augmented reality (AR) has shown enormous potential as a teaching tool for improving comprehension in technical and engineering fields. This research investigates the influence of AR on the comprehension and interpretation of assembly drawings, with a concentration on gear train assembly drawing. We created an AR-based web application called GTAR (Gear Train with Augmented Reality), using CATIA CAD software, AR-JS is an innovative platform intended for the development of AR experiences on the Web and GitHub is a development platform permitting developers to write, save, control, and share their code. The AR-based web application enables students to see animated simulation of gear train system on their smartphones. This tool aims to link static representations with functional comprehension through the integration of dynamic visuals into conventional technical drawings. This article discusses the making process of the AR-based web application, its educational advantages, and the outcomes of an experimental research study conducted with the second year of the mechanical science and technology baccalaureate students at a technical high school in Casablanca, Morocco. Initial results indicate that AR enhances spatial thinking, reduces cognitive strain, and elevates precision in reading assembly details. Post-test findings indicate a significant improvement in the average score, approximately between 15 and 16 points. This extensive enhancement demonstrates the beneficial impact of AR on students' comprehension of the assembly drawings.

This research is to assess the impact of augmented reality (AR) on the learning performance of technical drawing students. This will be achieved by creating a mobile learning application named AROP (Augmented Reality for Orthogonal Projection) using Catia, Blender, Unity 3D software and Vuforia. This study investigates the impact of using AR technology in technical education on the proficiency of the first year students at ANOUAL technical high school in Casablanca, Morocco. We will focus on the technical drawing course and, more specifically, the orthogonal projection section, which is the subject of our study. This study involved 36 students. The students divided into two groups, with each group including 18 students. The evaluations showed that students who used the AROP learning application to learn orthogonal projection (Group 1) had superior results compared to those who used conventional techniques with an average pre-test score of 10.67, a maximum score of 13 and a minimum score of 9. The mean post test score for this group was 15.33. The maximum score is 18 and the minimum is 14. While Group 2 had an average pre-test score of 10.67, with a maximum score of 13 and a minimum score of 9. The mean post-test score for this group was 12.67. The maximum score is 15 and the minimum is 10. Additionally, the students of Group1 expressed good comments about their learning experience.

Understanding student academic performance is a cornerstone for developing sustainable educational practices that benefit students, teachers, policymakers, and society. This analysis directly impacts students’ ability to engage in and promote sustainable practices, thereby shaping their future academic success. While many studies focus on predicting student performance based on a set of features, our study takes an approach by reducing these features into factors and analyzing their impact. We aim to identify the factors influencing student performance within the middle school education system using a combined approach of Factor Analysis for Mixed Data (FAMD) and Multiple Linear Regression (MLR). Our analysis is based on a robust and reliable large dataset of 1,073,450 observations, encompassing qualitative and quantitative features. FAMD analysis identified four underlying factors: prior academic performance, academic delay, socioeconomic status, and class environment; all these factors have good to excellent reliability, with Cronbach’s Alpha values ranging from 0.809 to 0.930. Feeding these factors into the MLR produces a robust model that explains 88.53% of the variance in the CGPA, indicating a strong fit. Prior Academic Performance factor emerges as the most powerful predictor, accounting for 76.6% of the explained variance. Academic Delay follows, explaining 14.34% of the variance. Socioeconomic Status contributes 6.02%, and Class Environment adds 3.03%, reflecting smaller but meaningful impacts. All predictors are statistically significant (p < 0.001), confirming their critical roles in influencing student performance (CGPA). The insights gained from this study are critically important in the field of education. They enable teachers and educational leaders to identify at-risk students early and develop targeted interventions that address the factors influencing their performance. This approach aims to enhance learning outcomes, improve educational practices, and promote sustainable education.

Scientific research plays an essential role in the development of mankind through its ability to provide innovative solutions and technological advances in various fields. In this perpetual quest, researchers must be active members of the scientific community, sharing their experiences and publishing their work and discoveries. All research work generally begins with a literature review (LR), which involves delving into previous work to examine the current state of knowledge, situate a research topic, and thus identify opportunities and explore new avenues of research. This important step can be arduous and complicated for new researchers. This paper aims to address the challenge faced by new researchers in conducting LRs, particularly in formulating search queries, identifying and selecting relevant studies, and extracting data from each paper, by proposing a methodological approach and applying it through a practical example in the context of using artificial intelligence in education. The approach is guided by the PRISMA (preferred reporting items for systematic reviews and meta-analyses) framework and focuses on recent studies conducted in 2021, 2022, and 2023. The proposed approach identified 52 out of 336 relevant studies. 65% of these studies were deemed to be of high quality (Q1 and Q2 rankings), and 40% of the articles were published in high-impact academic journals (Q1). This approach is versatile and can be adapted to different fields.