TOLUCATRAIN VR 1.0: Implementation and Evaluation of an Open Educational Resource for Railway Training

Introduction

In today’s educational landscape, digital transformations have significantly reshaped traditional instructional models, paving the way for more inclusive, equitable, and open approaches. According to UNESCO (2024b), open education plays a strategic role in enabling digital transformation processes that prioritize equity, inclusion, educational quality, and long-term sustainability.Within this framework, technologies such as Virtual Reality (VR) have emerged as democratizing tools for technical knowledge, by enabling active, accessible, and personalized learning environments. Ariza-Colpas et al. (2020) identify virtual reality as a strategic medium for advancing the democratization of knowledge, especially in processes that promote the social appropriation of science and technology.

In alignment with this vision, Mendoza-García et al. (2024) emphasize the value of immersive educational technologies as enablers of open, collaborative, and situated learning experiences, especially when co-developed with public institutions. Similarly, Iniesto et al. (2021) argue that accessibility and universal design must be core elements of Open Educational Resources (OER) to ensure that inclusivity is embedded from the design stage.

Nevertheless, gaps in specialized technical training persist. Railway engineering, a field considered strategic for sustainable development, faces limited academic coverage and restricted access to hands-on environments. According to the International Transport Forum (2021), rail transport is not only vital for decarbonizing economies but also for promoting inclusive growth and regional connectivity, making railway engineering a strategic discipline for achieving the United Nations Sustainable Development Goals (ITF 2021). Veraza, Vicario and González (2025) argue that railway engineering education faces significant challenges, particularly in providing meaningful practical experience without the high costs associated with physical infrastructure. This highlights the need to integrate emerging technologies with OER to bridge training gaps and foster a culture of continuous technical learning.

In Mexico, limited access to real-world learning contexts and insufficient infrastructure negatively impact technical education. The Organization for Economic Co-operation and Development (OECD 2020) warns that “the lack of adequate resources and the limited capacity of technical institutions hinder the implementation of effective training practices in real-world contexts”. This issue becomes even more critical in sectors like railway engineering, where operational costs are high. Asaff et al. (2015) reported that railway education in Brazil had significantly declined, becoming largely restricted to technical training conducted within private company maintenance workshops.

Traditional methods, which are predominantly theoretical, often fail to capture the operational complexity and systemic interactions inherent in railway systems. Vicario Solórzano et al. (2024) argue that conventional teaching approaches do not adequately represent the dynamic functioning of railway operations. They further contend that the scarcity of Open Educational Resources in this field constrains pedagogical innovation and limits the democratization and continuous updating of technical knowledge.

In response to these limitations, immersive environments provide safe, accessible, and controlled practical experiences. Veraza, Vicario and González (2025) describe the development of a virtual reality-based techno-educational resource designed to optimize learning through immersive and interactive environments. This resource supports the training of engineers for initiatives such as the Tren Maya, the Tren Transístmico, and the Tren Interurbano México–Toluca (TIMT), a key project aimed at modernizing the country’s rail infrastructure (García 2019; Yñigo & Flores 2024). The Instituto Politécnico Nacional (IPN) launched specialized training in this sector in 2020 through its Unidad Profesional Interdisciplinaria de Ingeniería y Ciencias Sociales y Administrativas (UPIICSA) campus in Mexico City and Unidad Profesional Interdisciplinaria de Ingeniería Campus Palenque (UPIIP) in Chiapas.

Recent literature confirms the effectiveness of immersive environments. Ka et al. (2025) found that VR-based teaching methods increased students’ interest, improved comprehension, and enhanced immersion-driven learning. Similarly, Wong, Hui and Kong (2023) reported higher levels of perceived usefulness and engagement in virtual reality environments designed for industrial operations training

Beyond their pedagogical potential, the development of such resources addresses specific training needs and aligns with the core principles of Open Educational Resources. The UNESCO Recommendation on OER (UNESCO 2019) establishes reusability, adaptability, accessibility, and international cooperation as foundational pillars of open education. In line with this framework, resources must be designed to be reusable, adaptable, and scalable (UNESCO 2024b).

This article documents the development of TOLUCATRAIN VR 1.0, an open educational resource created by a multidisciplinary team to enhance technical railway training through a digital twin of the TIMT. Developed using Unreal Engine and real-world scans, the resource merges theoretical content with practical applications in a highly immersive and realistic environment. Designed according to principles of accessibility and reusability, it offers a meaningful response to the current challenges of technical education in Mexico.

