Machine learning techniques for the prediction of performance in the learning process are primarily studied at the post-secondary level of education. Data sets are usually large at that level, resulting in predictive models that have a high accuracy. In contrast, limited research has been conducted at the secondary school level, mostly due to the typically small data sizes and the unique educational challenges of that level. In the present study we aimed to address such issues by implementing a model to predict the final performance of lower secondary school students in a course on Informatics, using a teaching scenario that combined the flipped classroom, a variation of the jigsaw technique and educational robotics. Given the student grades from the first third of the year, as well as demographics data, we used an IBM data analytics tool to create and test several predictive models. The CHAID algorithm achieved the highest accuracy (82.14%) and AUC value, outperforming others like Quest, C5, BN, and RF. The tool also pinpointed the educational activity that was the most significant predictor of final performance, indicating its strong instructional value. Despite the relatively small dataset, the results suggest that careful parameter selection can yield models that predict learner performance with a high accuracy and assist educators in continuously improving their teaching for better student outcomes.
Assessing competencies in engineering education increasingly requires digital assessment approaches that support learning regulation, instructional decision-making, and educational quality, rather than focusing solely on measurement efficiency. Computerized adaptive testing (CAT), grounded in Item Response Theory (IRT), provides a robust methodological foundation for personalized assessment. However, its pedagogical effectiveness in formative contexts depends critically on curriculum alignment, diagnostic capacity, and adaptive control strategies. This study proposes and evaluates a formative adaptive assessment framework for engineering education that integrates an IRT-based CAT engine with a Bayesian network- based diagnostic component. The framework is designed to support competency-oriented feedback, learning monitoring, and instructional interpretation within a curriculum-aligned assessment structure. Assessment relies on dichotomous multiple-choice items explicitly aligned with engineering learning outcomes, while item selection dynamically adapts to learners' evolving proficiency estimates. In parallel, probabilistic diagnostic modelling prioritizes under-assessed competencies throughout the adaptive process. Item calibration was conducted using empirical data collected from 612 university students in computer science, and system performance was examined through a simulation-based evaluation involving 500 simulated learners. Results demonstrate high estimation accuracy (r = 0.912) and satisfactory reliability for formative use across most learner profiles. Reduced precision at the extremes of the proficiency continuum and imbalances in item exposure were also observed, highlighting structural limitations primarily related to item bank coverage and curriculum representation rather than to the adaptive algorithms themselves. Overall, the proposed framework positions adaptive assessment as a pedagogically grounded tool for formative learning support, instructional decision-making, and quality assurance in engineering education.
The growing incorporation of generative artificial intelligence (GAI) in educational settings is transforming the teaching of technical writing in engineering education. However, there is little evidence on how students adopt these technologies in the development of technical reports, a key transversal skill in their future professional practice. This case study analyzes the relationship between GAI acceptance and self-efficacy in technical report writing among 158 engineering students at a national university in Peru. A quantitative, correlational approach and a non-experimental design were used. The results indicate that most students show moderate to high levels of technological acceptance and self-efficacy in writing technical reports, with a clear predominance of positive attitudes towards the use of GAI. Significant positive correlations were found between the dimensions of perceived use, ease of use, and intention to use GAI with the key stages of planning, drafting, and reviewing technical reports. It is concluded that the effective integration of GAI improves academic and professional engineering education by strengthening students' confidence and skills in specialized writing. Finally, it is recommended that future research incorporate variables such as intrinsic motivation and critical thinking, considering their application in different branches of engineering.
This study aims to identify opportunities for improving a project-based, design-oriented learning approach in design studies, focusing on mechanisms that promote student motivation, competence, and identity development in the context of interdisciplinary cooperation. The following research methods were employed: scientific literature analysis and focus group discussions to ensure data triangulation. Action pilot research was conducted during the development of a design project, with its stages aligned to the phases of the design process. The findings highlight key factors contributing to students' personal development and successful project outcomes. These include the organization and material infrastructure of the study process, the importance of collaboration across institutional units, among lecturers, and between lecturers and students. The study concludes that strengthening collaboration and stakeholder involvement in the improvement of the study process enhances students' learning experiences, facilitates knowledge exchange, and increases motivation and learning effectiveness.
