Intelligent Manufacturing and Mechatronics

COVID-19 caused a pandemic outbreak, resulting in many deaths and severe economic damage since 2019. Hence, the diagnosis of COVID-19 has become one of the major fields of research. Although RT-PCR has excellent reliability and precision, it is time-consuming and laborious. Therefore, the chest X-ray was used as an alternative and reliable diagnostic tool for COVID-19. However, it requires a radiologist to analyze the X-ray images, which is limited by the availability of experts and time. Henceforth, many researchers deployed automated computer-aided diagnosis with deep learning neural networks to speed up the diagnosis of COVID-19 with high accuracy and reproducibility. This study applied six state-of-art convolutional neural networks (DenseNet201, MobileNetV2, ResNet101V2, VGG16, InceptionNetV3, and Xception) with transfer learning. An integrated stacking ensemble method was used to concatenate DenseNet201, MobileNetV2, VGG16, and Xception to produce a robust and accurate diagnostic model for COVID-19. The proposed ensembled CNN model in this study produced a test accuracy of 0.9725, sensitivity of 0.9749, and F1-score of 0.9724.

The perception system is one of the important components of autonomous vehicles, as it provides the information that is required by vehicle control to make decisions on the manoeuvre of the vehicle. The study focuses on the development of target prediction using sensor fusion algorithm for Level 3 autonomous vehicle in Malaysian environment. The sensor fusion algorithm was developed to unify the data from the sensors and obtain useful information, where the closest object around the ego vehicle was determined in the project. In order to display the closest object around the ego vehicle, the relative distances of the objects were calculated. The closest object among the cameras, the closest object in each camera and warning for nearby object were displayed on the output images. To study the performance of sensor fusion algorithm developed in Malaysian traffic, the virtual environment model of MyAV Route A was developed by using RoadRunner. There were two cases developed to observe how would the algorithm perform. The first test case was on target prediction using sensor fusion algorithm on flat road, while the second test case was on target prediction along a slope. It was shown that the algorithm performed well from the two test cases, as the vehicles and pedestrians were detected and displayed successfully with confidence score of above 0.72 and 0.87, respectively, even with views from different angles and locations of cameras.

Damages such as a crack location and length in a vibrating component might cause catastrophic failures. The existence of cracks alters the physical properties of a structure, which alters its dynamic response characteristics. As a result, there is a need to comprehend the dynamics of cracked structures. Natural frequency analysis on railway rails has recently gained popularity as a method for identifying vibrational modes. However, studies on the impact of crack length, depth, and position on railway tracks have not yet been fully understood. This paper aims to investigate the effect of crack length, depth, and location on natural frequencies of railway track. For the model simulation, the most commonly used parameters adopted by the Malaysian Railways track is the UIC60 type of rail profile cross section has been selected for analysis. Free vibration analysis was developed using Finite Element Analysis (ANSYS) to evaluate the effect of various crack locations for 45 and 50 mm crack length on natural frequencies of railway track. To establish the precise finite element model for free vibration analysis of railway track, convergence analysis and numerical verification were conducted. The present numerical simulation results were in good agreement with experimental modal results. The findings demonstrated that mode shapes of vibrations were slightly different when changing the location of crack with crack length had been designed 45 and 50 mm. In general, this study has made important contributions to understanding the effect of crack length, depth, and location on natural frequencies of railway track.

Missing data poses a significant challenge in extensive datasets, particularly those containing time-series information, leading to potential inaccuracies in data analysis and machine learning model development. To address the issue, this paper compared and evaluated four imputation methods: MissForest, MICE, Simplefill, and Softimpute which utilized Random Forest Algorithm. The research examines the impact of missing ratios and temporal variations on the performance of the imputation methods. The results indicated that MissForest consistently outperformed other methods, exhibiting the lowest RMSE values and a high coefficient of determination (R2), indicating its accuracy and ability to explain the variation in the data. Furthermore, graphical analyses demonstrated the stability of MissForest over time, while MICE and Simplefill showed higher sensitivity to date changes. Softimpute demonstrated relative consistency but slightly lower performance compared to MissForest. Overall, this study highlights the effectiveness of MissForest as the preferred imputation method for AVL time-series data.

This paper digs into robots’ revolutionary role in the industrial landscape, highlighting present uses and future trends while addressing ongoing problems. It investigates how machine learning is altering industrial processes, increasing efficiency and production while simultaneously highlighting the challenges of data needs and model interpretability. The evaluation investigates the promise of blockchain technology in enhancing industrial security and transparency, while also recognizing the hazards of possible attacks and smart contract vulnerabilities. The transformational influence of additive manufacturing, particularly 3D printing, is discussed, as well as the constraints connected with printing speed, product quality, and material availability. The study emphasizes the potential of new materials such as bio-based polymers and 2D heterostructures in the advancement of robotic systems. Despite these encouraging achievements, the assessment finds gaps in existing research and suggests future strategies for maximizing the potential of these technologies in the industrial industry.

This review offers an in-depth review of current developments in patient monitoring technologies and prosthetic robotic arms, with a focus on their application for children with disabilities. These prosthetic arms’ design, development, and testing—which aspire to imitate the functions of human arms—are thoroughly explored. The paper also examines the application of virtual reality in user training and the significance of performance assessment in enhancing the functioning and design of the prosthetic. Additionally, numerous case studies are used to illustrate the various ways that robotic arms are used in industrial and rehabilitation contexts. It is emphasized as a potential way to raise the standard of care for kids with disabilities: the integration of patient monitoring systems and prosthetic robotic arms. The review attempts to highlight topics that need further research and lay a platform for future studies in this field.

This review paper delves into the advancements brought about by Industry 4.0 in the realm of intelligent machining systems for robotic end-effectors. Robotic end-effectors, which are the devices at the end of a robotic arm, have seen significant enhancements in their design, development, and application across various sectors, from manufacturing to healthcare. The integration of intelligent machining systems into these end-effectors has augmented their efficiency, precision, and flexibility. The paper also highlights the role of intelligent control systems in boosting the performance of these robotic systems. Despite the progress, challenges persist, such as improving machining accuracy, optimizing machining trajectories, and integrating machine learning techniques. The review concludes by identifying gaps in the current research and suggests potential areas for future exploration to further enhance the capabilities of robotic end-effectors.

