Supporting Educational Administration via Emergent Technologies: A Case Study for a Faculty of Engineering in Foreign Languages

Abstract

Although emerging technologies are increasingly adopted in teaching and learning, their potential to enhance educational administration remains underexplored. In particular, few studies examine how conversational agents, virtual reality (VR), and robotic process automation (RPA) can jointly streamline administrative workflows in multilingual and multicultural university environments. This study addresses this gap by presenting an integrated solution deployed on the website of an engineering faculty where programs are delivered in foreign languages. The proposed system combines a multilingual chatbot, a VR-based administrative guide and virtual tour, and RPA modules supporting certificate generation, password resets, and exam scheduling. Through an A/B usability test, usage analytics, and qualitative feedback, we evaluate the effectiveness of these technologies in improving access to information, reducing response time, and lowering administrative workload. Results show that this triad significantly enhances efficiency and student experience, particularly for international students requiring continuous support. The paper contributes a replicable model for leveraging emerging technologies in educational administration and offers insights for institutions seeking scalable and student-centered digital transformation.

1. Introduction

The rapid expansion of digital technologies has significantly reshaped higher education, influencing not only teaching and learning but also the less-explored domain of educational administration. While universities increasingly adopt digital platforms for instructional purposes, administrative processes—such as student onboarding, document issuance, data management, communication, and service provision—often remain dependent on traditional, labor-intensive workflows (Carmo et al., 2025; Hinojosa et al., 2025). This gap generates long response times, repetitive manual tasks, and challenges in supporting large, diverse student populations, particularly in multilingual and international institutions.

Although emerging technologies such as conversational agents, virtual reality (VR), and robotic process automation (RPA) have shown substantial potential in other sectors, their integration into educational administration remains understudied. The existing literature predominantly focuses on pedagogical applications of AI, immersive learning, or digital transformation in teaching (Janahi et al., 2023; Fowlin et al., 2025; Mulders, 2025; Morina & Perera, 2025; Sembey et al., 2024). Far fewer studies examine how these technologies can improve administrative efficiency, support communication, or streamline student interactions. Moreover, research rarely considers multilingual contexts, where administrative complexity is intensified by linguistic diversity and high numbers of international students.

To address these gaps, this study proposes and evaluates an integrated solution based on a triad of emerging technologies—chatbots, VR, and RPA—implemented on the website of an engineering faculty where teaching is conducted in three foreign languages. The purpose is to explore how these technologies can automate repetitive workflows, reduce administrative workload, provide accessible multilingual support, and improve students’ access to institutional information.

This paper makes four main contributions: (1) a novel, unified technological model that integrates a multilingual chatbot, an interactive VR administrative guide and virtual tour, and several RPA modules tailored to educational processes; (2) a real-world implementation deployed in a multicultural engineering faculty, demonstrating practical feasibility and scalability; (3) an empirical evaluation combining A/B usability testing, usage analytics, and qualitative feedback, offering evidence of improved efficiency and user experience; (4) a replicable framework for institutions seeking to modernize administrative workflows through emerging technologies.

The remainder of the paper is organized as follows: Section 2 reviews recent research on emerging technologies in higher education, with a particular focus on their transition from pedagogical applications toward educational administration. Section 3 introduces the institutional context and outlines the administrative challenges specific to a multilingual faculty of engineering. Section 4 presents the research design, materials, and methods adopted in the case study. Section 5 describes the implementation of the integrated technological solution, detailing the VR, chatbot, and RPA components. Section 6 reports the evaluation results for each technology. Section 7 discusses the findings in relation to existing frameworks, practical implications, implementation barriers, and future research directions. Finally, Section 8 concludes the paper and summarizes the main contributions.

2. Emerging Technologies in Higher Education: From Pedagogy to Educational Administration

Recent advances in digital technologies are accelerating transformation across higher education institutions, with artificial intelligence, immersive technologies (VR/XR), and intelligent automation identified as key drivers of organizational change (World Economic Forum, 2025). While these technologies have been extensively explored in teaching and learning contexts, their potential to support educational administration and student services remains comparatively under-researched.

Recent studies highlight a shift from technology-enhanced instruction toward technology-supported institutional services, including student guidance, administrative communication, and process automation (Fowlin et al., 2025; Mena-Guacas et al., 2025). In particular, emerging technologies are increasingly viewed as enablers of accessibility, responsiveness, and scalability in higher education systems facing growing student diversity and administrative complexity.

Conversational agents and large language model-based systems are being adopted to provide continuous, multilingual access to institutional information, reducing response times and administrative workload (Singh & Namin, 2025; Urbani et al., 2024; Dascalu et al., 2024a; Khennouche et al., 2024; Phokoye et al., 2025; TechTarget, 2024). At the same time, immersive technologies such as virtual reality are evolving beyond instructional simulations toward applications supporting orientation, spatial familiarization, and procedural understanding, particularly for international and first-year students (Mulders, 2025; Dascalu et al., 2024b; Uta et al., 2025). These tools offer intuitive and visual representations of administrative processes that are often difficult to communicate through traditional text-based channels.