This study also articulates a critical Latin American perspective on the creation of open, situated, and sustainable immersive environments. Unlike foreign models, TOLUCATRAIN VR emerges from institutional co-design, grounded in public capacities, interdisciplinary practices, and a logic of digital public goods. In contrast to commercial simulators such as OpenTrack (Switzerland/Germany) and those developed for operational efficiency training in Asia and Europe, which are primarily designed for professional or military-grade purposes and often rely on proprietary architectures, TOLUCATRAIN prioritizes public co-creation, openness, and alignment with educational purposes in Global South contexts (Muñoz-Muñoz, García-Beltrán & Valero-González 2024; Fernández-Ruiz, Roca & Moreno 2023).

The evaluation of the OER was designed around three complementary dimensions: usability, assessed via the System Usability Scale (SUS); cognitive gain, measured through pre/post comparisons of technical knowledge; and subjective appropriation, evidenced by qualitative testimonies. This mixed approach enables an assessment not only of the technical functionality of the immersive environment but also of its impact on learning and user perception. As this is a proof of concept (Technology Readiness Level 4), the study does not aim to generalize the findings, but rather to explore the technical and pedagogical feasibility of the OER. The validation conducted at two IPN campuses—UPIICSA in Mexico City and UPIIP in Palenque, Chiapas—each with distinct socio-technical profiles, confirms its scalability potential and adaptability to diverse educational and geographical contexts.

In comparison with various international railway simulation initiatives that rely on closed architectures—often developed by private companies or for military contexts—TOLUCATRAIN VR stands out as a distinct public co-creation initiative with an open, local character. The use of tools that align with this open approach, such as Unreal Engine, and the deliberate design of reusable and scalable content, position this resource as sovereign educational material aimed at technical training. Such an open-source strategy strengthens national capacities for educational technology development, while also enabling other actors in the ecosystem—public institutions, students, and educators—to become co-designers and active users.

Methods

The methodological design of this research was grounded in an applied research logic, supported by a techno-educational approach that integrates the design of immersive experiences with pedagogical, technical, and ethical criteria. From this perspective, an explanatory sequential mixed-methods design was adopted, allowing for a comprehensive understanding of the resource’s performance by considering both quantitative data (usability and cognitive learning outcomes) and qualitative data (student and faculty perceptions). The methodological strategy was structured to address two key dimensions: (1) the architecture and development of the virtual environment as an open digital good, and (2) the validation of its educational impact through proof-of-concept trials in real-world contexts. The following sections detail the development stages, participants, data collection instruments, and analytical procedures employed.

The evaluation relied on two main instruments. First, the SUS was applied to assess the perceived usability of the immersive learning environment. The SUS is a standardized questionnaire consisting of 10 Likert-scale items that evaluate the effectiveness, efficiency, and satisfaction of users interacting with a system. It provides a global score reflecting the overall user experience, which is particularly relevant for immersive interfaces where interaction design directly impacts engagement and learning.

Second, a pre/post knowledge test was developed to measure learning gains related to railway operations. This test included 20 multiple-choice questions covering key operational concepts, safety protocols, and system components addressed in the digital twin experience. The test was designed in alignment with the instructional content embedded in the virtual reality scenarios, ensuring content validity. Both instruments were validated by subject matter experts before implementation.

The SUS allows for the evaluation of various aspects of the user experience, such as ease of learning, system consistency, user confidence, and overall efficiency. These components are especially relevant in immersive environments, where the quality of interaction directly influences motivation and learning.

The knowledge test, in turn, was designed to assess key learning outcomes associated with railway operations, particularly the understanding of operational safety protocols, reading and interpretation of control panels, and knowledge of basic maneuvering and circulation procedures. These areas were selected due to their centrality in the content of the TIMT digital twin.

Development of the TOLUCATRAIN VR 1.0 resource

The TOLUCATRAIN VR 1.0 system was developed by the Instituto Politécnico Nacional (IPN) team using Unreal Engine 5.1.1, integrating three-dimensional scanning techniques (photogrammetry and LiDAR) of the Tren Interurbano México-Toluca cockpit. A Faro Focus3D scanner with a 6mm resolution and a high-definition camera (24 MP) were used to capture textures and geometries with high fidelity. The 3D models were optimized with levels of detail (LOD) and reduced polygon counts to ensure smooth performance on Meta Quest 3 and Meta Quest Pro devices, as shown in Figure 1.

Figure 1

The technical architecture adhered to the principles of OER, ensuring interoperability through the use of standard formats (FBX, OBJ) and accessibility in alignment with UNESCO guidelines (UNESCO 2024b). Supplementary materials (infographics, manuals, and interactive assessments) were integrated directly into the VR experience, providing immediate feedback and performance tracking.

Tools and digital modeling of the environment

The virtual environment was created using Unreal Engine 5.3 and digital content was modeled using Blender and Twinmotion. The design was informed by real-world visual references and structural elements captured from the TIMT project, using photogrammetry techniques and schematics provided by institutional collaborators.