This review examines the utilization of Artificial Neural Networks (ANNs) in the analysis of road crash data and their prospective contribution to autonomous vehicles. Artificial neural networks (ANNs) have exhibited considerable promise in the realm of modeling and forecasting crash occurrences, thereby offering valuable insights for enhancing road safety. Artificial neural networks (ANNs) play a crucial role in the functionality of autonomous vehicles by enabling them to effectively perceive their surroundings, make informed decisions, and operate with a high level of safety and efficiency. Nevertheless, the deployment of Artificial Neural Networks (ANNs) encounters various obstacles, such as their inherent opaqueness, the necessity for substantial quantities of meticulously curated data, and the demanding computational capabilities they demand. Possible solutions to these challenges encompass the advancement of methodologies for interpreting artificial neural networks (ANNs) and the utilization of more intricate ANN models. Notwithstanding the advancements achieved in this particular domain, it is imperative to acknowledge and rectify the existing deficiencies in the present body of research. These encompass the necessity for conducting more extensive research on the utilization of Artificial Neural Networks (ANNs) in the analysis of road crashes, the requirement for developing more resilient testing methodologies for these systems, and the demand for further investigation into the effective implementation of ANNs in autonomous vehicles. This review makes a significant contribution to the current academic conversation in the field, offering valuable insights for researchers, policymakers, and practitioners engaged in the domains of road safety and autonomous vehicles.

The role of robotics in precision machining and manufacturing has undergone significant evolution over time, as technological advancements have empowered robots to execute a diverse array of tasks with remarkable accuracy and uniformity. This comprehensive review provides an overview of the historical progression of robotics in the specified field, evaluates its present status, analyzes practical case studies of its applications, and investigates potential future avenues for development. Despite the notable progressions in the domain of robotics, there persist certain deficiencies and inadequacies. Several factors need to be considered in relation to the adoption of robotics technology, namely the significant financial investment required, the potential for job displacement resulting from automation, and the necessity of skilled personnel to effectively operate and maintain these machines. In order to bridge these gaps, it is imperative to undertake more research and development endeavors. Future research efforts may focus on developing economically feasible robotics solutions, specifically designed to meet the requirements of small and medium-sized organizations. Moreover, it is crucial to investigate approaches aimed at mitigating the adverse consequences of employment relocation. Furthermore, it is imperative to emphasize the need of establishing training programs that focus on providing workers with the necessary skills and knowledge to proficiently operate and maintain robots systems. This paper provides a thorough analysis of the application of robots in precision machining and manufacturing, with a focus on the potential of robotics to improve efficiency, accuracy, and flexibility in manufacturing operations.

Nanomaterials in robotics have transformed precision medicine, photothermal therapy, imaging, and medical materials and devices. This review covers nanomaterials’ current and future robotics applications. It also addresses nanomaterials in robotics research challenges and limitations. Nanotechnology is used to develop tiny robotic agents for surgery, therapy, imaging, and diagnosis. It also examines nanotechnology's role in metastatic cancer treatment and functional food development. This field has made progress, but research is still lacking. These include the lack of comprehensive studies on nanoparticles’ environmental and health effects, the lack of effective methods for regulating nanoparticle release, and the need for a regulatory framework that takes nanomaterials’ unique characteristics and potential hazards into account. The review identifies and examines literature gaps and suggests future research to fill them. Recognizing and addressing these issues can help nanomaterials and robotics reach their full potential.

Ensuring smooth machine operation and safety is crucial in most engineering plant, and fault diagnosis plays a critical role in achieving these goals. In recent years, machine learning techniques already being utilized extensively in the research of bearing fault diagnosis. One of the recent methods that has gained popularity is the extreme learning machine (ELM) method. This method offers several advantages, such as fast learning rate, generalization performance efficient, and easy to use. However, it is important to note that the ELM method can result in inaccurate diagnosis, if the values for input weight, hidden layer bias, and number of neurons are not selected properly. This paper introduces a new approach for bearing fault diagnosis, named GMO-ELM, which utilizes the ELM method and the geometric mean optimizer (GMO) to optimize ELM parameters. The proposed method was tested using sets of bearing vibration signal from Case Western Reserve University (CWRU) with four different operating conditions, including healthy baseline, outer race fault signal, inner race fault signal, and ball fault signal. Based on the result, the proposed method is able to provide 12% better performance by comparing to the conventional ELM and competitive diagnosis performance by comparing to other recent diagnosis model.

Remanufacturing of automotive components involves restoring cores from end-of-life vehicles (ELV) to a condition where the component can be used again. Implementing image processing technique in the visual inspection stage is used to reduce cost and time while increasing accuracy. This research aims to perform real-time surface visual defect detection that is simpler and less costly which commonly uses machine learning and/or deep learning techniques. The methods used in this research are a combination of image processing and computer vision techniques to great effect. The techniques used are a pixel-based contour detection and image subtraction. For this research, connecting will be the sample component to test the algorithm. This approach will take less time to train since it compares the non-defective connecting rod image with the defective connecting rod. The defects that are tested in this research are cracks and buckles, which commonly occur in connecting rod failure due to high load and stress. There are different settings of threshold and illumination that were tested, such as different thresholds of 50, 60, and 70, also illumination colour temperatures such as white, natural, and warm. After 270 trials with different settings and defects, the proposed algorithm achieved a 93% accuracy using a 70 per cent threshold and white light (3000–3500 K). The findings suggest that the implementation of a pixel-based contour detection through image subtraction holds promise as a cost-effective alternative to utilizing machine learning and/or deep learning techniques.

To conduct everyday tasks safely and effectively, it is essential to understand spinal kinematics. Even while there have been various attempts to look at the kinematics of the area during daily activities, very little study has specifically focused on the multi-segmental contribution of the thoracic spine. The credibility of actual multi-segment contributions is diminished since the thoracic area has only been investigated as a single segment. This study aims to evaluate the significance of the thoracic spinal segments during various pick-object approaches among healthy individuals. All tasks examined had interclass correlation coefficients (ICC) greater than or equal to 0.97 (0.958–0.996), which indicates a high level of reliability. There were statistically significant variations in the majority of the task completion percentages between tasks with and without a global section (P-value = 0.05). At the middle transition for every segment of the Begin cycle of pick-object, the thoracic spine reaches its maximum mobility. Picking up an object while squat had the least increase in thoracic kinematics as compared to Semi-squat and Stoop indicating the optimal postural implementation for minimizing spinal injury. The results give a clear explanation of the good spinal condition of asymptomatic people and might be helpful for ergonomic spinal rehabilitation.

