UTAR Institutional Repository (Universiti Tunku Abdul Rahman)
Not a member yet
6132 research outputs found
Sort by
Development of an small electric race car
Get PDFThis project aims to develop an electric racing car from a standard ride-on-car. Electric vehicles (EV) are recognized for the higher efficiency compared to traditional internal combustion engine vehicles. In addition, EV racing can be a good platform to understand everyday science. This can also be an effective method to spark children’s interest in STEM (Science, Technology, Engineering, and Mathematics) education, where the students, children and young learners can gain hands-on experience in problem-solving, engineering skills and creativity when modifying the electric ride-on-car within an affordable budget. Pursuing this project, it first involves researching electric ride-on-car modification techniques, and performance components. Selected ride-on-car is then analysed to determine suitable design changes for an adult to drive. Required parts were fabricated using methods suitable for children to handle and subsequently installed on the car. The car was reinforced to support the weight and handling requirements for adult driving. Additionally, the stock 12V motor was replaced with a 24V motor, different gear ratio and wiring for improving the ride-on-car’s performance. The modified car was then tested for weight distribution, shaft deflection, acceleration, and power consumption. The front wheel camber change due to heavy load is reduced about 75%. The rear shaft deflection was significantly improved 59% which would increase the durability of powertrain system. Performance improved notably, with the top speed increased 260% and the 0-13 m acceleration time dropped from 20.98 s to just 5.91 s. However, the upgraded motor consumed more electrical power which increase the power consumption from 0.0148 Wh/m to 0.1407 Wh/m. In short, this project successfully demonstrated that an electric ride-on car can be transformed into a cost-effective platform for STEM education.
Keywords: STEM awareness, racing, electric vehicle, modifications, energy efficiency
Subject Area: TJ1-1570 Mechanical engineering and machinery, TL1-4050 Motor vehicle
SeniorCare: An elderly assistance mobile application
Get PDFAs Malaysia's population continues to age, there is a growing need for new solutions to aid and assist elderly individuals in their safety, health, and well-being. Existing assistance applications available in the mobile application space, such as Medisafe and Family360, center around an isolation feature and do not deliver a consolidated system addressing the multiple challenges present among the elderly population at large. Thus, this project proposes the creation of a one-stop mobile application specifically targeting elderly individuals to aid and assist in many of its aspects. It would integrate various features, including reminders of medications and appointments, mental stimulation through mini games, an emergency alert system, use of the Global Positioning System (GPS) for location tracking, task assistance services, and remote monitoring and control by family members. The proposed solution aims at promoting the independence of senior citizens and, at the same time, gives peace of mind to their families wherever they may be around the world. The proposed application is developed using Android Studio, Flutter, and the Dart language. The primary application features include the medications and appointment reminder system, the emergency alert system, and the family locator system. The medication and appointment management system helps to keep scheduling, tracking, and getting reminders of medications and appointments at the right time. Thus, they will never miss critical treatments. An emergency alert system allows elderly individuals to quickly deliver emergency signals via diverse methods such as an inactivity timer, hand gesture, or verbal command. Messages will be sent instantly to emergency contacts to allow a quick response in cases of emergency. A family locator system provides position tracking, allowing family members to track the position of their elderly relatives. It also allows geofencing, where alarms will be raised if an individual goes outside or enters a predefined area. The mini game feature allows senior citizens to remain mentally active and moods elevated
Healthbuddy: Your personal health companion
Get PDFHealthBuddy: Your Personal Health Buddy is a health mobile application for smart phones designed to respond to the growing demand for holistic and personalized healthcare management solutions. The project was initiated to address the fragmentation of healthcare monitoring technologies, which inevitably results in wastage and decreasing user engagement. The proposed solution integrates different aspects of health into a single simple-to-use interface. The design features a Health Dashboard for tracking such measurements as weight, height and summary of their health details from other features, Medication and Appointment Reminders, Hydration Checks, and a Nutrition Planner recommending food options by input designated calorie intake. A mental health support chatbot utilizing artificial intelligence is proposed to offer mood tracking and stress management functionality and giving unconditionally support by understanding user’s requirement. Besides, Health and Fitness Challenges will be implemented to encourage user engagement through a reward system, and an Emergency Assistance is coded to provide immediate support in case of serious health events. Currently, the development is in prototyping of the features to determine their feasibility and user flow. By consolidating physical as well as mental well-being support under a single platform, HealthBuddy aims to provide the best possible service among the mobile health (mHealth) technologies
Facial expression recognition for human-computer interaction
Get PDFThis project proposes the development of a robust Facial Expression Recognition (FER) system
integrated within a chatbot framework, aimed at assisting individuals in recognizing and
interpreting their emotional states more effectively. The system employs a deep learning-based
approach, specifically utilizing the EfficientNet architecture, to improve the accuracy and
reliability of emotion detection. It integrates advanced Convolutional Neural Network (CNN)
techniques with preprocessing strategies to address variations in lighting conditions, patient
demographics, and facial occlusions, thereby ensuring consistent performance across diverse
clinical environments. Furthermore, the system delivers real-time feedback and personalized
guidance based on the detected emotions, fostering more empathetic and patient-centered care.
