Model Engineering College

This chapter explores the potential of artificial intelligence and data science in improving quality in education, focusing on aligning learners' needs, enhancing curriculum design, and improving assessment strategies. The chapter explains the uses of predictive analytics for identifying at-risk students, additional targeted paths for motivating the students, and AI-aided tools that foster collaborative learning. It also revises the problems and ethical concerns related to AI and data science applications in the educational context. By adopting these technologies, it will be easier for educators to create a much more engaging, open, and efficient learning environment, which is significantly better placed to help learners understand the nuances of the current learning environment in engineering and technology. This chapter aims to bolster the belief that AI and data science provide a promising avenue for enhancing engineering and technology education.

The comparator circuit is a fundamental building block in many digital systems, extensively used in applications such as arithmetic logic units (ALUs), memory systems, and microprocessors. Carbon nanotube field-effect transistors exhibit promising characteristics such as high carrier mobility, reduced leakage currents, and excellent mechanical properties when compared to MOS transistors making them a favorable candidate for next-generation electronic devices. This paper presents a 1-bit comparator utilizing Carbon Nanotube Field-Effect Transistors (CNTFETs) integrated with the Gate Diffusion Input (GDI) technique. The proposed design achieves significant performance improvements, demonstrating a 90.85% reduction in delay, a 45.73% reduction in power consumption, a 51.16% improvement in slew rate, a 2.39% increase in high noise margin, and an approximate 12.50% improvement in low noise margin.

In this paper, we consider a MAP/E_k/1 queue with working vacation. Customers arrive according to a Markovian arrival process and service time follows generalised Erlang distribution of order n. The first k stages of service are called preliminary service and the service in the remaining n−k stages is called the main service. When the system becomes empty at the time of completion of service, the server will go on a working vacation. Customers who arrive during the working vacation are provided only the main service. The server switches to normal mode when the vacation expires, or N customers are continuously served during working vacation, whichever occurs first. The customer in service at the working vacation expiration epoch is provided the service from the beginning. We analyse this model using the matrix-geometric method. We obtain the expected service time and waiting time of a tagged customer. Other performance measures are computed and a cost function is constructed to find optimal N value corresponding to the input parameter values.

Sign language is an important form of communication among the hearing and speech impaired community. However, the lack of interpreters and language barriers meant that it is often difficult for signers and non-signers to communicate effectively. This paper present a Sign Language Translation and Hand Gesture Identification using deep learning and computer vision techniques, in real-time. The operation of the proposed system has two modes: Letter Mode and Word Mode. In the Letter Mode, MediaPipe extracts the landmarks of the hand which are processed using the Random forest classifier for recognition of individual Individual Alphabets and Numerals. In Word Mode, sequences of 30 frames of video are fed into an InceptionV5-LSTM hybrid model for the recognition of both spatial and temporal features of the gesture in order to determine the word they represent. The system has been able to achieve accurate translation of hand gestures to text and show realtime prediction using integrated OpenCV interface. Experimental evaluation shows good performance under different lighting and background, which can stably keep high detection accuracy, as well as maintain not too low frame rate, ideal for real-time application.

Aluminium and its alloys have been garnering significant attention for a long time. These alloys find application in the automobile, marine, and in aviation industries. Al–Si–Mg alloys in particular provide a superb mix of ductility and strength. Multiple experimental investigations have demonstrated that the presence of precipitate phases significantly affects the mechanical strength of the alloys. Accordingly, to identify the cause, molecular dynamic simulations of U2-Al4Mg4Si4 and AlMg4Si3 precipitate phases were performed to investigate the uniaxial compressive behaviour of the Al–Si–Mg ternary phases. The total radial distribution function (RDF) was used to interpret the results obtained from the mechanical tests performed from a microstructural perspective. Further, size analysis of the phases was undertaken to glean into the evolution of the nature of the strength of the precipitate at nanoscale. The Young's modulus of the U2-Al4Mg4Si4 and AlMg4Si3 phases were found to depend on the chemical composition and steadily increased as the size increased. The RDF results reveal a closer atomic packing in the U2-Al4Mg4Si4 phase compared to the AlMg4Si3 phase, indicating that the former phase can withstand better plastic deformation. It was found that increasing the presence of the U2-Al4Mg4Si4 phase softens Al–Si–Mg alloys and improves ductility.