The use of flexible beam structures is widespread across various industries due to their numerous advantages, including reduced energy consumption, cost-effectiveness, faster movements, and improved efficiency when compared to rigid beam structures. This research focuses on employing a proportional-integral-derivative (PID) controller to regulate the vibration level of a flexible beam structure. The tuning of this PID controller is crucial to maximize its performance. Initially, the PID controller is heuristically tuned, successfully suppressing undesired vibrations in the flexible beam structure. Subsequently, an evolutionary algorithm, specifically the Firefly Algorithm (FA) and its advanced version, the Advanced Firefly Algorithm (AFA), is employed to optimize the PID controller. The findings demonstrate that the PID controller optimized through the evolutionary algorithm exhibits superior performance in vibration suppression compared to the heuristically tuned PID controller. Moreover, the comparison between FA and AFA reveals that AFA outperforms FA in enhancing the performance of the PID controller. The PID-FA achieved a reduction percentage of up to 96.7%, while the PID-AFA achieved a reduction percentage of up to 98.8%.

Bearing fault diagnosis techniques have been shifted toward using the deep learning (DL) model due to its ability to process the raw vibration signal. However, most of the deep learning models use high-speed datasets for bearing fault diagnosis; thus, the performance of the DL model on low operational speed datasets still needs to be solved. Therefore, this research proposed the integration method of particle swarm optimisation (PSO), stacked sparse autoencoder (SSAE), and empirical mode decomposition (EMD) for diagnosing the bearing fault at three-speed conditions (60, 780, and 1800 rpm) from three different experiment dataset including INV, Mafaulda, and CWRU datasets. The PSO-SSAE models produced 100% accuracy at 1800 rpm without any signal processing method involved. Meanwhile, the PSO-SSAEs’ performance drops 64 and 70% on 780 and 60 rpm datasets, respectively. The EMD method is used to preprocess these signals, and the prominence peak selection method is proposed for IMFs selection. Hence, the performance of PSO-SSAE models with EMD increased to 97% for 780 and 60 rpm datasets.

Much research has been done to investigate the most ideal grouser parameters to improve a robot’s wheel mobility in sandy terrain. However, their area of research mainly focused on the normal wheel with fixed grousers. This paper showcases the study of performance of the assistive grouser wheel moving on a flat surface sandy terrain attached with different grouser width parameters. Two different grouser shapes were fabricated which consist of C-Shape and Wider C-Shape, with two configurations forward facing and inverse. For each grouser set, one of the grousers was attached with 2 load cell sensors used to measure force. The grousers were tested across a total of 3 runs to identify its total average traction force acting on the grouser during the wheel’s movement. The results show increased maximum positive and negative forces when the grouser width is increased, which resulted in less net traction force. The inverse configuration also showed better results than forward facing. This shows that for assistive mechanism, increasing the width of the grouser may not increase the effectiveness of the grouser in generating traction force.

This paper presents an automated harmonic removal technique as an efficient method for identifying and removing the influence of harmonics from the output signal. The method involves disregarding user-defined parameters during system initialization and reconstructing the output signal automatically so that it can be used for system identification. While stochastic subspace-based algorithms (SSI) are generally reliable for modal parameter estimation, applying them to structures with rotating machinery and harmonic excitations presents challenges. Because the SSI method necessitates designating parameters, such as the maximal within-cluster distance, at the outset of each dataset analysis, the issue remains unresolved. In addition to modal identification, the current research concentrates on image-based feature extraction for aggregating and classifying harmonic components and structural poles directly from a stabilization diagram. Utilizing online data sets to validate the algorithm’s efficacy. Using a comparative analysis, the proposed method is compared to existing techniques, namely orthogonal projection-based harmonic signal removal and smoothing techniques based on linear interpolation. The results indicate that the proposed algorithm estimates modal parameters precisely and consistently both before and after the removal of harmonic components from the response signal.

There are various stages in human mental development. Among them are consciousness, drowsiness, and light sleep. These human mental stages and conditions can be affected by human emotions (Ali et al. in Wirel Pers Commun 125:3699–3713, 2022; Katmah et al. in Sensors 21(15):5043). Hence, human brainwaves or electroencephalogram (EEG) signals can be employed to analyze and interpret the development of human mental stage. In this research, 1-channel EEG device is employed to measure neural electrical activity from five people as they are engaged in three different cognitive exercises such as playing a video game, reading a book, and watching a movie. EEG signals are analyzed in LabVIEW software to reveal the unique features which are able to describe various human stages. The EEG power spectrum in terms of mean and standard deviation for each EEG frequency band (theta band, alpha band, and beta band) is computed. Then, the k-nearest neighbor (k-NN) classifier is employed to discover the best feature that is capable to indicate status of human mental stage. The findings of the study demonstrated that the mean EEG feature with the training and testing ratio of k-NN classifier at 80:20 could detect and categorize human stages with the classification accuracy of 81.57%. Meanwhile, LabVIEW graphical user interface (GUI) and block diagram are constructed to display the analyses of human stages of each subject for the specified human stage activities. In addition, a device is built to indicate human mental stage in an off-line manner.

Bearing is a critical component in an electrical machine which get continuous monitoring and included in scheduled predictive maintenance. The temperature of the bearing is a valuable information that may allow early fault detection, lubrication assessment, and overloading indication of the system driven. Using the temperature measurement of the bearing and comparing it to a baseline temperature in real time will allow early warning of any eventual fault. This paper proposes a thermal model for the bearing in a brushed DC machine, developed using transfer function that will predict the temperature increase contributed specifically by speed variation. The transfer function was found by identification using experimental temperature of the bearing at a speed ranging from 20 to 100% of its rated speed while being at no load. The result shows that the first-order transfer function was found to be the best with a model identification MSE of less than 0.23. The slight variation on the poles of the system indicates that the thermal system of the bearing inside an electrical machine does not obey exactly the LTI hypothesis.