Robotic process automation (RPA) represents another mature technological strand with growing relevance for educational administration. By automating structured and repetitive tasks—such as document generation, credential management, or scheduling—RPA contributes to operational efficiency, consistency, and error reduction (Jimenez-Ramirez et al., 2019; Plattfaut et al., 2022; Schlegel et al., 2024; Syed et al., 2020; Tarquini, 2018). The recent literature emphasizes that RPA adoption in higher education can free administrative staff from routine tasks, allowing greater focus on student-facing and strategic activities (Fombona et al., 2025).

Despite their potential, the successful integration of emerging technologies in educational administration depends on organizational readiness, staff digital competencies, and sustainable investment strategies (Gheisari et al., 2023). Financial constraints, resistance to change, and system integration challenges remain significant barriers, particularly in public institutions.

In this context, the convergence of conversational agents, virtual reality, and RPA emerges as a promising technological triad for supporting daily student–administration interactions. By combining communication support, experiential guidance, and process automation, this triad enables a transition from fragmented, manual administrative practices toward more scalable, inclusive, and student-centered administrative ecosystems in higher education.

The integration of all these technologies, on the one hand, automates interactions and, on the other hand, reduces the volume of repetitive requests that in most cases would have been solved manually.

3. Necessity of Introducing Emerging Solutions in Administration in a Faculty of Engineering in Foreign Languages

The Faculty of Engineering in Foreign Languages (FILS) (Faculty of Engineering in Foreign Languages, 2025) of the National University of Science and Technology POLITEHNICA Bucharest is an engineering faculty where teaching takes place entirely in three international languages: French, English, and German. This unique feature draws students from both Romania and other countries, creating a multicultural and international atmosphere. The faculty uses its good name around the world to benefit from numerous opportunities related to academic mobility, openness to foreign relationships, and cultural diversity.

As a renowned faculty, attractive to ERASMUS+ and non-EU students, FILS offers a wide range of bachelor’s and master’s programs in many different fields of technology. So, it is important to create a digital and inclusive learning environment that meets the demands of a varied group of students. FILS now has about 2000 students from more than 60 countries. Most of them speak English or French, and about 30% of them are international students. This high percentage shows that the faculty is well-known around the world and that its engineering programs are valuable around the world.

All this brings us many students, which naturally implies that they need constant information and assistance. We are talking about international students, who sometimes arrive late due to various logistical problems (visa delays, delayed acceptance letters to studies and so on), which is why the faculty website, the information it contains, and the administrative staff must represent a central point of support for them. The needs of students are diverse, as they come from different backgrounds and arrive at studies (especially non-European students) at different times, which is why the faculty’s administrative staff is busy throughout the year.

Streamlining and providing support in the interaction between students and the administration becomes essential, because the staff cannot constantly support a high volume of requests (especially at the beginning/end of the year, exam sessions, thesis defense, and other key academic periods). In addition, in the interaction with the large number of foreign students that the faculty accommodates annually, there is also the impediment of the foreign language proficiency of the secretariat, which can often complicate the situation.

Therefore, for a smooth and continuous functioning of the student–institution relationship, in an accessible and multilingual manner, emerging technologies can manage to align perfectly in this context:

• The conversational agent can operate 24/7 and provide answers to a large proportion of frequently asked administrative questions;
• VR can be used for interactive guides, virtual tours of the faculty, and administrative simulations;
• RPA can be used for automating processes or taking over various repetitive requests.

When applied to the profile of a technical faculty offering instruction in foreign languages and characterized by a strong international influence—like FILS—the triad of technologies (chatbot–PA–VR) in the area of educational administration can represent real support, as presented in this case study.

4. Materials and Methods

4.1. Research Design

This study adopts a case study research design, combined with a mixed-methods evaluation approach, to investigate the potential of emerging technologies to support educational administration in a multilingual higher education context. The case study methodology was selected due to its suitability for examining complex socio-technical systems deployed in real institutional environments, allowing an in-depth exploration of both technological implementation and user interaction.

Quantitative data were collected through questionnaires and platform usage analytics, while qualitative insights were obtained from open-ended survey responses and feedback provided by administrative staff. This mixed-methods approach enables a comprehensive evaluation of usability, efficiency, and perceived value of the proposed technological solution.

4.2. Research Context and Participants

The study was conducted at the Faculty of Engineering in Foreign Languages (FILS) within the National University of Science and Technology POLITEHNICA Bucharest, an engineering faculty where teaching is delivered entirely in English, French, and German. The faculty hosts approximately 2000 students from over 60 countries, with a high proportion of international and Erasmus+ students, making it a relevant context for studying multilingual administrative support.

