a suitable classification. This can help scientists and programmers design complex machine learning algorithms and understand how deep learning algorithms work.

Advancing LEAN methodology adoption in Serbian manufacturing SMEs: empirical insights and a strategic implementation framework

2026-06-09T13:46:43+02:00

As manufacturing industries become more competitive around the world, Lean methodology must be used to improve operations and keep the market stable. However, small and medium-sized businesses (SMEs), have a hard time fully implementing Lean because they don't have enough resources, and their organizations are stuck in their ways. This research provides an empirical analysis of the level of application of Lean in manufacturing SMEs in the Republic of Serbia. A structured survey was conducted in which 54 manufacturing SMEs took part, in order to determine the level of Lean adoption, the main obstacles in the implementation process and the influence of factors, such as company size, quality management system and product diversification. The Mann-Whitney U, the Kruskal-Wallis, the Chi-square test and a binary logistic regression were used for qualitative analysis. The results show that there is a strong link between ISO 9001 certification and Lean implementation. Larger enterprises are more likely to adopt Lean than smaller ones. The study presents a customized Plan-Do-Check-Act (PDCA) model for the Serbian SMEs, offering an approach to overcoming implementation barriers. Future research should be directed towards extending the analysis to the Western Balkans and comparing the results to support evidence-based industrial policy development.

The national cultural dimensions influence on celebrity endorsement

2026-06-09T12:27:50+02:00

Celebrity endorsement is one of the most effective promotional strategies and, as such, is widely present in both domestic and international marketing. The application of this promotional strategy in global marketing requires adaptation depending on the cultural patterns of consumer behavior in the international environment. Accordingly, the subject of this study is the influence of culture on the effect of celebrity endorsement within the global context. The study aims to examine the impact of national cultural dimensions, based on Hofstede's model (power distance and individualism/collectivism) on the perceptions of respondents from different cultural backgrounds regarding the appearance of celebrities in promotional activities, the perceived quality of products endorsed by celebrities, the attractiveness and credibility of celebrities. Primary data was collected using the method of personal interview, applying an online survey technique. A total of 353 respondents from 70 different nationalities participated in the study. The research hypotheses were tested using group comparison techniques, specifically the independent samples t-test. The results suggest that celebrity endorsement is more favorable in collectivistic and high power distance cultures in comparison to individualistic and low power distance cultures. The theoretical contribution of the study lies in providing new insights into the influence of culture on consumer behavior. The findings may serve as guidelines for marketing managers in developing a marketing strategy plan for international markets, which can be considered a practical contribution of the study

Real estate investment - a financial and bibliometric analysis

2026-06-09T12:27:50+02:00

The investigation of real estate risks, exuberance and housing bubbles have garnered a lot of attention, leading to significant implications on governments, industry and human beings. Their compliance with the competitive market economy has led to the identification of certain distinctive features and risks, such as the atomization of requests and offers, the piecing fluidity, the personalization of requests, and sales-focused production. Thus, much research has been conducted on the topic of real estate investment over the last two decades. However, discussing the impact of real estate investing in the context of radical cultural changes driven by overlapping crises, technological innovation pressures, and environmental concerns continues to lack systematic and in-depth bibliometric research. This paper aims to consolidate standard research on the bibliometric study of papers and articles published over the last 23 years, from 2000 to 2023. This paper aims to advance our understanding of this cross-cutting concern using a review of the bibliometric literature on the house price bubble thus providing both a quantitative and qualitative perspective on productivity related to this area of economic research. The results from 1,780 studies analyzed show evolving trends in the keywords financial investments, real estate investment, commerce, risk assessment and artificial intelligence. traditional commerce-related keywords and have increasingly begun to include elements of risk assessment and artificial intelligence, reflecting a continuous adaptation to new technological and economic challenges and opportunities. Future research may cover other databases for more comprehensive analysis

The role of leadership in driving corporate social responsibility and innovation: insights from Serbian organizations

2026-06-09T12:27:51+02:00

The objective of this study is to examine the impact of transformational and transactional leadership styles on Corporate Social Responsibility (CSR) and Organizational Innovation (OI) within Serbian organizations. Using responses from 204 employees across small and medium, domestic and foreign companies, the research explores how leadership behaviors influence CSR initiatives and innovative practices. Data were collected through an online survey using a structured questionnaire and analyzed in SPSS with descriptive statistics, reliability testing, and multiple regression to examine relationships. Stratified sampling was applied to ensure representation across company size and origin. Two hypotheses were tested and proven highlighting significant differences of this influence based on organizational size and origin, with transformational leadership proving more effective than transactional leadership as a predictor of CSR and OI outcomes. The research findings point to the importance of culturally aligned leadership approaches in fostering sustainable and innovative practices, offering valuable insights for leadership development in transitional economies.