As part of the technical development of the resource, advanced optimization processes for virtual reality environments were integrated. These included the use of Levels of Detail (LOD), culling techniques for real-time removal of non-essential data, and performance adjustments for mid-range platforms. These elements are not a separate phase in the evaluative methodological design but are integrated transversally throughout the digital creation cycle to ensure the resource’s stability, accessibility, and usability in low-cost computational environments. This technical strategy directly contributes to the user experience later evaluated through the SUS scale.

The creation of the virtual environment for the Tren Interurbano México-Toluca in the TOLUCATRAIN VR 1.0 project incorporated advanced techniques in data capture, processing, and digital modeling, enabling a faithful reproduction of both the railway infrastructure and its surrounding environment. This multidisciplinary approach combined precise methodologies and cutting-edge technological tools to transform physical data into an interactive digital model optimized for virtual reality. The technical phases of the process are described below:

1. Data capture and registration

The initial phase focused on obtaining precise information through various complementary techniques:

• Drones and Photogrammetry: Drones equipped with high-resolution cameras were used to capture aerial images, which were then processed to generate 3D models via photogrammetry.

• 360° Camera: For interior spaces, a 360° camera was employed to document areas with immersive fidelity.

• Physical Measurements and Scaling in AutoCAD: Accurate physical measurements were taken and integrated into AutoCAD to ensure precise scaling.

• LiDAR Technology: Applied in specific areas to capture volumetric details, thereby enhancing overall accuracy.

2. Processing and preparation of digital models

Once the data was collected, the information was converted into 3D models optimized for integration into virtual environments:

• Integration and Structuring in AutoCAD and SketchUp: Measurement data were first imported into AutoCAD to create a precise reference base. Subsequently, the models were transferred to SketchUp, an essential tool at this stage due to its ability to structure and segment models into layers. Using SketchUp, the model was divided into modular components, allowing for detailed organization and editing of individual elements such as the train cab, compartments, and other architectural details, thereby facilitating visualization and further optimization of the digital environment.

• Polygon Reduction and Optimization: Polygon reduction techniques and the generation of various levels of detail (LOD) were applied. This step was crucial to reduce model complexity without compromising visual quality, ensuring smooth performance on virtual reality devices such as Meta Quest 3 and Meta Quest Pro.

• Refinement and Texturing in Blender: The next step involved exporting the model from SketchUp to Blender, a key tool for final refinement. In Blender, high-resolution textures were applied, lighting adjustments were made, and geometric details were corrected. This allowed the integration of photorealistic materials and dynamic shading effects, as shown in Figure 2. Blender also facilitated the incorporation of real-time light and shadow simulations, contributing to the realism and immersion of the virtual environment.

Figure 2

3. Integration into the virtual reality platform

Once the digital model had been optimized and refined, it was integrated into Unreal Engine 5 (UE5) to create the interactive virtual environment that constitutes TOLUCATRAIN VR 1.0:

• Import and Configuration in UE5: Final models were exported in standard formats (such as FBX and OBJ) and imported into UE5. Within this game engine, advanced materials and realistic lighting systems were configured, including ray tracing techniques to emulate lighting conditions comparable to those observed in the physical environment.

• Optimization for VR Devices: Specific optimization strategies were implemented, such as dynamic LOD management and frame rate adjustments, to ensure smooth functionality on devices with limited resources.

• Integration of Interactive and Educational Elements: The integration of interactive resources, infographics, manuals, and interactive assessments was carried out in accordance with the guidelines of OER and UNESCO directives (UNESCO 2024b). This added an educational dimension to the virtual environment, providing immediate feedback and performance tracking for the users.

4. Technical challenges and implemented solutions

The development of TOLUCATRAIN VR 1.0 involved overcoming various technical challenges, which were addressed with innovative solutions:

• Interoperability and Use of (Application Programming Interfaces) APIs: The integration of multiple APIs in UE5 to enhance interactive functionalities led to occasional instabilities. Debugging protocols and incremental updates were implemented to ensure compatibility across various tools and devices, maintaining a coherent and stable workflow.

• Massive Data Management: The large volumes of data obtained from LiDAR and photogrammetry required the use of advanced algorithms for mesh cleaning and optimization, which helped reduce noise and correct distortions in the digital model.

• Operation in Confined Spaces: The use of drones in restricted environments posed operational challenges due to interference and limited space. This limitation was mitigated by using 360° cameras and optimizing flight routes, ensuring redundant coverage and integrity of the captured information.