This study explains undergraduate engineering students' engagement in blended learning through the lens of Herzberg's Two-Factor Theory and the mediating role of learning motivation. Survey data from 656 engineering students enrolled in blended courses at Hanoi University of Science and Technology were analyzed using PLS-SEM. Results indicated that motivator factors strongly influenced learning motivation, which in turn significantly predicted learning engagement, while hygiene factors had only a small direct effect on learning motivation but an indirect role through motivator factors. Mediation analysis confirmed a ing 58% of motivation and 78% of engagement variance. The findings highlight that stable course conditions enable pedagogy-centered motivators to activate motivation and sustain engagement. Practical implications emphasize the need for designing blended courses that combine reliable technical environments with intrinsically motivating learning designs.
As engineering higher education adapts to the demands of Industry 4.0 (I4.0), Faculty readiness to incorporate disruptive technologies in their classes becomes critical. This study investigates engineering Faculty's familiarity with I4.0 technologies (I4.0T), perceived importance of their incorporation in engineering education, and intentions to implement them in teaching. A sample of 107 Faculty members from local colleges and universities completed self-report questionnaires, with results analyzed quantitatively. Findings reveal moderate familiarity with I4.0T, low usage levels, and limited self-perceived competence. While half of the Faculty acknowledge the importance of I4.0T incorporation, fewer than one-third feel capable of implementing this incorporation. Approximately three-quarters express little interest or insufficient knowledge to form a relevant opinion. Faculty intentions to incorporate I4.0T are strongly associated with their frequency of use and perceived value. The paper recommends strategies to increase technology use, boost Faculty competence, and support the incorporation of I4.0T into higher education teaching practices.
A substantial percentage of the world's energy consumption (almost 40%) and carbon dioxide (CO2) emissions (around 37%) come from the construction industry, especially schools. This work presents a new hybrid artificial intelligence (AI) engineering model that aims to maximize energy performance on campuses in a holistic way. Modules for data-driven forecasting, metaheuristic optimization, and real-time adaptive control are all part of the concept. A thorough energy simulation of a university campus building is used in conjunction with the AI model to assess its performance through a co-simulation framework. Findings show that yearly peak electricity demand may be reduced by 18.7% and total site energy consumption by 22.4% when compared to a baseline building management system, all while keeping indoor thermal comfort levels high. According to the study, one effective way to make school buildings smart, eco-friendly, and energy efficient is to use a hybrid AI-driven method.
Soft skills development remains a critical challenge in higher education, especially in large, multidisciplinary learning environments where students vary in background, motivation, and self-regulatory capacity. Although problem-based learning (PBL) is broadly acknowledged as a successful method for developing critical skills and abilities such as planning, collaboration, and reflective thinking, traditional implementations often face constraints in personalization, coordination, and formative assessment. This study introduces an enhanced PBL model augmented by artificial intelligence (AI) Agents to support soft skills acquisition in blended learning contexts. The AI Agent, leveraging natural language processing (NLP) and automation via the n8n platform, functioned as a virtual assistant to facilitate task planning, peer coordination, self-monitoring, and timely feedback. A quasi-experimental study was carried out involving three groups of students (N = 263) participating in a course focused on developing soft skills, comprising one group using AI-supported PBL, another group using traditional PBL, and a control group taught through standard instructional methods. A mixed-methods analysis demonstrated that the group utilizing AI support exhibited statistically notable enhancements in their outcomes across six soft skill domains, particularly in planning, group work, and reflective learning (p < 0.001). Behavioral data from LMS logs and product assessments further validated enhanced collaboration, consistency, and accuracy in self-assessment among AI-supported learners. The findings demonstrate that integrating AI Agents into PBL not only reduces instructor workload and enhances instructional equity but also empowers learners through personalized, data-driven scaffolding. This approach offers promising implications for scalable, technology-enhanced soft skills education across disciplines.