In recent times, diverse machine learning models have been employed in this field of technology. Nevertheless, the implementation of learning models for image classification remains uncertain and has proven to be challenging. The utilization of transfer learning (TL) has been showcased as a potent technique for extracting crucial features and can significantly reduce training time. Moreover, the feature extractor model has demonstrated excellent performance in the TL method across numerous applications. As of now, there has been no evaluation of using these methods for ADHD classification through functional magnetic resonance imaging (fMRI) applications. The objective of this study is to identify an appropriate pipeline consisting of transfer learning and conventional classifiers for effectively discriminating between individuals with ADHD and those without. For feature extraction, InceptionV3, VGG16, and VGG19 models were employed, which were subsequently combined with either k-nearest neighbor (k-NN) or support vector machine (SVM) classifiers. A dataset consisting of 556 images was collected from the ADHD-200 competition dataset. The data were divided into an 80:20 ratio, with 80% used for training and 20% for testing. The hyperparameters of both k-NN and SVM were optimized using the grid search method. The experimental results revealed that the optimal pipelines were achieved using InceptionV3 coupled with k-NN classifier, where the best parameters were determined as the Minkowski distance metric and a k-value of 1. The pipeline demonstrated a macro-average classification accuracy of 1.00 for the training set and 0.95 for the test set. In summary, the results demonstrate that TL models have successfully exhibited the capability to differentiate fMRI images for ADHD classification.

In rehabilitation and medical offices, mechanomyography (MMG) is a noninvasive, painless technology that can be applied for a number of goals. The goal of this study is to present a thorough overview of recent studies on mechanomyography-based human limb rehabilitation. The present study illuminates the utilization of distinct transducers, including accelerometers, piezoelectric contact sensors, and condenser microphone sensors. Furthermore, it underscores the diverse results that these investigations have yielded. The main findings of this review, which apply to all of these forms of mechanomyography sensors, are that the ratio of sensor mass to muscle mass under observation is the most crucial factor in sensor selection. Therefore, it is believed that accelerometers are the most trustworthy devices for spotting MMG signals during both voluntary and induced muscular contractions.

Exposure of continuous vibrations toward the human body from a moving vehicle could reduce human comfort, provoke motion sickness, and affect a human’s health directly. The effect will be catastrophic for future autonomous vehicle implementation if the effect is not widely studied since the role of driver will be handled by the computer. In this study, Smart Campus Autonomous Vehicle (SCAV) simulation platform is coupled with a pregnant women biodynamic model to investigate the human body dynamic response to induced vibrations and assess comfort and motion sickness. The combined models are used to investigate the impact on the occupant’s head vertical accelerations from the accelerations induced by the vehicle movement. Real simulation platform by using Smart Campus Autonomous Vehicle (SCAV) is used to obtain vehicle acceleration data that is used as an input for pregnant human biodynamic model. From this combination, the vibrational effect on the human head can be obtained depending on the vehicle movement. Finally, the responses of head acceleration is obtained, and comfort and motion sickness incidence are assessed by using relevant models mentioned in the literature.

3D LiDAR-based object detection during autonomous vehicle navigation is a trending field in autonomous vehicle research and development. As 3D LiDAR is resistant to light interference while capable of capturing detailed 3D spatial structures of the detected objects, it is the main perception sensor for autonomous vehicles. With its improved accessibility in the recent years, the advent of deep learning had allowed feature learning from sparse 3D point clouds. Hence, this leads a plethora of methods in object detection for 3D sparse point clouds. In this research, an extensive experiment was conducted using various 3D LiDAR object detections for various forms like pillar-form, point-form and voxel-form onto multiple point cloud data sets captured using Robotic Operating System (ROS). Based on experiments conducted, pillar-form point cloud data is suitable for dense point clouds, while voxel-form is optimal for both indoors and outdoors environment.

The flexible structure offers various advantages such as being lightweight, efficient, fast system response, and low energy consumption. However, the light characteristic of the structure leads to excessive vibration, which can lead to system failure. Thus, eliminating vibration from external sources on the horizontal flexible plate structure is compulsory to preserve the performance and lengthen the system's life. To overcome these drawbacks, this project aims to develop an intelligent vibration controller based on bird mating optimizer into a PID controller for vibration cancelation purposes. Bird mating optimizer is known as a new metaheuristic algorithm that was initially proposed to solve ongoing optimization problems with an auspicious performance inspired by the intelligent mating behavior of birds. This algorithm is aimed to find an optimal value for PID controller parameters. The controller is developed using active vibration control technique in simulation environment, and the performance of the proposed controller is compared with classic Ziegler–Nichols tuning approach. It is indicated that, the PID controller tuned bird mating optimizer can outperform the PID tuned by Ziegler–Nichols by achieving the highest attenuation at the first mode of vibration with 27.20 dB attenuation using single sinusoidal disturbance, equivalent to 26.29% of vibration reduction.

This is a comparative study of Teaching Learning-Based Optimization (TLBO) as a human-based algorithm against other types of metaheuristic algorithm: single-agent finite impulse response optimizer (SAFIRO), simulated Kalman filter (SKF), particle swarm optimization algorithm (PSO), black hole algorithm (BH), and genetic algorithm (GA), in solving CEC2014 test suite. The TLBO algorithm is inspired by the process of teaching and learning in a classroom. The advantages of TLBO are it only has two main tasks: teaching phase and learning phase and has no parameter setting. The TLBO performance provides a balance between exploration and exploitation. Statistical analysis is then carried out to rank the TLBO results to those obtained by other type of metaheuristic algorithm. The experimental result show that the TLBO algorithm is a promising approach and comparative to SAFIRO and SKF and has better than PSO, BH, and GA.