Higher education in light of Industry 5.0: SERVQUAL-SEM quality dimensions

2026-06-09T12:27:51+02:00

The Industrial Revolution wheels the development of society, and its transformation reflects the upcoming Industrial Revolution. This led to Industry 5.0 and Society 5.0. in many areas, even in education. Therefore, universities should achieve a high level of service quality to align with new industrial trends where humans are expected to be innovative and collaborative in a shared working environment with new digital technologies. Thus, this paper aims to research the relationship among SERVQUAL quality variables of online higher education (Tangibility, Reliability, Readiness, Assurance and Empathy) and their impact on student perception of Industry 5.0 effects on higher education quality. To achieve this aim, the survey was carried out among 532 university students. Structural Equation Modeling (SEM) shows that all analysed relationships among SERVQUAL variables are confirmed and indicate a positive impact. Furthermore, the influence of the mentioned quality variables on the students’ perception of Industry 5.0 effects on higher education quality is also considered. Considering those relationships, Reliability is the only significant variable impacting students’ perception of Industry 5.0 effects on higher education quality. These conclusions could be used to develop a framework for quality improvement of online higher education in Industry 5.0, leading to the base for developing new Society 5.0

An analysis of the influence of shadow economy on economic development

2026-06-09T12:27:51+02:00

The shadow economy is an umbrella concept that covers different aspects of the informal economic sectors. Conventional wisdom states that an increased size of the shadow economy hampers economic growth. However, we adopt a more nuanced conceptual framework, considering the mixed effects exercised by the shadow economy on economic development and full employment of labour resources. Thus, we estimate the size of the shadow economy employing a Multiple Indicators and Multiple Causes model. To cope with the uncertainty that emerges from the large set of potential determinants of economic development, we employ the resulting proxy of the shadow economy in a Bayesian Model Averaging/Weighted-Average Least Squares. Alternative approaches to data are considered for a robustness check. We find that for a dataset of 124 post-industrial and developing economies covering the period between 1991 and 2017, the shadow economy can both positively and negatively affect economic development. Our results show that the ‘net’ effect exercised by the shadow economy on the official sectors is an open question. The main implication is that any policy aiming to reduce the size of the shadow economy should be accompanied by an improvement of the economy's formal sectors to absorb unemployment.

Innovation versus acquisition: panel evidence on the market value of intangible capital in Europe

2026-06-09T12:27:51+02:00

This study examines the influence of intangible investments on firm valuation in terms of the comparative influence of internally generated research and development and externally purchased intangible fixed assets. Using panel data for 247 listed European firms over the span of a decade, this paper uses fixed effects and Generalized Method of Moments estimators to adjust for unobserved heterogeneity and endogeneity problems. The findings reveal a strong and stable positive and statistically significant association between research and development intensity and Tobin's Q, stressing the focus of the market on endogenously driven innovation as the strongest driver of firm value. In comparison, the effect of acquired intangibles is less strong and statistically less intense, revealing that capital markets focus more on endogenous innovation efforts than on intangible acquisition. Results of the study contribute to the literature on intangible capital by offering empirical findings on the impact of investment styles on firm value, with significant implications for policy making, innovation management, and corporate finance.

Incorporating literary fiction in teaching "Brand Management" course - focus on consumer values

2026-06-09T12:27:51+02:00

The authors demonstrate how examples from literary texts deepen the understanding of consumer values in Brand Management course (according to Sheth, Newman and Gross’ classification: functional, social, emotional, epistemological, and conditional values). To understand the students’ interest in this topic and the usefulness of assignments with fragments of fiction, the authors conducted a study among MBA students who completed the Brand Management course, where these assignments were integrated. The results of the study showed that students consider these assignments interesting, useful, and broadening their erudition. Moreover, almost half of the respondents used the knowledge gained from completing the assignments with fragments of fiction in their practical activities.

Blockchain-based skill and employment verification: challenges and strategies for executive hiring in Dubai’s fintech industry

2026-06-09T12:27:52+02:00

Dubai's traditional executive recruitment process faces challenges, including the use of fake credentials, screening delays, and high recruitment costs. With its decentralized and immutable nature, blockchain offers a potential solution to these problems by ensuring secure, accurate and efficient verification. The paper examines how HR professionals and IT specialists from various Dubai FinTech companies perceive the effectiveness of blockchain in verifying credentials, reducing fraudulent claims, and enhancing recruitment efficiency. Additionally, the paper examines organizations' readiness to adopt blockchain technology and its potential impact on operational costs and efficiency. The research findings show that blockchain significantly improves verification accuracy by reducing human error and preventing fraud. Blockchain has been identified as an effective tool to curb fraudulent claims in executive hiring, positively impacting both time and cost efficiency. However, resistance to new methods and concerns about implementation costs were identified as potential barriers to widespread adoption. To overcome resistance to change and reduce implementation costs, adequate training is essential for successful adoption. The study proposes further research into the long-term adoption of blockchain and its impact on different industries.