To achieve efficient real-time processing on resource-constrained medical devices, the model
incorporates optimization techniques such as pruning, quantization, and lightweight CNN
architectures. At the end, this project successfully trained a robust Facial Expression
Recognition (FER) model with approximately 73% accuracy, integrated it with a chatbot
system, and deployed the complete solution on a website to enable real-time, emotion-aware
interactions
Parking finder mobile application
Get PDFWith the rise in the number of car owners in fast-growing metropolitan areas, the need for effective parking solutions is becoming more demanding. This project proposes a Parking Finder Mobile Application that will provide real-time information about parking space availability and the parking finding status of vehicles in the parking lot. In this system, computer vision and deep learning models such as YOLOv8 will be utilized for parking spaces and vehicle detection while the DeepSORT algorithm is implemented to track vehicle movement in real-time. The proposed solution tackles the limitations that existing parking systems have including the high cost of implementation and lack of real-time vehicle monitoring. Combining parking space detection with vehicle tracking, the program will shorten the parking search times and improve user experience using a colour-coded status indicator and a simple interface. It is anticipated that such an approach would optimize parking space utilization in cities and promote urban mobility
Automated sign language translation using deep learning
Get PDFThis project focuses on developing a system for automated static gesture sign language translation using deep learning. With the increasing demand for accessible communication tools, particularly for the hearing-impaired community, the need for reliable sign language translation systems is growing. The main challenge addressed in this project is the recognition and translation of static sign language gestures into text, which is less complex than dynamic gestures involving movement. The methodology involves processing images of static sign language gestures using hand landmark detection with MediaPipe. These landmarks are then normalized and input into a deep learning model, trained on processed dataset images, to predict the corresponding sign. The model architecture consists of multiple dense layers with batch normalization and dropout to ensure robust learning. The system is integrated into a user-friendly application that offers real-time sign language translation through a webcam feed, with features such as dynamic confidence threshold adjustment, translation history tracking, and a sign language dictionary. The results show that the system is capable of accurately recognizing and translating static sign language gestures with high confidence, as validated by the test dataset. The system is efficient, easy to use, and highly adaptable for future enhancements. This project demonstrates the potential of deep learning in bridging communication gaps for the hearing-impaired community and sets the groundwork for future work in dynamic sign language translation
Leveraging digital technology for environment, social and governance (ESG) compliance
Get PDFThe construction industry plays a pivotal role in economic growth but is also one of the largest contributors to environmental degradation and social challenges. With rising global pressure to achieve sustainability and meet stricter regulatory demands, Environmental, Social, and Governance (ESG) compliance has become an essential benchmark for responsible construction practices. ESG compliance is increasingly critical in the construction industry,
yet the integration of digital technologies to support ESG initiatives remains fragmented and limited to isolated applications. Thus, this study aims to examine how digital technologies can enhance ESG compliance, the challenges
impeding their adoption, and the strategies to support implementation in Malaysia’s construction sector. A quantitative research approach was employed, with structured questionnaires distributed to construction professionals in Klang Valley, yielding 117 valid responses. The analysis employed Cronbach’s Alpha, the arithmetic mean, the Mann-Whitney U test, the Kruskal-Wallis test, and Spearman’s correlation. Results revealed that digital tools are perceived as highly effective in enhancing transparency, regulatory reporting, and energy efficiency; however, high implementation costs and workforce skill gaps were
identified as critical barriers. Significant differences in perception emerged across professional roles and years of experience, highlighting diverse readiness levels for digital adoption. Correlation analysis further indicated strong linkages between digital upskilling and the successful deployment of ESG-focused technologies. The findings emphasise the need for a unified framework that
integrates digital technologies into ESG practices, supported by industry-wide collaboration, government incentives, and standardized reporting mechanisms. This study contributes to advancing sustainable construction practices by offering practical recommendations for leveraging digital innovation to strengthen ESG compliance, thereby fostering accountability, resilience, and long-term competitiveness in the sector.