Engineering is central to addressing the world’s urgent challenges, from energy and sustainability to access to clean water and safer infrastructure. The UN’s Sustainable Development Goals (SDGs) offer a globally accepted blueprint for these, for which the systemic embedding in engineering education is far from uniform. This paper explores pedagogical methods for integrating the SDGs into engineering education and discusses some of the challenges that educators are currently facing. Based on a literature review of worldwide practices, including project-based learning, problem-based learning, and interdisciplinary teamwork strategies for developing the sustainability competence of future engineers, obstacles such as curriculum overload, lack of faculty training, and a limited number of assessment instruments are emphasized. Based on the content taught in their engineering courses, as well as through compulsory climate change education for non-engineers, similar process flows are then considered to determine how they can contribute to addressing the challenges defined in SDG 9 (Industry, Innovation, and Infrastructure), SDG 12 (Responsible Consumption and Production), and SDG 13 (Climate Action).
This study investigates the integration of a smartphone-based virtual laboratory into a fourth-semester undergraduate fluid mechanics class on pump–piping systems. The virtual laboratory is designed according to constructive alignment and the SOLO taxonomy to foster deep learning. Students interact with realistic 3D system models, adjust component parameters, and receive real-time feedback based on physical simulations. To identify the effectiveness, a pre- and post-test with 26 paired responses showed a small overall improvement in general knowledge, with medium-to-large gains in specific methodological knowledge and selfassessed competence in handling real fluid systems. Student feedback was collected to assess the appeal of the teaching method. Students rate it highly positive (mean rating = 4.42/5), highlighting increased motivation, engagement, and active participation compared to conventional teaching. Future work will expand the app with additional levels targeting diverse learning objectives in fluid mechanics.
Makerspaces can potentially support learning and teaching across different learning environments using new ways to enhance 21st-century and engineering skills. Although there has been extensive work on makerspaces within museums and libraries, more work must be done concerning their effects on learning and teaching and how the spaces might enhance students’ engineering skills. This paper explores the notion of blended makerspaces—a new model, developed by the author, of blending hands-on, engineering, and technological skills— and their effects on learning. Based on a mixed-methods case study, this paper describes how the blended method used within the state-of-the-art makerspace stations creates critical thinking, inventiveness, and cooperation. This paper also examines how the spaces enhance life skills, such as flexibility and employability, preparing students for their prospective careers. The suggested blended makerspace method—learning in person, online, or both—emerges as a novel pedagogical structure. Furthermore, the study results indicate that through the infusion of blended activities in makerspace projects, students can enhance 21st-century engineering skills. This paper adds to the discussion on maker pedagogy. It suggests pathways for integrating makerspaces into educational contexts to enable students to develop engineering and other skills to be future-ready.
The rapid growth of immersive educational technology has opened up new possibilities in higher education, particularly in engineering education, where hands-on and experiential learning are crucial. This study examines the potential impact of utilizing the metaverse in lesson design on the way engineering is taught in schools. The study offers a comprehensive educational framework that incorporates gamified homework, collaborative work with avatars, 3D virtual worlds, and real-time simulations of engineering processes. It is based on theories of constructivist and experiential learning. The study employed a combination of methods and included undergraduate engineering students from core fields such as electronics and computer science. The participants utilized Unity and spatial platforms to learn in a metaverse-based environment, which replaced traditional lectures with coding worlds, virtual labs, and problem-solving scenarios. We used pre- and post-tests to determine how much students learned in mathematics. We also conducted focus groups and interviews with teachers to gauge their interest, ease of use, and perceived usefulness for teaching purposes. The results showed that students were significantly more motivated, engaged, and understanding when they used this method compared to traditional classroom methods. Students reported that using avatars made it easier for them to collaborate, think more clearly, and pay closer attention. The study also identified real-world challenges, including the need for faculty training in virtual instructional design, the difficulties of onboarding, and the limitations of technology. This study contributes to the growing field of metaverse-driven pedagogy by presenting a flexible and scalable approach to designing engineering curricula. It provides schools that want to incorporate immersive environments into their regular classes with helpful advice. The results show that the metaverse can help bridge the gap between theory and practice. This will make engineers more creative, flexible, and ready for the job market.