The coherent integration of these advanced data capture, processing, and modeling techniques was critical for the creation of the virtual environment in TOLUCATRAIN VR 1.0. This rigorous process not only faithfully replicates the infrastructure of the Tren Interurbano México-Toluca and its surrounding context, but also sets a new standard in the application of virtual reality technologies for educational purposes and infrastructure planning.

Virtual reality performance optimization

In the development of TOLUCATRAIN VR 1.0, specific adjustments were made in Unreal Engine 5 to ensure that the virtual reality experience operated smoothly without sacrificing visual fidelity. These optimizations focused on the configuration and utilization of the engine’s advanced features, such as:

• Nanite: Enabled the management of detailed geometry without extreme manual polygon reduction.

• Lumen: Provided dynamic global illumination in real time.

• Materials and Shadows: Optimized using virtual shadow maps, adjusting their detail according to visual relevance.

• Real-time Profiling: UE5 tools identified bottlenecks and allowed for real-time graphic resource optimization.

Dynamic management of Levels of Detail (LOD) and culling

To achieve efficient performance in TOLUCATRAIN VR 1.0, visual optimization techniques were applied to maintain a fluid experience without compromising graphical quality. A key strategy was the use of Levels of Detail (LOD), which allowed objects to be displayed with more or less visual complexity depending on their proximity to the user. This avoided unnecessary use of graphic resources on elements that did not require high visual precision.

In addition, techniques known as culling were implemented, which involve avoiding the rendering of objects that are outside the user’s field of view or hidden behind other elements. This optimization significantly reduced the amount of graphical information processed at any given time, resulting in a substantial improvement in the overall performance of the virtual environment.

The combination of these strategies, supported by the native capabilities of Unreal Engine 5, enabled stable execution with low response times and consistent visual quality. Consequently, TOLUCATRAIN VR 1.0 offered an immersive and visually appealing experience suitable for educational purposes.

Proof-of-concept design

In order to meet the criteria for Technology Readiness Level 4 (TRL4), a Mixed Methods Sequential Explanatory Design was adopted for the proof-of-concept testing, aimed at leveraging both the richness of quantitative data and the depth of qualitative perceptions (Creswell & Plano Clark, 2018). In the first phase, numerical data were collected regarding usability and the cognitive performance of participants through standardized instruments. In the second phase, semi-structured interviews and open-ended questionnaire items were conducted to explore users’ and instructors’ experiences and opinions in greater depth, enabling a more nuanced interpretation and enrichment of the statistical results. According to Salvador-Carulla et al., the Levels of Technology Readiness Level (TRL) framework has been adapted across various disciplines, including the assessment of technological readiness in higher education, providing a structured approach for implementing new resources and technologies (Salvador-Carulla et al. 2024). Furthermore, TRL can serve as a useful tool to assess the maturity and preparedness of different technologies, offering clear guidelines for their development and integration in educational and research contexts (Salvador-Carulla et al. 2024). The adoption of this taxonomy facilitates dialogue with educational innovation policies and evidence-based technology evaluation, particularly valuable in public education systems seeking to strategically incorporate VR resources, as is the case with this project.

Participants and sampling

The sample consisted of 196 engineering students from the Instituto Politécnico Nacional (IPN), distributed across two campuses: 102 participants from the UPIICSA, and 94 from the UPIIP. The sample size was determined through a statistical power calculation (G*Power 3.1), considering a medium effect size (d = 0.50), α = 0.05, and power = 0.80, yielding a minimum requirement of 128 participants; the final sample exceeded this threshold, ensuring robustness for analysis.

A stratified convenience sampling method was used, selecting students enrolled in railway engineering or related programs. Inclusion criteria were: (a) not having previously participated in similar VR studies, (b) providing signed informed consent, and (c) possessing the physical and cognitive ability to use VR devices. Participants’ ages ranged from 18 to 56 years (M = 22.8; SD = 5.4), with a gender distribution of 69% male and 31% female, as shown in Figure 3, Participant demographics. Additionally, 10 volunteer students (5 from each campus) and 5 specialized instructors were interviewed for the qualitative phase.

Figure 3

Measurement instruments

In addition to digital instruments, physical educational materials were provided to complement the virtual environment. These included printed infographics, technical manuals, and evaluation exercises, all designed to reinforce the knowledge acquired during the simulation. As shown in Figure 4, students accessed these resources during the sessions, which facilitated enriched learning through both visual and written elements. Each workstation was also equipped with electronic tablets displaying explanatory videos, enabling a multisensory learning experience.

Figure 4

To evaluate both the quality of the resource and participant satisfaction, the following instruments were employed:

• System Usability Scale (SUS) (Brooke 1996): Comprising 10 items on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree). Cronbach’s alpha was calculated to assess internal consistency.

