Virtual reality (VR) has emerged as a promising tool for transforming science, technology, engineering, and mathematics (STEM) education, yet its adoption in K–12 classrooms remains uneven and often limited to short-term pilots. While prior studies highlight VR’s potential to increase engagement and support conceptual understanding, questions persist about scalability, sustainability, and equity in implementation. This paper addresses these gaps by synthesizing recent scholarship and proposing a structured framework of best practices for integrating VR into K–12 STEM education. Drawing on academic literature, U.S. policy reports, and case studies, we identify persistent challenges that include high costs, lack of teacher preparation, infrastructure disparities, and overlooked accessibility concerns. We use these findings to inform a phased implementation roadmap. Our framework emphasizes assessment and planning, technical integration, teacher preparation, student implementation, and iterative evaluation, providing actionable strategies for schools and districts. Results of this synthesis indicate that successful VR adoption requires coordinated attention to pedagogy, funding, professional development, and equity. We conclude that moving VR from isolated novelty projects to sustainable and equitable tools in STEM classrooms depends on aligning technology with curricular goals, investing in teacher pipelines, and embedding VR within long-term evaluation and improvement cycles.

This paper describes the use of platforms for the development of virtual applications as tools for modeling of robot manipulators. The proposal is based on take advantage of the potential that these platforms currently have for solving the rigid body dynamics, which easily allows modeling the mechanical aspects of the manipulator. On the other hand, the possibility offered by these platforms of incorporate programming code in conventional languages allows to modeling the dynamic behavior of real physical systems, such as sensors and actuators, which allows implementing the development of the instrumentation and control stage of an industrial robot in the same way as a real one. Using these platforms allows the modeling from the bases of any manipulator robot. The modeling of a reconfigurable parallel robot is presented as a case study. En este trabajo se propone el uso de plataformas para el desarrollo de aplicaciones virtuales como herramientas para el modelado de robots manipuladores. La propuesta se basa en aprovechar el gran potencial que actualmente tienen estas plataformas para solucionar la dinámica de cuerpos rígidos, lo que permite modelar de forma sencilla los aspectos mecánicos del manipulador. Por otro lado, la posibilidad ofrecida por estas plataformas de incorporar código de programación en lenguajes convencionales, permite modelar el comportamiento dinámico de sistemas físicos reales, tales como sensores y actuadores, lo que hace posible la implementación de una etapa virtual de instrumentación y control tal y como se realiza en un robot real. El uso de estas plataformas permite modelar desde cero cualquier robot manipulador. El modelado de un robot paralelo reconfigurable es presentado como caso de estudio.

Currently, emerging technologies such as digital twins, through the application of frontier techniques, have achieved physics-based simulations that reduce time and costs. Hence, its application is of the utmost importance in the industry, mainly in the sugar drying process of sugar mills for an updated version of the process. Sugar mills lack process control, leading to unexpected issues. Sugar mills with poor process control cause operational problems. This article presents significant innovation in the field of industrial process optimisation through the integration of digital twins with the k-ε standard model in computational fluid dynamics (CFD). The primary objective of this publication is to predict the ideal conditions of a centrifugal sugar dryer using CFD through the k-ε standard model to analyse the aerodynamic behaviour of the ambient air by applying heat through heat exchangers to obtain a suitable mass flow. The mathematical model was carried out under an energy balance to the thermodynamic system to study the behaviour through a simulation in MATLAB R2017 and an air-fluid simulation of drying with software CFD 2015. The results proved that the model of the thermal system and frontier conditions, when applying CFD, carried our simulation and remained stable. The ideal operating conditions of the centrifugal sugar dryer can be predicted effectively, with an energy saving of 4.25%.

Large language models (LLMs) such as GPT-5, Claude, and Grok are increasingly incorporated into engineering education, yet their effectiveness on discipline-specific cognitive tasks remains insufficiently characterized. Engineering learning relies heavily on conceptual understanding, multi-step problem solving, and higher-order reasoning, making it essential to evaluate how LLMs perform across these contexts. This study investigates the performance of three LLMs across four undergraduate engineering courses, Thermodynamics, Computer Programming, Systems and Control, and Circuits, using multiple-choice questions aligned with Bloom’s Taxonomy. We examine differences in accuracy and reasoning quality across cognitive levels, compare zero-shot and context-assisted prompting, and benchmark LLM performance against human learners in Computer Programming. Forty-eight instructor-designed questions (12 per course) spanning all Bloom levels were administered to each model. Responses were evaluated using accuracy, a rubric-based reasoning score, and a Composite Performance Index capturing completeness, efficiency, and quality of support. Human performance data were collected through an IRB-approved assessment with undergraduate students. Results show that Claude demonstrated the strongest overall performance and consistency, GPT-5 excelled in analytical reasoning, and Grok exhibited variable accuracy with limited higher-order reasoning. Providing contextual information improved GPT-5’s performance but had minimal impact on Claude or Grok. In Programming, Claude and GPT-5 outperformed human participants across all Bloom levels. These findings suggest that LLMs can support engineering learning when paired with instructional strategies emphasizing verification and critical evaluation. However, persistent limitations at higher cognitive levels highlight the importance of guided, responsible integration, supported by a practical instructional guide for effective LLM use across engineering tasks.

The presence of silica in iron ore concentrate can have significant negative impacts on the efficiency and quality of steel production. As such, providing engineers with early and reliable information about the purity of iron ore concentrate is crucial for smooth mining operations. This paper reports on the development of a long short-term memory (LSTM) network and an ensemble data interpolation technique to enhance quality prediction in the froth flotation process of an iron ore mine. Our results demonstrate the ability of our model to accurately predict the silica content of iron ore concentrate on a minute-by-minute basis, as well as the ability to forecast hours in advance.

Generative artificial intelligence (AI), especially large language models (LLMs) that can write and debug code, is changing how students approach programming work in engineering education. Unlike more open-ended conceptual or modeling tasks, programming fits closely with what these systems do well: generating syntax, fixing errors, building procedural logic, and completing code structures. Hence, programming coursework may be one of the areas in which AI changes performance patterns in a measurable way. This study examines whether that shift appears in actual student outcomes. Using a retrospective pre/post design, it compares results from a pre-AI period (2021–2022) with results from a post-AI period (2023–2025), when generative AI tools became widely available to students. The focal assessment is a comprehensive programming project graded with the same rubric across multiple sections and terms. Performance is evaluated through descriptive statistics, distributional comparisons, and mastery thresholds (≥80%). The post-AI period shows a rise in overall scores, along with strong clustering near the top of the scale. Lower- and middle-range scores become much less common, most students fall in the highest score band, and overall variability declines. These results suggest that generative AI acts as a procedural equalizer in programming contexts, referring to the role of generative AI in reducing performance differences by assisting with rule-based, syntax-driven, and execution-oriented aspects of tasks, thereby raising baseline outcomes while compressing variation among students. It appears to raise lower-end performance and make outcomes more consistent, but it also narrows the spread among stronger students and creates a ceiling effect. That pattern raises questions about assessment validity, skill differentiation, and what “mastery” means when AI can handle much of the procedural work. Using multi-term data from authentic online courses, this study adds empirical evidence to the growing literature on AI in engineering education and identifies programming coursework as a setting where generative AI may have already changed performance dynamics in a structural way.