Graduates from the engineering field starting their careers in supply chain and logistics have begun to be balanced on cost efficiency, environmental sustainability, and social responsibility from the beginning of their careers, all considered the TBL (triple bottom line). This paper proposes a blended learning framework for TBL integration into the supply chain for undergraduate and postgraduate engineering students. The framework incorporates self-paced micro-learning, collaborative problem-solving seminars, work-integrated learning, and reflection to develop systems thinking, ethics, and quantitative decision-making. Practical advice on how to implement the blended approach is provided through a sample module, assessment rubrics, and program evaluation strategies. With relativity to other faculty, this approach is designed for single-course implementation all the way to program-wide flexibility. This framework builds on the gap between engineering education and blended sustainability frameworks, preparing graduates to manage a global supply chain ethically and sustainably.
Engineering capstones increasingly require students to deliver solutions that balance economic and environmental objectives under data scarcity. This paper proposes a guarantee-aware pedagogy that frames each project as a bi-objective program with explicit uncertainty sets and a lightweight correctness backbone. The pipeline comprises p-group reduction and factor screening; admissible surrogate modeling with a nearest-PSD repair for quadratic fits; e-constraint generation of representative Pareto sets (augmented to include selected non-supported points); and a transparent MCDA/LCA decision audit. Uncertainty is handled via three teachable Dials-Scenario (sample-average with concentration bounds), Budgeted-Robust (price-of-robustness parameter G), and Fuzzy a-cuts (a-dominance bands)-with an optional Wasserstein DRO extension. We define a single Credibility Index C = coverage – overfit that combines calibration and parsimony, and pair it with learning gains measured by Hedges’ g and optional 2PL IRT. A compact demonstrator (three e-levels) shows how the method yields interpretable Pareto fronts and auditable choices, while uncertainty dials trade protection for cost in predictable ways. Complexity tags and resource scheduling rules (assignment TU, queueing r < r\*) keep workloads feasible for classroom scale. Results indicate that the approach raises methodological transparency, improves reproducibility, and supports defensible suctainability-driven design decisions within a single semester.
Transforming engineering pedagogy is imperative in an era defined by global sustainability challenges. The United Nations’ Sustainable Development Goals (SDGs) offer innovative teaching methodologies by embedding necessary competencies to tackle complex virtual and real-world problems. The paper integrates engineering courses with SDG-6, SDG-7, SDG-9, and SDG-11. By viewing education pedagogy through the lens of the SDGs, this paper positions the shift from the traditional content-based model towards project- and problem-based models that cultivate ethical solutions, critical thinking, interdisciplinary collaboration, and the transformation of scholarly, industrial, and societal perspectives. At a policy level, this transformation entails reforming the curriculum and assessment framework that values both societal impact and technical competence. By viewing engineering education through the lens of the SDGs, this study positions pedagogy as a catalyst for preparing future engineers not only as innovators but also as responsible agents of sustainable global development.
With women underrepresented, engineering education in India continues to face persistent gender disparities. Despite various reforms and interventions across the decades, the imbalance is still apparent. This study examines gender representation in engineering education in India using both the systematic literature review and quantitative data analysis. The systematic literature review followed the PRISMA guidelines to review peer-reviewed open-access research articles. The review helped to identify the key factors that influence women’s participation in engineering education, such as social, cultural, organizational, and policy-based barriers. For the quantitative data analysis, the authors have referred to the data from the All-India Survey of Higher Education (AISHE) across multiple years (2012–2022). This study aims to understand the trends in female enrollment and graduation in engineering programs. The data was organized, cleaned, and represented. The graphical representation highlights year-wise patterns, progress, and areas where gender gaps persist. With the combined insights from SLR and AISHE data trends, the authors have proposed a conceptual framework that aims to improve female participation. The findings of this study are expected to inform institutions and policymakers on how to make the engineering education system more inclusive for female participants.