Cars, especially N1 Light Commercials, lack crumple zones. Poor design and rigid structure would not protect automobile and passengers. A vehicle frontal protection device/bullbar solves this issue. Solidworks developed three concepts from the existing design. Choice matrix selected conceptual design 3. A decision matrix was used to choose the optimum material for the design from steel, aluminium, and carbon fibre epoxy. ANSYS Explicit Dynamics did a full-frontal collision finite element research on conceptual design 3 and the outgoing design using 304 stainless steel and carbon fibre epoxy. Simulated total deformation, equivalent stress, kinetic energy, and internal energy. Carbon fibre epoxy and better design reduced point deformation. 304 stainless steel and carbon fibre epoxy had maximum deformations of 0.0081932 and 0.011181 m in the old design and 0.0095112 and 0.010074 m in the new design. The previous design with 304 stainless steel and carbon fibre epoxy had stress values of 1.496e+9 and 1.7807e+9 Pa, whilst the current design had 1.3063e+9 and 7.3077e+8 Pa. The new carbon fibre epoxy design provided the lowest maximum stress and fastest energy absorption. 304 stainless steel and carbon fibre epoxy had 5068.8 and 1077.1 J kinetic energy, whilst the new design had 12,677 and 2693.7 J.

Car assembly is assembling the components to become a complete car. The car assembly process consists of three lines: trimming, chassis and final. Assembling a car seat is one of the activities with a heavy workload by using several body parts in the work found in the final line. The work process is done manually and repetitively, such as taking, pushing, pulling and assembling the seat into the car. These activities can cause pain and injury in several parts of the worker’s body, called MSD. This injury will have an impact on the unevenness of the assembly process in achieving the production target. There are two seat assembly workers. This research aims to identify employee complaints to determine the level of MSD risk and the body parts of the employee that dominate feeling very sick by applying the NBM and WERA methods. The application of NBM gives the results of body parts that dominate feeling very painful during the work process which consists of the lower neck, back, waist, bottom, lower arm and ankle, which are included in the “Medium” risk level with each total score of 24 operators 1 and 27 operators 2. WERA gave the results of four activities included in the “medium” risk level with a total score of > 30, and one activity was in the “low” risk level with a total score of 25; hence, further review and action are needed. Corrective action can be given as an ergonomic design of automatic aids that can move independently without human assistance to facilitate the work process.

Injection molding is common manufacturing process that produces thermoplastic products by injecting molten plastic into a mold that emulates the product’s geometry. The product is ejected from the mold by ejector once it has solidified. This paper presents the study of thermoplastic injection molding process simulation and mold design of electrical appliance. The objective of this study is to reveal the important steps for plastic product development. In addition, it presents the requirement analysis on the part to simulate the important parameters specifically at the mold. The process of manufacturing plastic part begins with the part’s sketching, drawing, modeling and analysis followed by mold design and simulation. A 3-Pin Plug was selected as a case study. The mold with four cavities was designed for the mold. The straight in line cooling channel was included in the mold to offer higher heat transfer co-efficient. Injection molding process parameters were simulated in terms of filling time, plastic flow, injection pressure, average temperature and air traps. The results revealed that it has smooth flow with the filling time up to 0.5394 s. The highest average temperature and pressure recorded were 323.1 °C and 23.77 MPa, respectively. It recorded a very little air traps that probably occur in the product. In short, the development of the 3-Pin Plug with its mold design is promising with the supports and results produced from the simulation. It is recommended to study the cooling effects by replacing current conventional cooling with conformal cooling channel in mold design.

Snap-fit is a joining method that is used to join two or more parts together. It removes the need for and usage of external connecting components like screws, bolts, and nuts. Snap-fit is a cost-effective and time-efficient strategy that favours Design for Assembly (DFA). The focus of this paper is to conduct simulation and experimental study on retention forces of the snap-fits, which are affected by several parameters such as thickness of the beam (Tb), length of the beam (Lb), the width of the beam (Wb), and retention angle (β), so that the best snap-fit design can be produced. This study’s material is ABS, and the programme used to design the snap-fits is Autodesk Inventor. The snap-fit simulation is performed using ANSYS software, and the experimental analysis is performed using a Universal Testing Machine (UTM) to calculate the retention forces. The snap-fits are made with a 3D printing machine, and a total of 16 models have been tested. The lowest retention forces are by Model 2 with the value of 1.7219 N for simulation and 1.6573 N for experimental. High retention forces allow the snap-fits to break and can cause injury when disassemble. The factor that affects retention forces of the snap-fits is length of the beam.

A high prevalence of musculoskeletal disorders (MSDs) has been reported amongst university students related to prolonged demands and multiple study tasks. This study investigated the MSD symptoms and ergonomic risk amongst students who attended online learning classes and prolonged sitting on their study workstations. The study population comprised engineering, science, and technology students (n = 58). The Cornell Musculoskeletal Discomfort Questionnaires (CMDQ) and Rapid Entire Body Assessment (REBA) were used to evaluate body discomfort and posture, respectively. REBA worksheet was employed to assess the entire body posture of the students with 100 and above total body discomfort scores. The body part that obtained higher complaints amongst participating students was the lower back (19.28%) followed by the upper back (17.93%), neck (10.68%), and wrist (right and left) (9.66% and 7.84%). About 13% of participants were exposed to a very high-risk level, and 27% had a high-risk level. The overall mean score was 7.3, under the medium-risk range. MSD symptoms in students are almost highly prevalent. A mean REBA score of 7.3 equals a medium-risk assessment accompanied by guidance to “further investigate, change soon.” The management team in engineering, science, and technology faculties responsible for student health and comfort should prioritize methods to address and control musculoskeletal discomfort.

The performance of cutting tools is critical in the ever-changing world of machining and manufacturing. This research explores the novel approaches used to improve these instruments’ surfaces in an effort to maximize their effectiveness and durability. The idea of surface texturing, which is the painstaking process of engraving patterns onto tool surfaces to strengthen their tribological qualities, is fundamental to this investigation. By going over a number of methods in detail—from photolithography to laser texturing—the study highlights how significant these changes are. Specifically, micro-textured tools have demonstrated outstanding performance in the machining of difficult materials, such as Ti–6Al–4V alloy and aluminum alloys. In addition, innovations, such as laser interference patterning, have changed the game by drastically lowering machining forces and enhancing surface roughness. This study opens the door for new developments and improved manufacturing results by demonstrating the revolutionary potential of surface texturing in the machining sector.