Participants involved in the evaluation included the following:

• Undergraduate and master’s students who interacted with the VR environments and chatbot;
• Administrative staff members who were involved in or affected by the automated workflows supported by RPA.

Participation in all evaluation activities was voluntary, and no personally identifiable data were collected.

Participation in the questionnaire-based evaluation was voluntary. Prior to responding, participants were informed about the purpose of the study, the anonymous nature of data collection, and the use of responses exclusively for research purposes. No personally identifiable information was collected, and informed consent was obtained implicitly through voluntary completion of the questionnaire.

4.3. Technological Components Under Study

The research focused on an integrated triad of emerging technologies deployed on the faculty website:

• A VR-based virtual tour and administrative guide supporting orientation and procedural understanding;
• A multilingual conversational agent (chatbot) providing continuous access to administrative information;
• Robotic process automation (RPA) modules designed to automate repetitive administrative tasks such as password resets, certificate generation, and preliminary exam scheduling.

Each component was evaluated individually and as part of the integrated administrative support ecosystem.

4.4. Mapping of Administrative Processes and Supporting Technologies

To support the design of the case study and ensure methodological transparency, key administrative processes were first identified and mapped to the technological components under investigation. This mapping guided both the implementation and the evaluation strategy, providing a structured link between institutional needs and the proposed technological solution; see

Table 1. Mapping of administrative processes to VR, chatbot, and RPA components used in the case study.

Administrative Processes	Main Challenge	Proposed Technology
New student orientation	Lack of familiarity with the spaces	VR
Virtual campus tour	Difficult access for students from other cities or countries	VR
Presentation of the main administrative procedures in an interactive way	Ignorance of essential steps and procedures	VR
Student–administration communication	Long response time and overload of the secretariat	Chatbot
Student feedback	Slow and manual processing	RPA
Issuance of certificates	Frequent requests, repetitive and manual process	RPA
Exam scheduling	Complex coordination between teachers, students and rooms	RPA
Updating and verifying student data	Repetitive task, risk of errors	RPA

.

4.5. Data Collection Instruments

Several data collection instruments were employed:

• Questionnaires: Structured online questionnaires were used to collect quantitative data related to usability, intuitiveness, perceived usefulness, and user satisfaction with the VR environments and the chatbot. The questionnaires included both closed-ended questions (Likert-scale and multiple-choice) and open-ended questions for qualitative feedback.
• Platform Analytics: Usage statistics from the chatbot platform were analyzed, including the number of interactions, frequency of use, and geographic distribution of users.
• Administrative Staff Feedback: Qualitative feedback regarding the RPA components was collected through informal semi-structured discussions and observational reports from administrative staff involved in pilot testing, focusing on processing time, workload reduction, and perceived reliability.

4.6. Evaluation Procedures

The evaluation procedures were tailored to each technological component:

• VR Evaluation: An A/B usability test was conducted, comparing a traditional PDF-based administrative guide with its VR-based counterpart. Participants were asked to interact with one or both versions and complete a questionnaire assessing attractiveness, ease of navigation, clarity of information, and overall experience. The VR A/B evaluation was designed as an exploratory usability-focused study. The primary objective was to capture users’ perceptions and interaction tendencies rather than to perform hypothesis-driven statistical inference. The sample size of 32 participants was considered appropriate for early-stage usability evaluation, in line with prior human–computer interaction and educational technology studies, where small to medium samples are commonly used to identify usability issues and assess perceived usefulness.
• Chatbot Evaluation: The chatbot was evaluated based on usage analytics over a seven-month period and through a student questionnaire assessing frequency of use, satisfaction with responses, and perceived usefulness in an academic context.
• RPA Evaluation: The RPA modules were evaluated during a pilot implementation phase by analyzing task execution time, error reduction, and staff feedback regarding workflow efficiency and operational impact. The RPA evaluation was conducted as a qualitative pilot study. Baseline quantitative measurements (e.g., task processing time before and after automation) were not systematically collected during this phase, as the primary objective was to assess feasibility, usability, and perceived operational impact.

4.7. Data Analysis Methods

Quantitative data collected from questionnaires and platform analytics were analyzed using descriptive statistics, including frequencies, percentages, and aggregated usage indicators. Qualitative data from open-ended questionnaire responses and staff feedback were analyzed using thematic analysis, allowing the identification of recurring patterns related to usability, efficiency, and perceived benefits or limitations.

This structured methodological framework ensures transparency, coherence, and replicability, providing a solid basis for evaluating the impact of emerging technologies on educational administration.

