Engineering Journal (Faculty of Engineering, Chulalongkorn University, Bangkok)
Not a member yet
1223 research outputs found
Sort by
Constraint Programming in Single Machine Scheduling for Minimizing Makespan with Multiple Constraints
Get PDFThis study focuses on developing a scheduling model for sequencing a set of jobs with different release times in a single machine to meet non-similar due dates as well as to reduce total sequence-dependent setup time. A constraint programming (CP) model is proposed to solve the scheduling problem by minimizing makespan under multiple constraints, namely release times, sequence-dependent setup time, and due dates. The proposed constraint programming model is tested and compared with the baseline method derived from as-is scheduling of alloy wheels manufactures. The computational experiments show the proposed constraint programming model outperforms the baseline method in the average improvement in makespan and total setup time. For small-size problems, the proposed scheduling model were optimally solved in a short time, achieving the best average improvement in makespan of 4.8826% and the best average improvement in total setup time of 45.7924%. Despite increasing problem sizes, the proposed scheduling model's computational time deteriorates but continues to provide the best solutions, achieving the best average improvement in makespan of 7.4891% and the best average improvement in total setup time of 55.4033%
Electronic Tracking Device of Mass Public Transport Vehicle for Evaluating Driving Performance in Thailand
Get PDFThailand Mass Public Transport has a location tracking device along with a vehicle speed limiter attached to each public transport vehicle combined with Mass Public Transport Policy that aims to prevent dangers caused by improper driving behavior. These actions result in the troubles caused by inappropriate driving behavior decreasing drastically. However, risks from driving have more cases that the vehicle’s position and speed cannot determine and analyze. Thus, this research aims to develop a data-collecting device that collects linear acceleration, angular velocity, and magnetic field while transmitting data to an online database. The collected data enables the creation of a three-dimensional simulation from ten different public transport vehicle routes. The two main goals of developing a data-collecting device are to maximize the data collection frequency and evaluate its effectiveness in a real environment while the public transport vehicle is on duty. From ten vehicle routes, the device was able to collect stable data at a frequency of 50 Hz, with a reliability rate of over 50%. However, the device encounters various problems from external factors and bus layout diversity while testing in real environments on the road, which have been solved and are ready for real usage and statistical data analysis in the future
Downwash Investigation of Horizontal Tail Plane Configuration for 19-Passenger Aircraft Based on a Wind Tunnel Test
Get PDFAn investigation has been conducted on wind tunnel test data of 19-passenger aircraft to see the phenomena of the downwash effect. The objective of this investigation is to analyze the downwash effect on the variation of horizontal tailplane configuration that is installed on the vertical tailplane of the aircraft and on the variation of flap configuration. Each flap configuration represents the condition at the flight profile of cruise, takeoff, and landing. The horizontal tailplane configuration is varied by changing the angle of incidence and the vertical position on the vertical tailplane. Analytic calculation was conducted on wind tunnel result data to quantify the downwash effect. A computational fluid dynamic simulation is performed to obtain the visualization of the downwash effect and for wind tunnel data verification. From the investigation, it is found that the lower position of the horizontal tail has a smaller amount of downwash than other positions. The flap configurations have greatly affected the perceived amount of downwash. The greater deflection of the flap generated downwash even more
Cardiovascular Classification Using Efficient Net on Electrocardiogram Images
Get PDFCardiovascular disease ranks among the top causes of mortality, frequently caused by sudden obstructions within blood vessels. Timely identification and intervention are essential for minimizing the impact of the disease. This research employs image augmentation techniques to correct class imbalance in an ECG image dataset divided into five categories: Normal, Abnormal Heartbeat, Myocardial Infarction, Previous History of Myocardial Infarction, and COVID-19. The balanced dataset includes 6,322 images. To improve classification accuracy for cardiovascular diseases, three pre-trained models visual Geometric Group, Residual, Dense, and Efficient Network with Version 2, were trained on the balanced ECG dataset. Critical hyper parameters were fine-tuned, yielding optimal performance with a learning rate set at 0.00001, a dropout rate of 0.3, and utilizing the Adam optimizer. EfficientNet-V2 outperformed the other models, reaching a level of accuracies of 96.22%, precision 96.34%, recall 96.31%, 95.89%, 94.75%, and an F1-Score of 96.33%, thus exceeding the performance of Densenet 161, Densenet 201, ResNet50 and VGG16
Light Weight Residual Convolutional Neural Network for Atrial Fibrillation Detection in Single-lead ECG Recordings
Get PDFElectrocardiogram (ECG) analysis constitutes the most important approach able to classify heart infarction anomalies. These anomalies can be identified from the changes in various features of the ECG signal. In this paper, we are proposing a new features-based classification method of short-time single-lead ECG signals. The goal of this method is to classify these ECG signal into one of the following classes: normal, atrial fibrillation, other abnormalities, and too noisy as defined by the dataset. This is a challenging problem because of the severe imbalance between the classes, where the normal class makes up the majority of the samples in the dataset. The second challenge in this dataset is the fact that the sample ECG signals have a variable length (it varies between 3 to 60 seconds). The proposed method considers three main processes. The first process consists of detecting inverted ECG record by analyzing the signal range and mean in a sliding-window. The second process involves the extraction of many features effective in characterizing ECG signals and detecting abnormalities. These features include morphological, Heart Rate Variability, statistical, time/frequency amplitudes, and special Atrial Fibrillation (AF) features. The third process represents the main contribution by designing a lightweight residual Convolutional Neural Network (CNN) model for the classification of short-time single-lead ECG signals. This model is composed of five layers with two residual connections where advanced CNN concepts such as Batch Normalization, DropOut, and Leaky-ReLU are used. Compared to state-of-the-art solutions, the proposed method achieved the best performance with F1-score of 95.11% using inversion correction