Keywords: Environmental, Social, and Governance (ESG); digital technology; construction industry; sustainability; governance; digital transformatio
Bim-enabled collaboration and communication in the construction industry: a comparative study of the practices among conmstruction practitioners
Get PDFBuilding Information Modelling (BIM) has reshaped the global construction industry, with strong adoption in countries like the UK, Germany, and Singapore. In Malaysia, BIM implementation is growing, especially in high value public projects. However, challenges persist, particularly in stakeholder collaboration and communication. This study aims to compare the practices of various construction practitioners in BIM execution, focusing on how they
engage in collaboration and communication. The literature review showed that construction professionals use BIM in different ways, highlighting the critical importance of integrated collaboration and effective communication across
disciplines. This study adopts a pragmatist philosophy, emphasising a mixed method approach through a questionnaire comprising both closed-ended and open-ended sections. Data was collected through an online survey, with a total
of 137 valid responses from architects, engineers, quantity surveyors, and chartered builders across Malaysia. The data were analysed using descriptive statistics and inferential tests, including Cronbach’s alpha, Kruskal-Wallis, and
Spearman’s correlation. Findings reveal that BIM is most used by quantity surveyors and junior-level practitioners. Collaboration is strongly prioritised by engineers and executive-level professionals, while communication is more
common among junior practitioners, especially in design coordination. A moderate-to-strong correlation between collaboration and communication indicates their interdependence. Meetings and discussions were the most
frequent BIM-enabled activities, and higher BIM proficiency corresponded with deeper BIM engagement. In conclusion, the study contributes to understanding how BIM is practised among Malaysian construction practitioners, highlighting
the need for strategies to enhance collaboration and communication. Future research could explore BIM adoption in SMEs, infrastructure projects, and behavioural factors influencing implementation.
Keywords: Building Information Modelling; Collaboration; Communication; Malaysian Construction Industry; BIM Practices
Subject Area: TH1-9745 Building constructio
Disease prediction web application using machine learning
No full textIn recent years, the prevalence of diseases has increased and the demand for quick diagnosis tools is growing. This has highlighted the need for machine learning-based web applications for disease predictions is important in the healthcare system for early diagnosis. This project presents the design and development of a web-based disease prediction application that employs machine learning and natural language processing technologies to assist users in identifying potential health conditions. The motivation for this project is to improving access to early diagnosis, reduce the burden on medical staff and getting general medical advice anywhere and anytime. The methodology involved develop and train machine learning models on Symptom-Disease Prediction Dataset (SDPD) to achieve precise predictions, integrate the model into web application built on Flask and React, and employ Google Gemini to generate general medical recommendations and extract symptoms. System testing was conducted through multiple testing methods, including unit testing, integration testing, user acceptance testing (UAT) and user interface design feedback collected through Google Forms. The results indicate that the machine learning model achieved a prediction accuracy at approximately 97%. User acceptance testing validated that over 90% of users rated the usability and ease of use of the system at 4 or higher on a 5-point Likert scale. The study concluded that the system successfully achieved its objectives, delivering a practical, user-friendly, and intelligent healthcare support system. However, it also acknowledged limitations such as dependence on dataset quality, lack of coverage for rare or new diseases, and multilingual support. Future work will focus on expanding dataset variety, integrating multilingual support, and incorporating of contextual health data to further enhance prediction accuracy and precision.
Keywords: Disease prediction, Machine Learning, Web Application, Large Language model, Natural Language Processing
Subject Area: QA76 – Computer Scienc
Web based smart iot-based system for optimized Japanese melon farming: data-driven approach to enhance yield and quality
No full textThis project presents the design and implementation of a web-based smart IoT system for Japanese melon cultivation, addressing the critical need for real-time monitoring, actionable analytics, and decision support in high-value crop farming. The system integrates IoT sensors to capture environmental parameters such as soil moisture, pH, electrical conductivity, temperature, and light intensity, with data first ingested via ThingSpeak and subsequently synchronized into a Supabase PostgreSQL database through an automated Edge Function and Cron Job. The application layer, developed using Spring Boot, manages business logic including threshold-based rule evaluation and integrates with Firebase Cloud Messaging to deliver real-time alerts and recommendations. Angular, Ng Zorro, TailwindCSS, and embedded Grafana dashboards form the presentation layer, providing farmers with intuitive visualizations such as time-series graphs, Soil Health Index computation, and correlation heatmaps. System testing and evaluation demonstrated reliable data integrity (99.81% completeness), accurate threshold-based suggestions, and efficient performance with an average application start time of 1.55 seconds. Functional and integration test cases confirmed robust user management, sensor threshold configuration, and task scheduling features. The findings highlight that the system effectively transforms raw IoT data into interpretable insights, enabling timely interventions that improve yield consistency and fruit quality. While the study faced limitations in full-scale deployment and hardware connectivity, the outcomes establish a scalable, cost-effective foundation for precision agriculture. Future work is recommended to expand deployment across full cultivation cycles, incorporate predictive analytics, and integrate advanced automation for irrigation and ventilation control.
Keywords: smart farming; IoT; Japanese melon; Supabase; Grafana; Firebase; soil health index
Subject Area: T57.6–57.9