5. Implementation of the Integrated Solution

We integrated the triad of technologies (chatbot–RPA–VR) into the FILS website (Faculty of Engineering in Foreign Languages, 2025), with the purpose of increasing the attractiveness of the website and making the information more accessible for the age group of our students, thus reducing the support workload for administrative staff and professors.

5.1. VR Integration

We have chosen to integrate a faculty virtual tour and an interactive VR guide into the website of the faculty. The goal of creating a virtual tour for FILS is to provide an accessible way for students, alumni, prospective students, partners, and general users to explore our faculty without the need for a physical visit. Therefore, in addition to the idea of promoting the institution, creating a virtual tour offers a sense of comfort and the chance for users to familiarize themselves with certain locations and to visually answer the frequently asked questions that students—especially international ones—ask. Moreover, in order to relieve and support administrative staff, we adapted the pdf version of the students’ guide into a 3D format, allowing the transmission of useful information (for example how to obtain a room in the student residence) in an intuitive way. Students may find their first interactions with the university environment chal ()lenging, particularly if they are first-year or international. There are numerous steps to follow and a lot of information to absorb, and rules or procedures may vary slightly. Thus, the centralization of information is crucial, and we claim that a guide in VR may ease the integration into the university environment and may also allow the administrative staff to focus on more important tasks instead of repeating the same instructions over and over.

A virtual tour is a simulation of an existing location, usually consisting of a sequence of videos, still images, or 360-degree images. The purpose of virtual tours is to familiarize visitors with certain locations or to give them a perspective on it without requiring them to be physically there and to possibly provide them with information about the site. The accessibility offered by virtual reality to these tours is based on the use of images of the spaces concerned, generally generated. While there are many ways to create virtual tours, there are two main categories: the use of commercial platforms or the use of WebVR/WebXR development frameworks.

We chose to create the FILS virtual tour using A-Frame (A-Frame, 2015)—an open-source framework that allows the creation of interactive VR/3D scenes directly in the browser. HTML and JavaScript were also used to implement various features, such as a menu, moving between scenes, or displaying the current room.

The tour, accessible on the FILS website (https://fils.upb.ro/en/virtual-tour/ (accessed on 10 November 2025)), created by capturing several 360° images with an Insta360 camera in the FILS building and in various rooms of the university (see Figure 1), is integrated into a single HTML file, including JavaScript logic (for transitions and conditional display of hotspots, managed based on the tour scene at a given time) and an interactive navigation menu.

Because A-Frame works with WebVR/WebXR, the virtual tour may be viewed with reasonably priced VR glasses, like the Google Cardboard model. In this instance, users interact with hotspots using the fuse feature, which is a virtual reality interaction technique that is particularly useful when users lack a controller or are unable to use a mouse, as in the case of cheaper glasses. The menu provides a different method for the user to navigate. Especially helpful for the browser version, it is separated into submenus according to the floor of the building, enabling rapid navigation without engaging with quick access sections.

The VR guide contains five scenarios in total (see Figure 2): verifying credentials, signing the study contract, obtaining accommodation, a short questionnaire related to the faculty, and the virtual visit mentioned above, all essential stages for new students.

From a technological point of view, the guide was also developed using A-Frame, and the interactive logic is realized using HTML and JavaScript. For specific functionalities, such as signing contracts, a <canvas> element was integrated that captures the mouse path and simulates a real signature, and interaction with objects in the environment is also achieved through 3D models. There is also an integrated database (MongoDB) for storing data and interactions that students have, and the integration was achieved through a Node.js backend that takes POST requests from the application and saves them in dedicated collections.

In the first scene (Figure 3), the user is given instructions on the wall upon entering the scene, and by interacting with the books on the desk, they can open the guide (in PDF format) and verify whether their institutional address is valid on the computer (in this scenario, an address is deemed valid if it ends with @upb.ro or @stud.fils.upb.ro), verification performed using a regex.

The second scene (Figure 4) simulates the signing of the study contract.

Because the process is usually carried out with the help of the year tutor, the user first checks who the tutor is, and then he must access the contract and sign it using an HTML <canvas> that captures the user’s drawing with the mouse using JavaScript events to track mouse movements and regex for name validation.

In the third scene (Figure 5), the user goes through the process of obtaining accommodation. One can check the fees which must be paid depending on the form of financing of the studies and the dormitory type. The amount is displayed, and once completed and validated via JavaScript, the system withdraws the amount of money from the user’s budget, generates a proof of payment, and, as in the previous scene, simulates the signing of the contract.

The fourth scene (Figure 6) is designed as a quiz with multiple-choice questions about FILS, where seven random questions are chosen from a predefined list, with the goal of testing users on their knowledge of the faculty (the questions being oriented towards identity, history, partnerships, study programs, and so on).