Cultivating Knowledge: Design and Development of the Infographic (Baja Sawit) Mobile Application to Enhance Oil Palm Fertilization Practices among Smallholders
Get PDFThis study presents detailed information for developing the Infographic Oil Palm Fertilize Mobile Application called Baja Sawit. The Infographic Baja Sawit Mobile Application development aims to promote good practices in palm fertilization among smallholders in a simple and accessible manner using Mayer's 12 Principles of Multimedia Learning approach. The study illustrates the systematic development of the application's content, marked by precision in integrating insights derived from a comprehensive literature analysis and interviews with five highly skilled professionals in oil palm fertilizers. The workshop done with the Malaysian Palm Oil Board (MPOB) used the Infographic Baja Sawit App, monitoring participants throughout. This study utilized the Mobile Application Development Life Cycle (MADLC) methodology, encompassing seven key phases: identification, design, development, prototyping, testing, deployment, and maintenance. The resulting Infographic Baja Sawit Mobile Application offers substantial advantages to oil palm smallholders by actively promoting best practices in palm fertilization. This study has substantial implications for developing a digital information infographic culture tailored to oil palm smallholders from various regions in Malaysia
Survey Studies of Software-Defined Networking: A Systematic Review and Meta-analysis
Get PDFSoftware-Defined Networking (SDN) represents a novel technological paradigm expected to dominate the next-generation networking. Since the emergence of SDN, there has been a significant increase in publications addressing a wide range of issues, leading to a proliferation of surveys and reviews. Consequently, due to the growing number of survey studies in the SDN domain, it has become imperative to establish a comprehensive taxonomy for these papers. This paper presents a systematic taxonomy for classifying, categorizing, and analyzing state-of-the-art survey research within the SDN field. Our systematic taxonomy process involves selecting reviews and surveys related to keywords such as ‘SDN,’ ‘survey,’ ‘challenge,’ ‘taxonomy,’ ‘review,’ and ‘state-of-the-art.’ We sourced these papers from reputable digital databases, including Web-of-Science (WoS), ScienceDirect, Scopus, and the Institute-of-Electrical-and-Electronics-Engineers’ Xplore, all of which comprehensively cover recent literature. In total, we analyzed 442 survey and review studies published between 2012 and 2021, covering various journals and conferences with a focus on both general topics and specific subtopics of SDN. This paper represents the first epistemological study conducted on the literature of SDN. Our study aims to serve as a valuable resource for researchers, journal editors, and funding agencies, facilitating the identification of research gaps and making a significant contribution to future studies
Application of Collaborative Robots for Increasing Productivity in an Eyeglasses Lenses Manufacturer
Get PDFThis research focuses on a framework for making decisions when adopting collaborative robots (cobots) to collaborate with or replace human workers. Top management at a real-life case study firm that manufactures a variety of eyeglasses lenses wants to implement cobots in the sorting process since such a repetitive task has been shown to have a significant negative influence on workers' ergonomic ailments. Its current procurement decision-making process focuses solely on financial perspectives without taking into account any other significant criteria. Therefore, the purpose of this study is to investigate the elements that are crucial in deciding whether to use cobots in manufacturing lines., Multivariate statistical methods, comprising the exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), are applied to analyse the elements that are associated with the latent variables such as safety, ergonomics, productivity, quality, system, internal organisation and external organisation. In addition, alternative deployments of cobots in the case study are validated through the ARENA simulation software. More specifically, the results showed that using cobots in the workplace might boost output while lowering WIP, waiting times, the number of tasks in queue, and the workforce. In addition, cobots may reduce employee ergonomic risk and enhance workplace safety
A Review of Building Information Modeling and Simulation as Virtual Representations Under the Digital Twin Concept
Get PDFBuilding Information Modeling (BIM) is a highly promising technique for achieving digitalization in the construction industry, widely used in modern construction projects for digitally representing facilities. Nevertheless, retains limitations in terms of representing construction operations. The digital twin concept may potentially overcome these limitations and initiate advanced digital transformation in the construction industry as it has revolutionized the product lifecycle management in the manufacturing industry. This research provides a critical review of applying digital twin in the construction industry. Altogether, 140 papers from related journals and databases were reviewed. The digital aspect of twinning consists of BIM and simulation modeling. These two techniques have been used to create virtual or digital representations of actual buildings and real-world construction processes. However, integrating and applying BIM and simulation modeling according to the digital twin concept remains to be fully studied. Comprehensive evaluations of BIM, simulation modeling, and digital twin will provide a well-defined framework for this research, to identify direction and potential for digital twin in the construction industry, thereby progressing to the next level of digitalization and improvement in construction management practice
Bi-level Planning Model for Optimal Battery Energy Storage Allocation Considering Optimal Daily Scheduling Using Mixed-Integer Particle Swarm Optimization
Get PDFThis paper proposes a bi-level optimization (BLO) approach for optimal battery energy storage system (BESS) allocation (OBA) in distribution network (DN) considering optimal BESS daily scheduling (OBDS). The objective is to obtain the best locations and daily scheduling of BESSs that minimize total energy loss in DNs. In the upper-level of the proposed BLO method, the OBA is solved by mixed-integer particle swarm optimization (MIPSO). Meanwhile, the OBDS is solved as a sub-problem by particle swarm optimization in the lower-level of BLO. The proposed BLO based OBA considering OBDS algorithm had been tested with IEEE 33-bus radial distribution test system using load profile of Thai’s power system during summer, winter, and rainy seasons comparing to mixed-integer genetic algorithm (MIGA) method. The simulation result shown that the proposed lower-level OBDS can efficiently minimize the total daily loss by BESS scheduling. Moreover, the proposed algorithm can also achieve the optimal placement of BESS