In the dynamic field of composite manufacturing, there exist various critical areas of research that offer potential for future progress. The utilization of natural fiber-composites presents an opportunity for a more sustainable approach, with a focus on improving the extraction and refining processes, as well as enhancing their inherent properties such as resilience and environmental resistance. The domain of epoxy tooling, which is being shaped by the impact of Industry 4.0, holds the potential to improve production efficiency and drive the adoption of environmentally friendly epoxy resins. In the realm of self-healing materials, there exists a captivating domain that exhibits promise in refining its inherent self-repair capabilities and uncovering novel avenues for utilization. Furthermore, the prioritization of agility and autonomy in composite fabrication highlights the necessity for advanced mobile and autonomous tools, enhanced control over material movements at various scales, and predictive simulation techniques. The examination of these research dimensions has the potential to significantly transform the field of composite manufacturing, with a particular focus on enhancing aspects such as quality, sustainability, and innovation.

The field of 3D printing, also known as additive manufacturing, is currently at the forefront of technological progress. Its impact is substantial and wide-ranging, affecting various sectors such as tissue engineering and advanced mechanical applications. The integration of Polyethylene (PE) based composites is a key aspect of these technological advancements, as they are highly regarded for their strong mechanical properties and ability to be tailored to specific requirements. The latest advancements have brought attention to the utilization of High-Density Polyethylene (HDPE) nanocomposites combined with Titanium Dioxide, resulting in a notable increase of 37.8% in tensile strength. Moreover, the utilization of sandwich structures featuring syntactic foam cores is advancing the frontiers of marine applications, particularly when enhanced with Glass Micro Balloons (GMB). The combination of wood flour and recycled polyethylene showcases its potential for utilization in extrusion and injection molding processes, while also enhancing its resistance to thermal degradation, specifically up to temperatures of 180 °C. Concurrently, the utilization of degradable PCL/PEG composite meshes, in conjunction with the incorporation of the antibiotic azithromycin, presents a highly encouraging prospect for medical interventions, particularly in the treatment of Pelvic Organ Prolapse (POP) and targeted infections. Research further highlights the multifunctionality of Additive Manufacturing (AM) technologies in examining sandwich structures, uncovering enhanced flexural stresses in reentrant structures. The utilization of carbon fiber reinforcements in 3D printed components highlights the presence of distinct anisotropy in shear properties, which is influenced by the printing orientation. This observation emphasizes the promising prospects of employing various material combinations in the realm of 3D printing. This overview provides a concise summary of the swift advancements in the field of 3D printing, emphasizing the significant impact of material science and technology in influencing forthcoming applications.

Hybrid Flow shop Scheduling (HFS) problem is one the most sought after researched work either in dealing with modelling of the schedule or finding optimum ways to solve the problem. However, there are still gaps in the literature where the study on multi-objective HFS with energy utilization (EE) remains unsolved. The proposed study presents a model to solve scheduling in HFS and several optimization approaches to solve EE-HFS problem. The aim of this work is to present the best approach to minimize both energy utilization and completion time in HFS. The work will consider unrelated machine capabilities that are independent of one machine to another. The optimization of EE-HFS was performed utilizing the Artificial Bee Colony Optimization (ABC) across 12 benchmark HFS problems. Based on the optimization results, it was observed that the ABC algorithm exhibited superior performance compared to 8 other algorithms in most of the problem scenarios. The ABC algorithm performed better than 46% of the optimization objectives from other algorithms and demonstrated the most stable convergence when compared to other algorithms dependent on iterations under consideration.

Collaboration between university and Small Medium Enterprises (SMEs) is a concept for achieving innovation through knowledge creation but challenges are huge due to resource constraints and limited capabilities. In Malaysia, despite of SMEs being the major contributor to the national economic landscape, innovative capacity is still below expectation indicating the poor relationship between the collaboration actors. One method to enhance innovation capabilities through interaction is to develop a new collaborative model with embedded attitudinal element that reflects the favoured mode of interaction for SME Malaysia. This work intends to study the perception of the researchers and industry practitioners towards effective collaboration using interaction process parameters that promote co-creation of knowledge in enterprises based on trust and honesty. A survey is instrumentally developed to study the respondents’ perception on the interaction attributes. The data obtained are analyzed based on Taguchi method and linear regression. The results show that impactful collaborative engagement is achieved if interaction between actors is informal. The findings support the argument on SME preferring an informal mode of interaction to foster successful SME-University collaborations through enhanced interaction and help to complement the current national innovation policy which is yet to be updated since 2012.

Shoe racks are immensely popular in furniture stores and there are more designs on the market today. Shoe racks are furniture pieces designed to organize and store shoes neatly. They come in various styles, including basic shelves, shoe cabinets with doors, benches with built-in storage, over-the-door organizers, and wall-mounted options. Shoe rack designs cater to different storage capacities, space constraints, and aesthetics, providing functional solutions to keep shoes easily accessible and clutter-free. However, despite their popularity, there remain notable gaps, defects, or deficiencies in the racks that hinder their effectiveness. The study will focus on incorporating anthropometric data specific to Malaysians into the final product's dimensions. The aim is to enhance the ergonomic fit of the shoe rack for Malaysian users, designers, and engineers. This will involve collecting anthropometric measurements from a representative sample of Malaysian students, taking into account factors such as gender, body size, and height. By addressing the specific needs of Malaysian users, this study aims to contribute to the development of more functional, user-friendly, and ergonomic automatic shoe racks. The findings of this research will not only benefit consumers in the market but also aid designers and engineers in creating more effective and user-centered furniture solutions.

The employment of deep learning architecture for defect detection in the manufacturing industry has gained due attention owing to the advancement of computational technology. Conventional means of defect detection by manual visual inspection by operators are often deemed laborious as well as prone to mistakes. In the present study, a feature-based transfer learning approach is used to classify surface defects. The KolektorSDD database is used in the present study. Two pipelines were developed to investigate its efficacy in detecting the defects, namely the VGG16-kNN and VGG16-SVM pipelines, respectively. It was demonstrated from the study that the VGG16-SVM pipeline was more superior compared to the VGG16-kNN pipeline as no misclassification transpired in either the test or the validation dataset. It could be concluded that the proposed pipeline is suitable for the classification of surface defects.