Finally, the fifth scene is actually the virtual tour of the faculty, as previously presented. In each scene and for each case, the steps are explained and validated, and the information is in accordance with the regulations of the faculty and the university, so going through the guide in the form of a simulation can allow users to get an idea of how these administrative processes take place.

By combining all these elements, the administrative journey of a student is reproduced through a simple and intuitive orientation experience, thus reducing the possible requests addressed and allowing discovery in a controlled, repeatable, and easy-to-understand environment for all students, regardless of their level of familiarity with the university system.

5.2. Chatbot Integration

Although the fundamental principles of chatbots have gained popularity, the increasing complexity associated with varying user needs requires careful attention to the tools and platforms used to develop these conversational agents. The majority of chatbots have certain fundamental generic traits in common, and each framework has a very different set of technology, development approaches, and other features.

A framework is a combination of libraries for creating, training, and deploying custom and machine learning models for conversational interfaces. These frameworks allow developers to easily create conversational agents. Therefore, selecting the appropriate one for chatbot development is essential for reaching the intended objectives, be they straightforward scripted dialogues, sophisticated AI-driven dialogues, multilingual support, system integration, or a mix of these applications (Urbani et al., 2024).

A chatbot or conversational system responds to queries in natural language, which means it mimics a real person’s response. This is a time-consuming procedure that uses numerous algorithms. First comes message preprocessing, which involves examining the user-input content. Next comes natural language comprehension, purpose classification, which involves comparing the message to pre-existing phrases, and lastly, the answer (Lee et al., 2024).

Although there are numerous options available today for developing chatbots (Dialogflow, Rasa, Pandorabots and others) (Singh & Namin, 2025), in the context of an engineering faculty offering instruction in foreign languages and characterized by a strong international focus, Fastbots (FastBots, 2025) was chosen—a platform that allows the creation of virtual assistants, capable of querying knowledge bases to obtain answers. These bots can respond to users with automatic versatility, facilitated by the integration of artificial intelligence, which makes the flow and understanding of user intentions much less rigid.

For our conversational agent, managing the interaction from the time the user provides an input until they receive a response is achieved using natural language processing techniques, ML, and artificial intelligence (Kovari, 2025). The first phase, which is preparing the text in a few steps, is made easier by natural language processing (NLP): tokenization splits the user’s input into words (tokens) and changes any uppercase letters to lowercase letters. Next, any diacritical or differentiating signs, as well as words that do not mean anything (such as conjunctions and prepositions), are taken out. In NLP, this is called “stop-word removal.” To classify intentions, logical regression approaches (for simpler streams) or transformer-based models (BERT or GPT) are employed to obtain the intent from the text that has already been processed. After the intent is categorized, we must find a semantic match in the knowledge base. This is just a comparison between what the user wanted and what the knowledge base knows. The system then recognizes the entities so that it can give the user a response (if there is a match) or a fallback message.

Our chatbot also integrates GPT-4o mini as the artificial intelligence language model meant to simplify discussions, but also for the multilingual understanding, adaptation, and communication it offers. The chatbot’s creativity percentage is also permitted to be set by the platform; in this instance, it is set at 0% because it is being included into the website of a public institution, so the information should not include personal notes or additions brought by the creative potential of artificial intelligence.

Several main approaches were used in the training process and in building the knowledge base:

• Text input: Example sentences and user intents were manually added;
• Document upload: Reference documents containing structured academic information were linked;
• Potential question generation: This was guided by a direct mapping between expected questions and possible correct answers (see Figure 7).

Conversation flows were constructed by connecting predetermined user inputs to dynamic responses using the visual editor, so the development time was greatly reduced by that compared with drafting conversation rules by hand.

All of the key features, including chat history, live conversation tracking, training statistics, deployment, and appearance (we added a name— FILSbot—a welcome message, and the colors of the faculty logo; all these elements contribute to the image and maintaining identity), are easily accessible through the dashboard.

A command prompt is also included in the conversational agent’s architecture, which lets you model how it would behave based on the activities and goals it must accomplish. For example, in our case, we explain the purpose, as well as some basic rules to follow (e.g., only respond in certain predefined languages—Romanian, English, or French—not provide information outside its knowledge base, not offer personal opinions, etc.). And there was the question of what happens when the database does not have the necessary information. Therefore, a general answer (or fallback answer) was established to give a kind response and to let them know how to get in touch with the faculty staff. The instructions were as follows: “Please reply in the specified language (Romanian, English, or French)”/“Sorry, I don’t have that information, but I encourage you to contact the FILS team for further assistance”, if users ask for information that is not available or if the relevant details are not covered by the data provided.

Finally, for the deployment phase, the chatbot is directly integrated into the faculty website (Faculty of Engineering in Foreign Languages, 2025), thanks to the platform’s embedded code. It can be accessed from all pages of the website, thus offering students instant, fluid, and practical assistance (see Figure 8).