• Ad hoc Satisfaction Questionnaire: Contained 8 items on a 5-point Likert scale, validated by a panel of experts in railway education (Content Validity Index, CVI > 0.80) and pilot-tested with 20 students.

• Knowledge Test: 25 multiple-choice questions designed under the supervision of the Instituto Politécnico Nacional (IPN) instructors. It was administered both before (pre-test) and after (post-test) the VR experience. The user’s difficulty and satisfaction index was calculated.

Procedure

Sessions were conducted in two periods (March and April 2025) in laboratories equipped with Meta Quest 3 and Meta Quest Pro headsets. Each group of 10 participants followed the protocol below, with a total session time of 90 minutes:

• Welcome and Instructions: Explanation of objectives, distribution of digital and educational materials.

• Pre-test: Administration of the knowledge test using Google Forms.

• Training: A brief 3-minute tutorial on how to operate the VR devices and navigate the environment.

• VR Experience: Execution of predefined tasks (panel inspections, operation procedure simulations) with a maximum duration of 8 minutes.

• Questionnaires: System Usability Scale (SUS) and the satisfaction questionnaire.

• Post-test: Re-application of the knowledge test.

Throughout the entire process, a researcher was present to provide technical assistance and record qualitative field observations.

Data collection and management

All data, both quantitative and qualitative, were collected using Google Forms. Responses were anonymized through alphanumeric codes to preserve participant confidentiality. The data was then organized and analyzed within the same platform, eliminating the need for external systems or specialized software.

Qualitative data analysis

The qualitative analysis was based on open-ended responses collected via Google Forms. The thematic analysis approach proposed by Braun and Clarke (2006) was applied, following six steps: data familiarization, initial coding, theme identification, theme review, theme definition, and final reporting. Coding was independently performed by two researchers to ensure agreement in interpretation. Results were validated through triangulation and participant review to ensure coherence and credibility in the analysis.

Quality control and study conditions

Basic control measures were established to ensure data validity and a positive user experience. Although specialized statistical software was not used, the digital forms allowed for effective data organization and filtering, preventing duplicate entries and ensuring consistency.

In addition to addressing possible side effects associated with VR technology use, specific measures were implemented to safeguard participant well-being throughout the experience. Continuous monitoring was conducted before, during, and after each session to detect potential symptoms of cybersickness (e.g., dizziness, nausea, visual fatigue). To ensure optimal conditions, sessions were held in spaces with stable lighting and temperature, and VR devices were calibrated before each use. As shown in Figure 5, an immediate interruption protocol was in place in case of severe discomfort, ensuring the safety, comfort, and physical integrity of all participants at all times.

Figure 5

Application at another academic site

As part of the validation process for the virtual environment and the educational resources developed, an additional proof-of-concept test was conducted at the Unidad Profesional Interdisciplinaria de Ingeniería (UPIIP) of the Instituto Politécnico Nacional (IPN), located in Palenque, Chiapas, as shown in Figure 6. This activity replicated the protocol used in the previous proof-of-concept tests, employing the same virtual reality devices (see Figure 7), printed materials, multimedia resources, and assessment methodologies implemented during the initial experience.

Figure 6

Figure 7

The implementation of this test involved the collaboration of Professor Benjamín Jacobo Donnadieu, Head of the Railway Engineering Department at UPIIP, whose participation was essential for the academic and logistical coordination of the exercise. Students from the Railway Engineering academic program also participated. This inter-institutional collaboration helped demonstrate the adaptability of the virtual environment in different educational contexts, as well as its potential for scalability for future applications at other Instituto Politécnico Nacional campuses.

Results

In the usability dimension, a digital questionnaire based on the System Usability Scale (SUS) was administered. As shown in Figure 8, the average result of 78.4 out of 100 indicates that most users found the resource easy to use and functional.

Figure 8

Cognitive performance showed significant improvement after the VR experience. Before using the resource, students achieved an average of 62.3 out of 100 on the technical knowledge test. After the VR experience, the average increased to 81.7, representing a substantial improvement in understanding. The difference between pre-test and post-test results indicates that the resource had a positive and significant impact on the students’ technical comprehension, with an average gain of nearly 20 points. This reinforces the value of the immersive environment as an effective tool for railway engineering training.

Figure 9 shows that the average interaction time per participant was 34.2 minutes. Additionally, 85% of students completed the practical tasks on their first attempt. Only 15% required a second attempt, primarily in tasks related to fault diagnosis.

Figure 9

From the qualitative analysis, four core themes emerged: (a) motivation and engagement: participants highlighted the playful and immersive nature as a motivating factor; (b) realism and knowledge transfer: the fidelity of the virtual environment and its potential for transferring knowledge to real-world contexts were valued; (c) initial barriers: some users reported difficulties with the joystick and VR lens calibration during the first few minutes; and (d) suggestions for improvement: participants recommended including maintenance scenarios and emergency simulations.