This study presents a comprehensive simulation model for the ultrasonic shot peening (USP) process applied to aluminum A380 alloy, a material widely used in the automotive industry. The model accurately predicts deformation and stress distribution on the alloy’s surface under varying shot parameters, including shot material and initial velocity. The results indicate that the maximum von Mises stress (VMS) produced by the USP process should lie between the yield strength and the ultimate tensile strength (UTS) of the aluminum A380 alloy to avoid excessive plastic deformation and initiate beneficial crack propagation. The study found that the required initial velocity for the stainless steel shot is 60–120 ms−1, while the tungsten carbide shot only requires an initial velocity of 60–90 ms−1 to produce hydrocompressive residual stress on the surface layer of the aluminum A380 alloy at optimum conditions. The simulation model developed in this study provides a valuable tool for optimizing the USP process parameters for aluminum A380 alloy, enhancing the performance of automotive parts made from this material, and reducing the resources required for experimental testing. Future work should focus on validating the simulation model against experimental results and exploring the effects of other process parameters on the alloy’s mechanical properties and ductility.

This paper assesses maintenance costs and reliability associated with additional risk reduction measures in a methanol plant to achieve an “as low as reasonably practicable” (ALARP) target risk level of 1 × 10−7 yearly. It proposes an approach to evaluate maintenance and reliability using economic loss risks, focusing on incidents involving piping and reactor operations. The study examines economic losses from fatalities, injuries, equipment damage, business interruptions, and emergency services due to toxicity, thermal radiation, and overpressure events at different reactor pressures. Case studies involve comparing five plants: a 76 bar normal methanol plant with a 42 m3 reactor and four modified plants with 7.6 m3 reactors at pressure of 76, 200, 350, and 500 bar. The methanol reactor contains hazardous substances: hydrogen, carbon dioxide, carbon monoxide, and methanol. Losses, including fatalities, injuries, equipment damage, are estimated by combining consequence analysis outcomes with individual and equipment values. Business disruptions consider downtime and industry value added per employee, while emergency service losses amount to two percent of the total. Results indicate the normal methanol plant needs RM 12.7 million annually for maintenance to achieve ALARP, while modified plants reduce costs by 75% to 91% compared to the normal methanol plant.

Carbon dioxide (CO2) is a greenhouse gas that occurs naturally, but since the industrial revolution, the quantity of CO2 in the atmosphere has increased by roughly 50%. This increase in CO2 changes the earth’s climate. Carbon capture and storage is a significant countermeasure to climate change. Membrane filtering techniques, absorption, adsorption, and cryogenic separation techniques are often used to handle carbon dioxide separation from natural gas. Distillation columns are sophisticated pieces of equipment used to separate mixtures of liquids and gases. A major difficulty with distillation columns is the collection of pollutants on the column’s surfaces, which can diminish the separation process’s efficiency. The objective of this study is to investigate the rotating separator in cryogenic conditions by using computational methods. The design of RZB was simulated in transient-state, multiphase Eulerian model with the renormalization group (RNG) k − ε model equations with standard wall function. The simulation shows how both phases of cryogenic temperatures had an impact. The separation effectiveness was strongly impacted by the cryogenic temperature. Results of methane and carbon dioxide show 97% separations in liquid outlets and gas outlets.

This study demonstrates how disposing of the cold energy (ice) produced by the conversion of liquefied natural gas to natural gas has a substantial impact on the industrial sector in particular. In addition to lowering the consumption of electricity, which can have a detrimental influence on the environment, the produced cold energy can be employed in several industrial industries. Because coal is one of the primary sources of electricity in Malaysia, high energy use can cause the ozone layer to thin more quickly. The use of the cold energy available includes the generation of cold water, dry ice, cooling or freezing districts, and more. Intermediate fluids have a variety of characteristics, including critical temperature, critical pressure, density, and latent heat. Using computational fluid dynamics (CFD) and simulation software tools, a double-pipe heat exchanger will be modeled in this study. Besides that, the objective of this study is to optimize design parameters [temperature (T), tube length (L), diameter (d), pressure (P), and velocity (v)] for heat exchanger, and last but not least, to design a heat exchanger that will transmit cold energy to the cold chamber. The results of the study found that the temperature on the outer pipe produced a difference of 3.1 K between the initial temperature and the new temperature while the inner pipe shows 10.1 K differences between the initial temperature and new temperature.

Since natural gas (NG) now accounts for 20% of the world’s primary energy needs, the demand for gas is expected to increase by a total of 140 billion cubic metres (bcm) between 2021 and 2025. NG also referred to as wet gas is mostly made up of methane with traces of other molecules; if left untreated, it will cause several operational issues. An environmentally beneficial method of processing natural gas is the inline centrifuge separator, which is one way to recover parameter-free water and NG stream without using any chemicals or inhibitors to stop hydration formation from happening. The inline centrifuge hydrate inhibitor, however, offers improved possibilities for preventing hydration while providing a straightforward design, low production cost, and excellent efficiency. RNG k–ɛ is turbulent, and mixed multi-phase models were connected during the conceptional design stage to determine flow behaviour using computational fluid dynamics (CFD). Computational methods for developing and improving products are used in this way. The purpose of this work is to investigate and simulate the impact of flow characteristics such as pressure and velocity from the proposed blade. The results of the numerical simulation showed that the blade tended to produce the closest results from the validation design decreasing from 40.5 to 37.9 kPa.

Using value stream mapping (VSM) and simulation techniques, the study aims to improve production line performance in the electronics manufacturing industry. The study focuses specifically on analysing and optimising the assembly line operations of an assembly line in an electronic manufacturing company. The primary objectives of this study are to develop a current state VSM, simulate its performance using discrete event simulation software, and propose a future state VSM for increased efficiency. This study employs lean principles and simulation technology to identify and eliminate production line constraints and inefficiencies. The results of this study demonstrate the ability of VSM and simulation techniques in generating substantial improvements in key performance indicators. The results demonstrate significant improvements in throughput, with the proposed models obtaining significant improvements over the current state VSM. Additionally, the implementation of lean tools and simulation techniques results in increased value-added percentages and enhanced line efficiency. The decrease in average waiting time and blocked status contributes further improvement to the production line’s overall performance and productivity. The study concludes that the integration of VSM and simulation techniques offers valuable insights for optimising production line performance, minimising waste, and enhancing overall efficiency. The findings can serve as a foundation for future optimisation efforts and as a guide for decision-makers seeking to increase production line productivity and efficiency.