5.3. RPA Integration

Because RPA facilitates the reduction in processing time and the reduction in human errors (Schlegel et al., 2024), we have identified the following administrative processes that we streamlined, processes that currently require significant effort from staff:

• Generating certificates: these are frequently requested and the process involves retrieving data in platforms, completing the document, validating it, and submitting it—feasible with the help of RPA;
• Automated exam scheduling: a process that receives input data (teacher preferences, room restrictions, unavailability dates), runs a scheduling logic, and provides an initial proposal—also feasible with the help of RPA;
• Resetting passwords for access to university platforms.

Currently, our focus in the RPA is on optimizing request handling regarding password resets or assistance for institutional accounts. Although seemingly simple, these requests are frequent, recurring, and often urgent, especially at the beginning of the academic year or before exam sessions. In the near future, RPA can be extended to validate scholarship applications, generate diplomas, or even synchronize data between internal university platforms; see our three use cases of exploiting RPA to automate administrative processes in Figure 9.

6. Results and Evaluation

This section presents the results obtained from the evaluation procedures described in Section 4. The evaluation strategy examined the extent to which the three technological components—VR, chatbot, and RPA—improved administrative efficiency and student experience. A mixed-methods approach was adopted, combining an A/B usability test, interaction analytics, and qualitative feedback from students and administrative staff. This triangulated design ensured that both user perceptions and system performance metrics were considered.

6.1. VR Evaluation

To evaluate the impact of VR elements introduced (the tour and the guide), an A/B test (also called split or group testing) was applied.

The form that was bilingual (French/English) was designed to collect both demographic data and users’ previous experience with emerging technologies and ultimately their opinion on what they tested.

Of the total number of responses—32—the profile of respondents is as follows: most are between 18 and 24 years old, and over half of them are students (bachelor’s or master’s level). Regarding their frequency of interaction with technology, over 95% responded that they use it daily (see Figure 10), and over 55% of them have interacted with VR in the past (see Figure 11).

There were both closed questions (to establish a rapport regarding previous experience with technologies) and open questions (to obtain more qualitative feedback or suggestions for improvement).

For example, for the question “If you tested both (PDF version and VR version), which one seemed more attractive?”, although both versions received generally positive feedback, the VR version was considered more immersive, easier to use, and more effective in facilitating understanding of the administrative steps.

Furthermore, also for the closed questions, in terms of evaluating the experience, the highest scores were also obtained for the ease of navigation in VR and the degree of intuitiveness, and most users appreciated a high degree of comfort when using the gamified version. A score close to the maximum was obtained for the question “How likely are you to recommend it to other students?”, reflecting the general level of satisfaction of the users, with over 95% of them appreciating the experience as immersive (see Figure 12).

In addition to the quantitative evaluation, qualitative data were collected through open-ended questions aimed at capturing users’ perceptions and experiences when interacting with the system. The responses were analyzed thematically and grouped into key categories reflecting usability, accessibility, interface design, and suggested improvements.

Table 2. Summary of qualitative feedback obtained from open-ended questions.

Feedback Category	Rephrased User Statements (Representative Themes)
Usability and Ease of Use	Participants highlighted the system as intuitive and easy to use, particularly beneficial for first-year students unfamiliar with administrative procedures.
Accessibility and Navigation	Users appreciated the accessibility of the navigation, describing the interaction as straightforward, engaging, and enjoyable.
Interface Design	The interface was perceived as clear and user-friendly, with a simple layout that facilitated rapid understanding and interaction.
User Experience Enhancement Suggestions	Some participants suggested adding ambient sound elements to increase immersion and improve the overall experiential quality of the system.

summarizes representative themes derived from the open-ended feedback, illustrating how users perceived the overall interaction and identifying potential directions for enhancing the user experience.

Given the limited sample size, the results are reported using descriptive statistics, primarily percentages and frequency distributions. No inferential statistical tests were applied, as the evaluation aimed to explore comparative tendencies and user perceptions rather than establish statistical significance.

6.2. Chatbot Evaluation

For the chatbot, since its launch in January until July 2025, 1283 unique interactions (where an interaction is a chat that includes one or more messages with a user) have been recorded (see Figure 13), seeking to explore its knowledge by asking various questions in the conversation flows.

The most frequent questions concerned admission procedures, exam sessions, application requests, and timetables—recurring inquiries that would normally be handled by the secretariat office. To estimate the potential administrative impact, we assumed that each request would have been processed manually by administrative staff and that approximately 5 min (including interaction, information retrieval, and response provision) would have been required per request. Based on this assumption, 1283 interactions correspond to a total of 6415 min, equivalent to 106.91 working hours, or approximately 14 full working days. This estimation illustrates the potential operational impact of automating recurring administrative interactions.