Artificial intelligence (AI) is entering engineering courses rapidly, yet tool-led adoption can weaken assessment validity and academic integrity. This paper presents the E-AIP (Engineering–AI Pedagogy) Framework, which centers three pillars learning Outcomes, Process Evidence, and Integrity & Ethics Guardrails and links them to design levers (AI function, task authenticity, feedback granularity, locus of agency). We define seven constructs, state eight propositions about alignment, validity moderation, authenticity, and agency, and operationalize E-AIP through a compact matrix (AI function × outcome type with required process evidence and guardrails). Two design patterns (CS1 debug-with-defense; circuits param-twins) illustrate classroom use; a lightweight adoption toolkit (two rubrics and an integrity or privacy checklist) supports immediate deployment. Additional patterns and full matrices appear in the online supplement. E-AIP enables instructors to capture AI’s benefits while preserving what scores validly claim to measure.
This study evaluated the developed cost-effective stepper motor laboratory equipment for undergraduate engineering students, addressing the challenges of financial constraints and limited lab access. Utilizing the ADDIE model, the equipment was designed to support basic to intermediate microcontroller-robotics applications through five experiments. Student performance was assessed by comparing traditional practice, simulations (Tinkercad), and the actual lab equipment across three lab activities, revealing significant improvements with the physical setup. For the remaining two experiments, where Tinkercad lacked the necessary components, a mixed-methods approach was employed. Quantitative survey results demonstrated strong agreement and satisfaction, with Cronbach’s alpha exceeding 0.90, confirming reliability. Qualitative thematic analysis, using Braun and Clarke’s 6-step method, highlighted user-friendliness and component-specific features as key strengths. Minor suggestions primarily focused on improving physical design. These findings validated the effectiveness of the developed stepper motor lab equipment in enhancing practical learning and bridging the gap between theoretical knowledge and microcontroller-robotics applications, particularly by overcoming the limitations of simulation-only learning.
This study aimed to analyse the impacts of artificial intelligence (AI) enabled creativity on divergent and convergent thinking among higher education students, focusing on how key factors—cognition, behaviour, interaction, ethics, and emotion—shape creative outcomes. The AI tools integration with academic environments has enhanced how students generate ideas, solve problems, and express originality. However, AI influences on critical cognitive processes and ethical considerations associated with creativity evidence are limited. The research adopts a quantitative approach, surveying a diverse sample of higher education students across various disciplines and institutions. Data were collected using a standardised questionnaire based on a Likert scale, covering variables such as divergence, convergence, metacognition, dependency, risk-taking, feedback, collaboration, transparency, confidence, and implementation. Structural equation modelling (SEM) was used to test the proposed relationships between these constructs. The findings reveal significant positive effects of cognition on emotion and behaviour, and of behaviour on creativity, while ethics and interaction showed complex, partly indirect pathways influencing creative outcomes. The model fit indices confirmed the robustness of the proposed framework, with acceptable values for CMIN/DF, RMSEA, and CFI. The study emphasises the need for educational institutions to design AI-integrated learning environments that promote ethical engagement, emotional well-being, and critical thinking. The results provide actionable insights for curriculum designers, educators, and policymakers seeking to harness AI for fostering student creativity while safeguarding academic integrity.
This study aims to critically explore the impact of peer learning on conceptual understanding in online engineering education. By leveraging current peer-reviewed articles, the research provides an extensive analysis of the efficacy, variations, and challenges of peer learning approaches in virtual engineering classrooms. A comprehensive systematic literature review (SLR) methodology is used in the study, adhering to PRISMA for transparent identification, screening, and assessment of suitable studies. The major databases searched for articles were PubMed, Scopus, and Google Scholar. The inclusion criteria were peer-reviewed articles published from 2015 to 2025, in the English language, and aligning with the scope of this study. The selected articles were then analyzed using a thematic analysis approach to identify patterns related to peer learning approaches like peer instruction, collaborative projects, and discussion forums, improving the conceptual gains of engineering students. Comparative findings were drawn from these parameters related to the study. Based on the extensive review and analysis of literature, a conceptual framework is presented in this paper, highlighting the relationships between peer learning approaches and their implications for online pedagogy. Recommendations focused on optimization of peer learning structures for the online environment in engineering and identifying support mechanisms for diverse learning groups. The review also identified research gaps, promoted adaptive peer learning models and longitudinal studies, and significantly contributed to the advancement of inclusive and effective online engineering education.