The voices of students and teachers who experienced TOLUCATRAIN VR 1.0 provide a deeper understanding of its educational value. Beyond its technical utility, their testimonies reveal emotions, expectations, and modes of appropriation that are difficult to capture with structured instruments. Below are excerpts that illustrate how the immersive environment was reinterpreted as a meaningful, innovative, and, in many cases, transformative experience:

Regarding possible improvements or changes they would make to the viewer, responses ranged from full satisfaction to specific suggestions:

Finally, participants were asked what aspects of the viewer they found unpleasant. The responses reflect a largely positive evaluation of the experience:

These direct quotes strengthen the narrative of experiential learning and confirm that, while there are areas for improvement, the virtual environment was perceived by its users as an effective, accessible, and enriching educational tool.

To effectively develop and evaluate the outcomes during the two proof-of-concept trials, an evaluation quadrant was created, as shown in Figure 10. This quadrant categorizes learning resources based on their usability and effectiveness. Quadrant One identifies potentially effective resources with poor usability. Quadrant Two represents the ideal: high usability and high effectiveness. Quadrant Three groups resources with low usability and low effectiveness. Finally, Quadrant Four describes resources that are easy to use but have low impact on learning outcomes.

Figure 10

For TOLUCATRAIN VR 1.0, the goal was to position the educational resource within Quadrant Two, ensuring that the interaction was intuitive and significantly facilitated learning about railway engineering. The evaluation process aimed to determine whether the resource required improvements in usability (Quadrant One), effectiveness (Quadrant Four), or both (Quadrant Three) in order to achieve high user satisfaction and effective learning outcomes.

The results indicated that 88% of participants would recommend TOLUCATRAIN VR to colleagues, while 85% positively evaluated the technical support provided during the trials. Teachers also highlighted the didactic value of the resource in complementing laboratory practice and reinforcing theoretical content.

The integration of quantitative and qualitative data demonstrates that the VR experience not only enhances the learning of railway standards and procedures, but also strengthens students’ motivation and confidence when approaching complex tasks in controlled environments.

The qualitative data collected through the instruments applied to participants confirm and enrich the quantitative findings related to usability, motivation, and knowledge appropriation. Student statements such as “I felt like I was really inside the train” or “I had never understood so well how a track switch works” support the high SUS scores and illustrate how the immersive experience enhances both comprehension and engagement with complex concepts linked to technical knowledge. Likewise, comments highlighting the value of the resource for self-directed learning and understanding the railway environment reveal an effective process of appropriation. Altogether, the evidence suggests that the use of open immersive technologies is viable not only from a technical standpoint but also from a psycho-pedagogical perspective.

Reliability analysis revealed Cronbach’s alpha of .89, demonstrating high internal consistency. 66% of participants rated navigation as easy or very easy, and 78% reported being satisfied or very satisfied with the resource, supporting the conclusion that the educational tool is viable for learning.

Discussion

The implementation of the TOLUCATRAIN VR 1.0 system as an OER represents a strategic contribution to strengthening technical railway education from the perspectives of educational justice, pedagogical innovation, and equitable access to specialized knowledge. This initiative aligns with UNESCO (2024b) guidelines and the principles of critical educational technology, recognizing Virtual Reality (VR) as a key mediating tool to reduce structural barriers in accessing practical learning environments.

The Dubai Declaration (UNESCO 2024a) emphasizes that open education serves as a driver for inclusive, human-centered digital transformation. In this context, TOLUCATRAIN VR 1.0 is positioned as a public digital good, offering an immersive, safe, and accessible environment for the situated learning of complex railway operations. This perspective is reinforced by Ariza-Colpas et al. (2020), who identify VR as a privileged medium for the social appropriation of technical knowledge.

The system was developed using Unreal Engine and 3D scanning technologies (photogrammetry and LiDAR) to create a digital twin of the Mexico–Toluca Interurban Train, enabling an immersive experience that supports experiential learning. This approach aligns with Mendoza-García et al. (2024), who documented improvements in student performance when exposed to virtual environments.

The evaluation was based on an explanatory sequential mixed methods design (Creswell & Plano Clark 2018), combining quantitative and qualitative evidence. Cognitive gains were observed (an average increase of +20 points in post-test results), alongside high usability scores (78.4 on the SUS scale) and user satisfaction (78%). The reliability of the instruments (α = 0.89) reinforces the validity of the methodological approach.

The qualitative thematic analysis (Braun & Clarke 2006) revealed five key themes: motivation driven by immersion, fidelity of the virtual environment, transferability to real-world contexts, initial difficulties with controls, and suggestions for new scenarios. This dialogue highlights the importance of user-centered design and the pedagogical value of OER.

