Journals Published by Vilnius Tech
Not a member yet
12596 research outputs found
Sort by
Research on the valuation of internet enterprise data assets based on value chain theory
Get PDFAs data assets grow strategically important yet remain difficult to value in internet enterprises, this study analyzes the factors influencing their valuation. Using value chain theory and a system dynamics model, it uncovers the mechanisms of value formation. Results show that data asset value is realized dynamically across stages – collection, analysis, mining, and application – shaped by internal attributes and external factors. The process follows a diminishing marginal return pattern and exhibits significant value lag. Therefore, data asset assessment should account for the full life-cycle, intrinsic properties, and technological conditions
Quotation impact factors for general contractors: a viewpoint from private sectors in Taiwan
Get PDFThe quotes provided by private sector entities to the project’s general contractor vary based on actual considerations. This research aims to identify the factors that influence these quotes for general contractors. Through an extensive review of literature and interviews with experts, four main aspects were identified, encompassing a total of 38 impact factors related to quoting patterns. A survey was conducted among construction professionals, resulting in 123 valid responses out of 146 distributed surveys. The survey’s validity was confirmed with a Cronbach’s Alpha value exceeding 0.7. Subsequently, a factor analysis was performed, confirming 38 impact factors and their respective weights. The results from the Principal Component Analysis (PCA) offer valuable managerial insights, highlighting the most impactful factors: external demand, corporate adjustment, and financial condition. These findings provide practical guidance for practitioners. These insights encompass areas such as information sharing, project limitations, and subcontractor capabilities. Contrary to stereotypes presented in prior research, factors such as reputation and social connections were found to not significantly impact quotation outcomes
Artificial intelligence (AI) for innovative products development
No full textIn today’s economic environment, where technological progress is rapid and product life cycles are getting shorter, companies not only need an innovative product development strategy but also an understanding of the factors that determine the success of new products in the market. One solution is the development of innovative products using artificial intelligence. This article aims to investigate how the integration of artificial intelligence affects the process of innovative product development. The article explores the concept of innovation and artificial intelligence, the link between these processes and classification models. A text analysis is carried out according to the developed research methodology. Based on the literature analysis, a popular model of innovative product development is selected. By integrating artificial intelligence into this model, various aspects of its application are examined, such as process automation, data analysis techniques and decision-making improvement. Links between the chosen model and the integration of artificial intelligence are found. A simulation approach is used to propose a structured model to assess the impact of AI on innovative product development. This approach helps to understand the relationships and interconnections between the results and the elements of the empirical study. At the end of the study, a simulation approach is carried out to test the proposed structured model. The paper concludes with conclusions and suggestions.
Article in Lithuanian.
Dirbtinio intelekto (DI) taikymas inovatyviems produktams kurti
Santrauka
Dabartinėje ekonominėje aplinkoje, kurioje technologinė pažanga vyksta labai sparčiai, o produktų gyvavimo ciklas trumpėja, įmonėms reikia ne tik inovatyvios produktų kūrimo strategijos, bet ir supratimo apie veiksnius, lemiančius naujų produktų sėkmę rinkoje. Vienas iš sprendimo būdų – inovatyvaus produkto kūrimas pasitelkiant dirbtinį intelektą. Šio straipsnio tikslas – ištirti kaip dirbtinio intelekto integravimas veikia inovatyvaus produkto kūrimo procesą. Straipsnyje analizuojama inovacijų ir dirbtinio intelekto samprata, sąsaja tarp šių procesų ir klasifikavimo modeliai. Remiantis sukurta tyrimo metodologija, atliekama mokslinės literatūros analizė, o pagal ją pasirenkamas populiarus inovatyvių produktų kūrimo modelis. Į šį modelį integruojant dirbtinį intelektą, nagrinėjami įvairūs jo pritaikymo aspektai, tokie kaip procesų automatizavimas, duomenų analizės metodai ir sprendimų priėmimo tobulinimas. Identifikuojamos sąsajos tarp pasirinkto modelio elementų ir dirbtinio intelekto integravimo galimybių. Siekdami pasiūlyti struktūruotą modelį, įvertinantį dirbtinio intelekto įtaką inovatyvių produktų kūrimo procesui, autoriai taiko modeliavimo metodą. Jis padeda suprasti empirinio tyrimo rezultatų ir elementų, ryšius bei sąsajas. Remiantis šiuo metodu, sukuriamas modelis. Tyrimo pabaigoje, norint patikrinti pasiūlytą struktūrizuotą modelį, taikomas simuliacijos metodas. Straipsnio pabaigoje pateiktos tyrimo išvados ir rekomendacijos.
