Hou Kit Mun

Electronic waste (e-waste) continues to increase globally, yet conventional bins cannot distinguish hazardous batteries and devices from recyclable metals. This article presents an AI- and IoT-enabled smart dustbin that automatically identifies and segregates general waste, metals, and electronic or battery-based hazards while providing real-time monitoring through a cloud-based dashboard. The system integrates inductive sensing, Time-of-Flight detection, an Espressif Systems Platform 32 (ESP32)-CAM module, and Google Gemini 1.5 Flash for image classification. The prototype achieved a waste segregation accuracy of 93.5% with a total cycle time of 4–6 s per item. The touch-free lid, swift mechanical actuation, and compact 59 × 59 × 100 cm footprint make the dustbin suitable for deployment in campuses, offices, and shopping malls. Dual ESP32 controllers, cloud connectivity through Message Queuing Telemetry Transport (MQTT), Firebase, and a Streamlit web interface enable automated alerts through Discord and email, demonstrating a scalable and energy-efficient approach to sustainable e-waste management.

Diapers are essential for managing incontinence in infants, but delayed diaper changes can lead to discomfort, skin rashes, and infections. Previous research has focused on ultra-high frequency (UHF) radio-frequency identification (RFID) moisture sensors with limited read ranges and binary (dry/wet) detection capabilities. This study proposes a textile-based UHF RFID bowtie antenna using high conductive material as a passive sensor for diaper moisture sensing. The purpose of the sensor is to detect and measure moisture levels in infant diapers, providing immediate notifications to parents and caregivers. The sensor, measuring 78 mm × 32 mm × 0.09 mm, offers multilevel wetness detection (0–60 ml) and achieves a maximum read distance of 12.1 m (dry) and 3.3 m (wet) within the US UHF RFID passband (902–928 MHz), which, to the best of our knowledge, represents the highest reported read range among passive UHF RFID diaper moisture sensors in the scientific literature. Additionally, the sensor is cost-effective, reusable, flexible, and can seamlessly attach to reusable diapers, making it an ideal solution for practical and sustainable diaper monitoring applications.

In real-life applications, electroencephalogram (EEG) signals for mental stress recognition require a conventional wearable device. This, in turn, requires an efficient number of EEG channels and an optimal feature set. This study aims to identify an optimal feature subset that can discriminate mental stress states while enhancing the overall classification performance. We extracted multi-domain features within the time domain, frequency domain, time-frequency domain, and network connectivity features to form a prominent feature vector space for stress. We then proposed a hybrid feature selection (FS) method using minimum redundancy maximum relevance with particle swarm optimization and support vector machines (mRMR-PSO-SVM) to select the optimal feature subset. The performance of the proposed method is evaluated and verified using four datasets, namely EDMSS, DEAP, SEED, and EDPMSC. To further consolidate, the effectiveness of the proposed method is compared with that of the state-of-the-art metaheuristic methods. The proposed model significantly reduced the features vector space by an average of 70% compared with the state-of-the-art methods while significantly increasing overall detection performance.

Mental stress state recognition using electroencephalogram (EEG) signals for real-life applications needs a conventional wearable device. This requires an efficient number of EEG channels and an optimal feature set. The main objective of the study is to identify an optimal feature subset that can best discriminate mental stress states while enhancing the overall performance. Thus, multi-domain feature extraction methods were employed, namely, time domain, frequency domain, time-frequency domain, and network connectivity features, to form a large feature vector space. To avoid the computational complexity of high dimensional space, a hybrid feature selection (FS) method of minimum Redundancy Maximum Relevance with Particle Swarm Optimization and Support Vector Machine (mRMR-PSO-SVM) is proposed to remove noise, redundant, and irrelevant features and keep the optimal feature subset. The performance of the proposed method is evaluated and verified using four datasets, namely EDMSS, DEAP, SEED, and EDPMSC. To further consolidate, the effectiveness of the proposed method is compared with that of the state-of-the-art heuristic methods. The proposed model has significantly reduced the features vector space by an average of 70% in comparison to the state-of-the-art methods while significantly increasing overall detection performance.