The voices of students and educators offer a deeper understanding of the educational value of the resource. The following excerpts illustrate how users reinterpreted the experience:

• “I had never been so close to a real train. I felt like I was inside, operating it, and that helped me understand things I only saw in diagrams during class.” (Student, UPIICSA)

• “It’s a very useful tool. Students are more motivated and ask questions they didn’t ask before.” (Teacher, UPIIP)

• “I would like to see more scenarios, like what to do if there’s a failure in the electrical system.” (Student, UPIIP)

• “It’s not just the technology. It’s the way we can learn without putting anyone or anything at risk.” (Teacher, UPIICSA)

The project incorporated well-being best practices to prevent adverse effects associated with VR use: short sessions (max. 8 minutes), ergonomic setup, and device calibration. In line with Wong, Hui and Kong (2023), the study acknowledges that comfort influences technological acceptance in education.

Of particular note is the use of the usability and effectiveness evaluation quadrant, which placed the system in the high-usability/high-effectiveness area, validating its potential as a model for VR-based technical training.

Among the limitations identified are the absence of a control group, convenience sampling, and the short duration of the experience, which restrict the assessment of long-term effects. Longitudinal studies are recommended to expand content offerings, and validate curricular integration.

Despite considering accessibility standards (open formats, documentation, short sessions), challenges remain in achieving full inclusion of individuals with sensory, motor, or cognitive disabilities. The integration of voice commands, subtitles, and sensory adjustments are suggested to embrace accessibility as an ethical principle.

The system’s Technology Readiness Level 4 (TRL4) indicates it has moved beyond laboratory validation and is ready for real-world scenarios, as demonstrated at UPIICSA and UPIIP. This paves the way for curricular integration and scalability.

TOLUCATRAIN VR 1.0 demonstrates that collaboration between academia, industry, and the public sector can produce high-quality immersive OER with local relevance. As stated by Iniesto et al. (2021), open resources designed ethically, accessibly, and collaboratively can transform technical education.

This project proves that high-quality immersive solutions can be developed in the Global South, even within contexts of technological limitations. By integrating technological sovereignty, educational justice, and situated design, it stands as a valuable contribution for other contexts seeking a more equitable, open, and sustainable approach to technical education.

The data collected during the trials not only demonstrate benefits for student learning and motivation, but also prompt critical reflection on contemporary educational technologies. In this sense, the implementation of an immersive OER like TOLUCATRAIN VR resonates with the principles of cognitive justice, equitable access, and technological reappropriation promoted by the open education and science movements. The opportunity to train railway engineers through environments that are free, locally designed, and contextually relevant contrasts with dominant commercial options, which often fail to be adopted in less advantaged countries. This positions the resource not only as a didactic tool, but as an act of public policy in its role as a common good.

Conclusions

This contribution provides an overview of the development and proof of concept (TRL4) of an immersive OER focused on training railway engineers. Although the results are encouraging in terms of usability, cognitive gain, and user appropriation, they should be interpreted in light of methodological limitations. In particular, it should be recognized that, given the project’s current development stage as a proof of concept (TRL4), as opposed to a pilot test, the sample size was limited, a complete experimental design has not yet been applied, and the assessment of learning was restricted to a single intervention point, without longitudinal validation. Despite these limitations, the experience demonstrates the feasibility of designing, implementing, and validating immersive OER prototypes from public capacities, with an open-source approach, and in response to strategic educational needs.

The implications for educational practice are clear: these tools can complement and enrich technical education in fields where access to physical infrastructure is limited or costly. Furthermore, it is recommended that such resources be integrated into curricula gradually and flexibly, alongside teacher training strategies. For future research, it is suggested that longitudinal validations be conducted to measure the sustained impact of the OER, as well as comparative studies with other learning environments. Additionally, enhancing the accessibility of the resource for populations with disabilities and exploring its application in other educational levels and industrial sectors is crucial.

The TOLUCATRAIN VR 1.0 project constitutes an academic, technological, and pedagogical experience that invites a rethinking of the role of Virtual Reality (VR) as a catalyst for the transformation of technical education in Latin America. Through its design, implementation, and evaluation, new pathways can be opened for building more equitable, accessible, and sustainable educational futures, particularly in strategic fields such as railway engineering.

Immersive technologies not only complement but transform the possibilities for accessing high-fidelity practical learning environments. In a context where the national railway system is being revitalized as part of a priority public policy, the availability of a resource like TOLUCATRAIN VR 1.0 provides a tangible means to accelerate the training of technical talent with territorial relevance and public value.