Reikšminiai žodžiai: dirbtinis intelektas, dirbtinio intelekto integravimas, inovacijų strategija, inovatyvus produktas, inovatyvus produktų kūrimas, simuliacijos metodas
The relationship between earnings management and audit opinion
Get PDFThis study is conducted to investigate the relationship between earnings management (EM), audit quality (AQ) and qualified opinion of financial statements (AO) of listed firms on Vietnam Stock Exchange. Data were collected from 499 listed firms from 2018 to 2020 with 1,497 observations, of which 101 are observed with qualified opinions. Regressions of Logit and GLS are employed. The results reveal that earnings management has a positive relationship with qualified opinion, while audit quality has a negative association with qualified opinion and earnings management. In this research, we consider the interaction of audit quality with earnings management and qualified opinion. An interesting finding is that the impact of audit quality is weaker than that of earnings management and qualified opinion, so the interaction variable has a positive relationship with both qualified opinion and earnings management. Based on the findings, several suggestions are proposed for users of financial statements and auditors as well to assess the appropriateness of the opinion that is not an unqualified opinion
Overcoming barriers: the struggle for recognition of female successors in Indonesian family business
Get PDFFamily businesses are vital to Indonesia’s economy, making succession a critical process. However, female successors often face a “glass ceiling,” due to ingrained gender biases and traditional family roles, limiting their advancement into leadership roles within their family business. Recognising their unique identities can function as an alternative pathway to leadership, ensuring business continuity by challenging gender bias. This study aims to explore how recognition empowers female successors in Indonesia family business to overcome gender bias and pursue leaderships roles. Employing Interpretative Phenomenological Analysis (IPA), this study explored the lived experiences of nine female successors based on specific criteria. Data was gathered through semi-structured interviews. This research reveals that female successors face gender bias and credibility gap. To gain recognition, female successors actively imitate their fathers’ behaviour and building trust by fostering acceptance through dialogue. Drawing on Honneth’s recognition theory, female successors’ leadership transition is aided by verbal affirmation, paternal mentorship, legal adherence, and role appreciation. Future research should examine a larger sample and more diverse sample of female successors, tracking their experiences over a longer period, to provide a more comprehensive understanding of the challenges and opportunities they encounter within the unique context understand their evolving experiences in family business
Quantitative assessment of human capital in Latvia
Get PDFThe importance of human capital has been a subject of discussion for several centuries. Over time, various methods for its quantitative assessment have been developed, each offering its own insights into the role of human capital in economic growth of a country. The study focuses on the case of Latvia by analyzing different types of quantitative human capital indicators. The study reviews existing methodologies used for a quantitative assessment of human capital, including indicator-based, cost-based and income-based methods, as well as six human capital indicators such as the human capital index by the World Bank, the human capital index by the Institute for Health Metrics and Evaluation, and the global human capital index by the World Economic Forum. The results show that, in five out of six indicators, Latvia ranks between the 21st and the 39th place in the world. However, some indexes offer outdated information or place a big focus only on one specific metric. Based on these findings, the authors of the paper propose a new quantitative approach for assessing human capital that integrates labour market, education and health indicators and metrics. The new approach shows that there is room for human capital potential in Latvia
Lightweight deep models for video anomaly detection: a comparative study of autoencoders and MobileNetV2 on the avenue dataset
Get PDFVideo anomaly detection aims to identify unusual events in surveillance footage, yet many existing deep learning solutions remain too computationally heavy for real-time deployment on resource-limited hardware. This study presents a systematic comparison of three lightweight deep learning models for frame-level anomaly detection on the Avenue dataset, including a baseline 2D convolutional autoencoder, an enhanced reconstruction-based autoencoder with refined feature representation and decoding strategy, and a MobileNetV2-based supervised classifier fine-tuned for anomaly recognition. The baseline autoencoder achieves moderate detection performance, with an approximately AUC of 0.75. In contrast, the enhanced autoencoder improves reconstruction quality and raises the AUC to approximately 0.84 through more effective feature abstraction rather than increased architectural depth. The strongest results are obtained by the MobileNetV2 classifier, which achieves an AUC close to 0.99, high precision and recall, and a stable confusion matrix. These results demonstrate that lightweight architectures, when combined with appropriate training strategies and careful handling of class imbalance, can outperform more complex models. Overall, the study confirms that architectural efficiency and learning paradigm selection are more critical than model depth alone, making lightweight models well-suited to practical, real-time video anomaly detection scenarios.
First published online 02 February 202
A Vieta–Lucas collocation and non-standard finite difference technique for solving space-time fractional-order Fisher equation
Get PDFThe purpose of the article is to analyze an accurate numerical technique to solve a space-time fractional-order Fisher equation in the Caputo sense. For this purpose, the spectral collocation technique is used, which is based on the Vieta–Lucas approximation. By using the properties of Vieta–Lucas polynomials, this technique reduces the nonlinear equations into a system of ordinary differential equations (ODEs). The non-standard finite difference (NSFD) method converts this system of ODEs into algebraic equations which have been solved numerically. Moreover, the error estimate is investigated for the proposed method. To show the accuracy and efficiency of the technique, the obtained numerical results are compared with the analytical results and existing results of the particular forms of the considered fractional order models through error analysis. The important feature of this article is the exhibition of variations of the field variable for various values of spatial and temporal fractional order parameters for different particular cases
Nonstationary heat equation with nonlinear side condition
Get PDFThe initial boundary value problem for the nonstationary heat equation is studied in a bounded domain with the specific overdetermination condition. This condition is nonlinear and can be interpreted as the energy functional. In present paper we construct the class of solutions to this problem
A numerical scheme to simulate the distributed-order time 2D Benjamin Bona Mahony Burgers equation with fractional-order space
Get PDFIn this study, a new class of the Benjamin Bona Mahony Burgers equation is introduced, which considers the distributedorder in the time variable and fractional-order space in the Caputo form in the 2D case. The 2D-modified orthonormal normalized shifted Ultraspherical polynomials are derived from 1Dmodified orthonormal normalized shifted Ultraspherical polynomials and 2D-modified orthonormal normalized shifted Ultraspherical polynomials and the orthonormal normalized shifted Ultraspherical polynomials are applied to approximate of the space and time variables, respectively. Moreover, the convergence analysis of these basis functions is investigated. Due to the time variable being in the distributed-order mode and the space variable being in the fractional-order case, to apply the desired numerical algorithm for this type of equation, operational matrices of ordinary, fractional and distributed-order derivatives are computed. In the proposed method, once the unknown function is approximated using the mentioned polynomial, the matrix form of the residual function is derived and then a system of algebraic equations is adopted by applying the collocation approach. An approximate solution is extracted for the original problem by solving constructed equation system. Several examples are examined to demonstrate the accuracy and capability of the method