The statistics from the platform also revealed the fact that, predictably, most of the interactions were from Romania. In the last 28 days, in terms of popular countries, 138 interactions were recorded from Romania, and after that, in the top, we have France followed by Pakistan. It is also worth noting that interactions have occurred across the globe, with contributors from Germany, Bangladesh, Italy, Morocco, and so on.

According to the results of the questionnaire we administered to the faculty students, 584 agreed to fill in the form, 511 of them admitted to having used the chatbot at least once, half of them admitted that they used it out of curiosity and to “check if it really gives the correct answers”, and the other half used it out of necessity; 90% of them were satisfied with the given answers, but all of them admitted that a chatbot “could be useful in a faculty context”.

6.3. RPA Evaluation

The evaluation of the RPA components was conducted during a pilot implementation phase within the faculty’s administrative services. Feedback was collected through qualitative methods, including semi-structured discussions and observational analysis involving administrative staff directly interacting with the automated workflows (two IT experts).

The evaluation focused on perceived workload reduction, task execution time, error frequency, and overall usability of the automated processes. Observations were complemented by short reflective notes provided by staff members after completing routine administrative tasks supported by RPA, allowing the identification of operational benefits and limitations, over a 7-month period. Due to the exploratory nature of the case study and the limited number of administrative staff involved, the RPA evaluation prioritized qualitative insights over large-scale quantitative measurement, in line with similar studies on early-stage automation in educational institutions.

The RPA evaluation focused on early-stage implementation results for automating certificate generation, password resets, and preliminary exam scheduling processes. Administrative staff perceived a reduction in manual effort and workflow complexity following the introduction of the RPA components, based on qualitative feedback collected during the pilot phase.

The automated password reset workflow reduced processing time from several minutes to under one minute and helped eliminate delays caused by staff unavailability. In addition to password reset requests, the RPA workflows supported other routine administrative processes, such as certificate generation and standardized document handling, which previously required several manual processing steps and a few minutes of staff time per request, based on conservative internal estimates. Staff members emphasized that the consistency of RPA execution minimized errors and improved workflow predictability.

Although still in the pilot phase, the RPA module demonstrated substantial potential to streamline more complex procedures, including scholarship application validation, diploma generation, or inter-system data synchronization. These results align with broader evidence on the value of RPA in administrative environments, providing a path toward more efficient and resilient institutional processes.

6.4. Summary of Findings

Across all components, the triad of technologies demonstrated complementary benefits:

• VR supported greater comprehension and engagement with administrative procedures.
• The chatbot provided fast, multilingual, and accessible communication, reducing student uncertainty and staff workload.
• RPA improved workflow consistency, reduced processing time, and minimized human effort in repetitive tasks.

Together, these results show that integrating emerging technologies can create a more supportive, efficient, and student-centered administrative ecosystem in higher education institutions.

7. Discussions

The results obtained in this study indicate substantial improvements in usability, accessibility, and perceived efficiency of administrative services following the integration of the proposed VR–chatbot–RPA triad. Beyond descriptive gains, these findings can be meaningfully interpreted in relation to recent studies investigating the role of emerging technologies in higher education administration and student support services.

Consistent with prior research on AI-driven conversational agents, the high levels of user satisfaction and frequent utilization observed in this study confirm the potential of chatbots to enhance institutional communication and reduce response latency, particularly in multilingual environments (Urbani et al., 2024; Singh & Namin, 2025; Opranescu & Ioniță, 2024). Similarly, the positive feedback regarding the VR-based guidance component aligns with recent findings emphasizing the value of immersive environments for orientation, procedural understanding, and anxiety reduction among first-year and international students (Mulders, 2025; Mena-Guacas et al., 2025; Namazi & Raiessi, 2025; Newman et al., 2021). The qualitative and quantitative outcomes suggest that VR can support not only learning but also administrative comprehension, an area that remains underexplored in the existing literature.

From an operational perspective, the RPA evaluation highlights efficiency gains and perceived workload reduction among administrative staff, corroborating previous studies that identify automation as a key enabler of administrative resilience and scalability in higher education institutions (Plattfaut et al., 2022; Schlegel et al., 2024). Unlike studies that examine these technologies in isolation, the present work contributes novel insights by demonstrating the synergistic value of integrating communication, experiential guidance, and automation into a unified administrative ecosystem.

7.1. Framework Support and Novelty of the Proposed Triad-Based Model

While the results demonstrate the practical benefits of the proposed VR–chatbot–RPA integration, it is important to position this model within existing digital transformation and educational technology frameworks. Current frameworks generally support the integration of artificial intelligence, automation, and immersive technologies in higher education; however, they tend to address these components either at a conceptual level or in isolation, rather than as a unified administrative service ecosystem (Gheisari et al., 2023; Fombona et al., 2025; World Economic Forum, 2025).