The resource’s open nature, its code, formats, documentation, and modular structure, position it as an exemplary OER, transcending the technical to become a tool for the democratization of knowledge. This approach aligns with a tradition of Latin American educational technology with its own identity, centered on equitable access and situated learning.

The methodological combination employed, integrating quantitative and qualitative findings, contributes to ongoing discussions on how to comprehensively evaluate the impact of immersive technologies in education. This fusion of metrics and user voices allows for the construction of pedagogically meaningful evaluations aimed at continuous improvement and learning through practice.

The meaningful appropriation of the resource by students and educators, as reflected in the proof-of-concept tests, demonstrates the value of a co-creative pedagogy. The suggestions from those who used it help enhance its design and move toward new, more adaptive and functional versions.

This development also highlights the capacity of public institutions to generate impactful technological solutions. The collaboration among units of the Instituto Politécnico Nacional (UPIICSA, UPIIP, ESIA, UPIIH) shows that innovation can emerge from the public sector when objectives, technical capabilities, and social commitment are aligned.

From an educational policy perspective, this project represents a commitment to expanding the ecosystem of open resources in Latin America, with a sovereign and critical approach. In contrast to imported solutions, TOLUCATRAIN VR 1.0 offers a locally grounded model, based on academic partnerships, internal capacities, and situated knowledge. It is a replicable pathway for other contexts in the Global South.

Challenges remain. In terms of accessibility, it is necessary to advance toward more inclusive versions for people with sensory, motor, or neurodivergent conditions. The design must incorporate principles of universal accessibility and comply with standards such as Web Content Accessibility Guidelines 2.1 (WCAG 2.1) (W3C 2018).

It will also be necessary to scale its use, validate its curricular integration, and ensure interoperability with institutional platforms. Advancing toward higher levels of technological maturity (TRL5–TRL7) will require longitudinal validations and robust feedback mechanisms.

This project prompts reflection on what kind of educational experiences we want to build, for whom, with what purpose, and from what ethical frameworks. Beyond being a tool, TOLUCATRAIN VR 1.0 represents a methodology, an institutional vision, and an ethic of care.

Embodying the institutional capacity to generate high-quality OER, this initiative offers an innovative and contextually relevant model for technical education. TOLUCATRAIN VR 1.0 embodies the real possibility of articulating institutional capacities for the design of high-quality Open Educational Resources, with technical, pedagogical, and social impact. Its development and validation reaffirm that Latin America can produce, from the public sphere, tools that not only respond to the demands of inclusive digital transformation, but do so with identity, educational justice, and territorial commitment. Its integration into the Latin American OER ecosystem supports the advancement of teaching and learning practices with identity, with technologies serving the common good. This case underscores the importance of continuing to invest in open, critical, and identity-driven education in Latin America.

Data Accessibility Statement

The data generated and analyzed during the development of the TOLUCATRAIN VR 1.0 project, as well as the learning object itself, cannot be publicly shared at this time, as they are currently undergoing copyright registration. Once this process is complete, their publication will be considered in alignment with the commitment to open science and quality technological education, through various institutional platforms and the research group’s microsite workspace: http://giies.com.mx.

Ethics and consent

This study involved human participants (engineering students and instructors) and complied with institutional ethical standards for educational research at the Instituto Politécnico Nacional (IPN). The study was classified as minimal risk, as it involved non-invasive educational interventions and anonymous data collection.

All participants provided signed informed consent prior to participation. Participation was voluntary, and confidentiality was ensured through anonymization of responses using alphanumeric codes.

According to institutional regulations, formal ethics committee approval was not required for non-clinical, minimal-risk educational research conducted within regular academic activities. However, the study adhered to principles of voluntary participation, data protection, and participant well-being.

Competing Interests

The authors declare no conflict of interest. This development responds to an institutional need of the IPN and reflects the academic commitment of the Laboratory of Educational Informatics and Sociocybernetics at UPIICSA to generate high-quality open educational resources.

Author contributions

• Claudia Marina Vicario Solórzano: General coordination of the project, methodological design, contribution to the writing of the introduction and theoretical framework, writing of the abstract, discussion, and conclusions, as well as revision for coherence and stylistic editing of the document. Responsible for change control and responses to reviewers.

• Victor Joohvan Veraza: Lead programmer in Unreal Engine 5, technical lead for the proof of concept, author of the methods and results sections.

• Emmanuel González Rogel: IT and logistical coordinator of the project, co-author of the summary, introduction and conclusions sections for change control and response to reviewers.

• Aquiles Raziel Rojas Martínez: Corresponding author, writer of the introduction, and contributor to the discussion section.

• Karla Josette Salas Montoya: Logistics coordinator for the proof of concept, co-author of the methods and results sections, and translator of the article.