Optimizing mechanical Computer Aided Design (CAD) learning methodology through intelligent metadata management for a sustainable engineering education research field: Online, distance and hybrid learning for tutoring CAD in mechanical engineering

Abstract

The author, motivated by her twenty-year experience in engineering education at different levels of the pedagogical scale, has been challenged to investigate novel processes of improving and eventually optimizing the learning procedure, under novel virtual learning environments. The bedtest of this research have been the two first year modules of Computer aided mechanical design (CAD I & CAD II). The research took place from July 2019 until April 2022, in real learning environments in higher engineering education. The stages have been based on the teaching modes applied: face to face, online and hybrid. The experimental procedure from which all outcomes have been revealed is divided in two major stages: the first stage refers to exclusive online learning environments and the second to hybrid learning modes, combining face-to-face, online and even mixed teaching spaces. On March 2020, for the first time in the history of education, online learning modes have been exclusively performed worldwide due to pandemic, directly affecting the global learning population, including students and educators. During exclusive online learning environments, a novel methodological approach has been applied, by using recently emerged technological tools that aimed to promote the relation of the module toughed with future tasks in mechanical engineering. Based on the experience under emergency remote teaching environments, the author has applied a more detailed learning strategy, performed in the entire population attending the online module. Several aspects of the educational procedure have been researched: evaluation of the learning strategy by a data analysis performed on the way tasks have been assessed online, asynchronous support by a social media video channel, students’ academic achievements expressed by their final grades through grade prediction and students’ satisfaction while attending their module expressed by student satisfaction prediction. Two hybrid models have been created, for grade prediction, and for predicting student’s satisfaction based on the new factors that have been revealed to affect students’ academic achievements under exclusive online learning environments. Both models were trained and forecasted students’ grades and students’ satisfaction with an R-squared equal to 1 after data fitting. The impact of different teaching approaches, based on the usage of two distinct learning tools (MS Teams and Moodle) and a combination of those two has been also investigated, and conclude that the use of multiple online platforms complicates the organisation of tasks and result to higher levels of students’ dropout. During hybrid learning spaces, a comparative analysis has been performed, in the performance of the social media channel (YouTube) that has been created and widely used under exclusive online spaces, and reapplied under hybrid ones. The author has investigated learners' viewing behaviors and engagement patterns through educational video analytics. The most important outcome is that there has not been a statistically significant difference in students viewing habits under different learning environments, proving its acceptance, as well as the long-term viability of the specific aspect of the learning strategy. A new GLAR model has been created revealing fourteen variables that have a statistical importance in the population of 2021-2022 out of which seven were validated from the previous model and the remaining seven were new, and revealed their importance in mixed learning environments. The statistically significant variables revealed have been optimized in order to determine their dominance over others through their optimum value and reveal the combination of variables that leads to an optimal allocation of the measures that need to be taken. The actions gradually performed have proven to respond to the needs of learners when attending the of Computer aided mechanical design module.