Prior research has investigated partial integrations of these technologies. Several studies report the use of chatbots to support student services, institutional communication, and administrative information access (Urbani et al., 2024; Singh & Namin, 2025), while others focus on RPA as a means to automate repetitive administrative workflows and improve operational efficiency in organizational and educational contexts (Plattfaut et al., 2022; Schlegel et al., 2024). In parallel, virtual and immersive environments have been predominantly examined in relation to learning, orientation, or experiential activities, with limited attention to their role in administrative guidance (Mulders, 2025; Mena-Guacas et al., 2025).

However, these approaches typically examine single technologies or dyadic combinations, without addressing end-to-end administrative processes that span user interaction, procedural understanding, and back-office execution. In this context, the novelty of the proposed model lies not in the individual technologies themselves, but in their coordinated integration across frontend and backend administrative functions. By combining immersive orientation through VR, continuous multilingual access via conversational agents, and workflow automation through RPA, the proposed triad extends existing frameworks toward a student-centered, service-oriented administrative model tailored to multilingual and multicultural higher education environments.

7.2. Implications for Educational Administration Practice

The findings have important implications for educational administration at both national and international levels. At the institutional level, the proposed triad-based model offers a scalable framework for supporting increasingly diverse student populations, particularly in multicultural and multilingual universities. By providing continuous access to information, immersive orientation tools, and automated administrative workflows, institutions can improve service consistency while reducing dependency on manual processes.

At a broader level, the model is relevant for higher education systems undergoing digital transformation, especially in contexts characterized by high international mobility, limited administrative staffing, or fragmented service delivery structures. The approach supports inclusive administrative practices by lowering linguistic and procedural barriers, thereby contributing to more equitable access to institutional services.

7.3. Barriers and Implementation Challenges

Despite its potential, the implementation of the proposed model is not without challenges. Financial constraints remain a significant barrier, as the deployment of immersive environments and automation technologies requires initial investment in infrastructure, software, and staff training. Organizational resistance to change, limited digital competencies among administrative personnel, and integration difficulties with legacy information systems may further hinder adoption, particularly in public institutions.

Additionally, cultural and regulatory differences across national contexts may influence the scalability and transferability of the model. Institutions operating in highly centralized or regulation-intensive educational systems may face additional constraints in automating administrative procedures or deploying AI-driven services.

7.4. Limitations and Future Research Directions

While the study provides valuable insights, it is limited by its single-case design and exploratory evaluation approach. Future research should consider multi-institutional studies, longitudinal analyses, and comparative evaluations across different cultural and regulatory contexts. Further investigation into the long-term impact of such integrated systems on administrative efficiency, student satisfaction, and institutional performance would also strengthen the evidence base. Future evaluations will incorporate baseline quantitative indicators, such as task processing time and error rates measured before and after automation, enabling a more rigorous assessment of the operational impact of RPA in educational administration.

8. Conclusions

This study explored how emerging technologies—conversational agents, virtual reality (VR), and robotic process automation (RPA)—can be integrated into the administrative ecosystem of a multilingual engineering faculty. Addressing a notable gap in the literature, which focuses predominantly on pedagogical applications of digital technologies, the paper demonstrates the value of applying these tools to educational administration, a domain where innovation remains comparatively limited.

The proposed triad-based model was implemented in a real institutional context and evaluated through a combination of A/B usability testing, interaction analytics, and qualitative feedback. The results show that VR enhances students’ comprehension of administrative procedures, particularly for those unfamiliar with local academic practices. The multilingual chatbot improves accessibility and responsiveness, offering continuous support and reducing the volume of repetitive inquiries received by staff. Early-stage RPA automation demonstrates clear potential to streamline repetitive workflows, minimize errors, and reduce administrative burden during peak periods.

Collectively, these findings highlight the benefits of integrating emerging technologies to build student-centered administrative services that improve efficiency, clarity, and inclusiveness. The work contributes a practical, replicable model that other higher education institutions can adapt according to their linguistic, organizational, or infrastructural needs.

Nevertheless, several limitations must be acknowledged. The evaluation was carried out in a single faculty, and long-term usage patterns have not yet been assessed. Some technologies—particularly RPA—are still in pilot stages and require further scaling and refinement. Additionally, continuous updates to the chatbot knowledge base and VR scenarios are necessary to ensure sustained relevance.

Future work will focus on expanding the VR environments, enhancing chatbot conversational abilities through large language models, extending RPA workflows to additional administrative processes, and integrating predictive analytics or student behavior modeling. Longitudinal studies across multiple faculties and universities would further validate the generalizability and long-term impact of the proposed framework. By continuing to develop and evaluate such tools, higher education institutions can accelerate digital transformation and offer more responsive, equitable, and efficient administrative experiences.