Modern machining techniques like wire electrical discharge machining (WEDM) enable the cutting of complicated shapes. Parameter optimization is necessary during the machining process of titanium alloy. This optimization may help in cost reduction in machining. The objective of this study is to determine the best parameters for machining processes based on single and multi-objective optimization. The study focused on three machining responses: material removal rate (MRR), gap size (GS), and surface roughness (Ra) in the EDM machining process. To achieve the most optimal outcome, teaching–learning-based optimization (TLBO) was employed by comparing the results obtained from optimized data with experimental data based on WEDM parameters. The optimization using TLBO demonstrated that the optimized data produced better results than the existing experimental data for MRR, GS, and Ra. This method is superior and more efficient than the traditional approach used for parameter optimization in machining processes. It is specifically designed to optimize the parameters of the EDM machine learning process.

Inconsistency data from the knowledge management system comes from unstable data and entities that have been defined. Data not being saved on the proper platform will result in inconsistent data. This research aims to develop an ontology for the selected joining process that provides a standard understanding structure to support user interoperability across heterogeneous data. Based on that, it is important to develop the ontology by identifying and classifying the correct entities. The basic formal ontology has been adopted as the top-level ontology for the development of the ontology for the joining process. The ontology is then expanded with entities related to the joining process. The ontology needs to be evaluated based on consistency, accuracy, and adaptability. A sample of data from research related to the welding process was used to test the capability of the ontology to infer information. As a result, proper ontology development will solve the inconsistency of data in the knowledge management system.

Detecting stroke is important to reduce the likelihood of permanent disability and increase the chance of recovery. Brain stroke lesion segmentation is an important procedure, especially when a specific brain portion needs to be analyzed. In this project, a brain stroke lesion segmentation algorithm using a modified U-Net (MUN) architecture will be developed. The MUN has a dimension-fusion capability, in which the images are analyzed separately using 2D U-Net and 3D image downsampling processes, before being fused at two points during the downsampling processes. The MUN accuracy is then compared with a regular 3D U-Net (UN). Three training options are further developed and compared. It is found that the MUN architecture produces higher training accuracy, but slower training duration compared to UN. Despite the capabilities of MUN, it cannot be further validated due to software limitations. Further improvement on the algorithm using other libraries is essential to enhance the capability of the MUN.

This study compares the performances of the GA-FOPID and MOGA-FOPID controllers, which are fractional-order proportional–integral–derivative (FOPID) controllers tuned using genetic algorithm and multiple-objective genetic algorithm for position tracking accuracy of robotic manipulator, respectively. The tuning process of six control gains in the three FOPID controllers is technically challenging to achieve high position accuracy of robotic manipulator. This study is performed to objectively assess the performances of genetic algorithm and multiple-objective genetic algorithm in tuning the six control gains in the FOPID controller. From the simulation study, it is interesting to note that the GA-FOPID and MOGA-FOPID controllers produce approximately 8.2990 and 14.6307% reductions of the mean square error in the angular position accuracy response of robotic manipulator as compared with the GA-PID controller. It is envisaged that the GA-FOPID and MOGA-FOPID controllers can be useful in designing effective position tracking accuracy of robotic manipulators.

The introduction of vehicle steer by wire (SBW) system technology permits customization dynamics of vehicle behavior by ability to tune the steering system response. This improves the vehicle performance, maneuver, and stability. However, the challenging issues in SBW system, is how to generate the steering torque. The important of steering torque is for a driver steering feel. With this sense of feel, driver has more confidence level during maneuver. Moreover, this torque used as function steering wheel returnability is whereby, when driver releases their handoff from the steering wheel, a steering return back to its initial position. In conventional steering system, this torque is created from the contact between tire and road surface. This torque then follows through column shaft and directly to the driver. However, in SBW system, this column shaft is eliminated and replaced with sensors and actuator. Therefore, in SBW system, this torque has to be artificially generate and control. This paper proposes a method of generate steering torque for SBW type of distributive steering system. The torque of front wheel motor left and right is taken into account to generate steering torque. The compensation torque is applied to improve the driver steering feel and classical controller used to control the torque. The mathematical numerical analysis is conduct to show the effectiveness of proposed control algorithm and compare with EPS steering system as a benchmark. Based on finding results, the proposed control algorithm able to generate steering torque in parallel provides driver steering feel during maneuver.

Distracted driving causes most road accidents and injuries. Cell phones, food, radios, and passenger conversations are all distractions. Distractions may slow a driver's response time and increase the risk of accidents. Studies reveal that even minor distractions may impair a driver's ability to drive safely. This study examines how distracted driving affects male drivers. Using US and Malaysian databases will do this. This research included drivers with at least two years of experience to guarantee a representative sample. Each dataset chose 35 and 58 drivers. Driver distraction level, a new class characteristic, has four levels: no, mild, moderate, and severe. Weka software was used for “data mining” to get insights from a vast dataset. Weka is a strong data mining and machine learning program including algorithms for data preparation, classification, regression, clustering, and visualization. We applied these algorithms on their datasets using its GUI or command-line parameters. Speed, braking, acceleration, steering, lane offset, lane position, and time were used to assess driving performance. Male drivers were more likely to be distracted driving based on their driving skills which is identified by the driving performance indicator (DPI).

Potholes are a common problem on roads, caused by weather, vehicle activity, and poor maintenance. Potholes can be hazardous for drivers, cars, and motorcycle riders. Potholes are often filled with asphalt or concrete. A methodology for automatically identifying potholes on road surfaces using computer vision methods is potholes detection utilizing image processing. This technique can be used to improve road maintenance by quickly locating potholes, enabling early repairs, and lowering the risk to drivers and their cars. This study emphasizes a Gaussian noise filtering technique for the developed infrastructure of image pre-processing stage. Thus, this study also suggests four methods for segmentation detecting potholes in images: image thresholding (Otsu), Canny edge detection, K-means clustering, and fuzzy C-means clustering. The effectiveness of the different image segmentation techniques was tested in MATLAB 2019a, and the results were generated in terms of accuracy and precision. The results were compared with each other to draw a conclusion on their viability.