Exposure to mental stress for long period leads to serious accidents and health problems. To avoid negative consequences on health and safety, it is very important to detect mental stress at its early stages, i.e., when it is still limited to acute or episodic stress. In this study, we developed an experimental protocol to induce two different levels of stress by utilizing a mental arithmetic task with time pressure and negative feedback as the stressors. We assessed the levels of stress on 22 healthy subjects using frontal electroencephalogram (EEG) signals, salivary alpha-amylase level (AAL), and multiple machine learning (ML) classifiers. The EEG signals were analyzed using a fusion of functional connectivity networks estimated by the Phase Locking Value (PLV) and temporal and spectral domain features. A total of 210 different features were extracted from all domains. Only the optimum multi-domain features were used for classification. We then quantified stress levels using statistical analysis and seven ML classifiers. Our result showed that the AAL level was significantly increased (p < 0.01) under stress condition in all subjects. Likewise, the functional connectivity network demonstrated a significant decrease under stress, p < 0.05. Moreover, we achieved the highest stress classification accuracy of 93.2% using the Support Vector Machine (SVM) classifier. Other classifiers produced relatively similar results.

In conjunction with the 4th Industrial Revolution, many industries are implementing systems to collect data on energy consumption to be able to make informed decision on scheduling processes and manufacturing in factories. Companies can now use this historical data to forecast the expected energy consumption for cost management. This research proposes the use of a Temporal Convolutional Neural Network (TCN) with dilated causal convolutional layers to perform forecasting instead of conventional Long-Short Term Memory (LSTM) or Recurrent Neural Networks (RNN) as TCN exhibit lower memory and computational requirements. This approach is also chosen due to traditional regressive methods such as Autoregressive Integrated Moving Average (ARIMA) fails to capture non-linear patterns and features for multi-step time series data. In this research paper, the electrical energy consumption of a factory will be forecasted by implementing a TCN to extract the features and to capture the complex patterns in time series data such daily electrical energy consumption with a limited dataset. The neural network will be built using Keras and TensorFlow libraries in Python. The energy consumption data as training data will be provided by GoAutomate Sdn Bhd. Then, the historical data of economic factors and indexes such as the KLCI will be included alongside the consumption data for neural network training to determine the effects of the economy on industrial energy consumption. The forecasted results with and without the economic data will then be compared and evaluated using Weighted Average Percentage Error (WAPE) and Mean Absolute Percentage Error (MAPE) metrics. The parameters for the neural network will then be evaluated and fined tuned accordingly based on the accuracy and error metrics. This research is able create a CNN to forecast electrical energy consumption with WAPE = 0.083 & MAPE = 0.092, of a factory one (1) week ahead with a small scale dataset with only 427 data points, and has determined that the effects of economic index such as the Bursa Malaysia has no meaningful impact on industrial energy consumption that can be then applied to the forecasting of energy consumption of the factory.

Machine health monitoring is the main focal point for now as many industries are evolving to industry 4.0. Industry 4.0 is the revolution in industrial that involve the Internet of Things (IoT) and artificial intelligence toward automation and data sharing for production efficiency improvement. The existing established methods for machine health monitoring were not in real-time and there was no real-time correction of data from the load and processing of data on the computer. In tracking machine health efficiency this approach wasn’t very successful. Real-time machine health monitoring can improve overall equipment effectiveness (OEE), reduce electricity consumption, minimize unplanned downtime, and extend machine lifetime. In this research paper, we propose to design a real-time machine health monitoring system using machine learning with IoT technology that can analyze the supply balancing condition on a 3-phase system. This system is built with compact physical hardware and can capture the electrical data from the load then send it to the server. The server will progress data and train the data using machine learning. The system was installed on a blender machine in a factory. In this research, a system which is able to monitor the machine operation and classify the operation stages of the machine was developed. Besides that, the system also capable to monitor the load balancing condition of the machine.

Automation technologies are often impacted undesirably by downtime caused by machine failures. An approach to overcome machine failures is through failures prediction, achievable via the monitoring of power consumption. Unusual power consumption could serve as an alert for possible machines faults. Real‐time power consumption monitoring is made possible with the deployment of Internet of Things (IoT), which brings together digital cloud computing and conventional automation technology to rapidly increase the efficiency of production. In this work, the authors have developed an IoT‐based power monitoring system to detect unusual power consumption of machines and present the data to users remotely through Android application and cloud storage. The competent integration of Wi‐Fi module ESP8266, CT sensor module and field‐programmable gate array DE1‐SoC board allows the system to accurately measure power consumption and instantaneously store the data in cloud storage. Performance of the developed system was characterised by current measurement of five different loads. The absolute errors, given by the difference between current measured by digital multimeter and the developed system, range from 0.01 to 0.26 A. The close agreement between the two measurements endorses the accuracy and reliability of the real‐time power monitoring system, on top of its easy‐accessibility gifted by IoT.