Journal of Advanced Research Design

Part Detection Model for Aerospace Manufacturing Quality Control using Convolutional Neural Networks

Wed, 04 Jun 2025 00:00:00 +0700

Ensuring the precision and efficiency of quality control processes in the aerospace manufacturing industry is critical to remain competitive. This paper presents an advanced approach to address the challenge of identifying and categorising similar aerospace components quality effectively using Artificial Intelligence’s computer vision and convolutional neural network’s deep learning technique to resolve the issue of parts misclassification. Specifically, in the development of a comprehensive database for object detection using Phyton programming and tailored to the unique requirements of the aerospace industry under study. The methodology involved collecting a diverse dataset comprising 50 images per class, annotated with bounding boxes and class labels, covering a pilot of D232B and D233B machining parts. The dataset is partitioned into training, validation, and test sets to facilitate the model training and evaluation. Furthermore, tools for managing and accessing the dataset were introduced, including the interface for image labelling. Leveraging on TensorFlow, the effectiveness of this approach was able to be tested in training the part detection model with 94% accuracy for part D233B and 97% accuracy for part D232B, showcasing its suitability for real-world aerospace manufacturing quality inspection applications. Overall, this project represents a significant advancement in improving quality inspection process in the aerospace industry, offering a valuable approach for enhancing efficiency and accuracy in component identification.

A Study on the Feasibility of Producing Concentrated Cricket Protein Powder using Ultrasound-Assisted Extraction

Sirirat Pungchompoo, Supasit Chooklin — Wed, 04 Jun 2025 00:00:00 +0700

This research aims to examine the feasibility of extracting cricket protein and investigates the marketing factors influencing consumer decisions to purchase cricket protein products, utilizing a sample group of 303 respondents. The analysis, employing the Chi-square statistic (p < 0.05), identifies product features and distribution channels as the most influential factors affecting the purchasing decision. Nevertheless, the impact of product price on purchasing decisions varies depending on occupational status. Consequently, sales promotion strategies should be tailored according to gender, age, and income. Pearson's product moment correlation coefficient indicates a correlation between personal factors and marketing variables. From a technical perspective, the study suggests that cricket protein should be extracted using the ultrasound technique at a temperature of 35 degrees Celsius, with a solvent ratio of 1 N Sodium Hydroxide at an S/L (Solid to Liquid) ratio of 1:6, and an extraction time of 15 minutes, which yields an optimal extraction rate of 60%. Furthermore, the economic analysis reveals that with a daily production capacity of 100 kilograms, a project duration of 10 years, and an initial investment of 4,317,532 baht, the net present value of the project is 1,777,605 baht, resulting in an internal rate of return (IRR) of 26% —exceeding the minimum required rate of return of 15.05%. The project also boasts a benefit-to-cost ratio of 1.42 and achieves full capital recovery in just 5.08 years. Consequently, it is deduced that this project is a viable investment opportunity.

Analyzing Criteria for Potential Adaptive Reuse of Pre-War Shophouses: A Multiple Linear Regression Approach

Wed, 04 Jun 2025 00:00:00 +0700

Pre-war shophouses are among the national historic structures that needs conservation work. However, choosing between adaptive reuse and other options is difficult because of the numerous factors involved. Therefore, before selecting a potential adaptive reuse, the adaptive reuse criteria must be taken into account. The criteria for adaptive reuse include six major categories: economic, environmental, social, architectural, technological, and legislative. These criteria represent as independent variables meanwhile the potential adaptive reuse represent as dependent variable. In this study, a specific group of respondents such as town planners from local governments, valuers from the valuation and Property Service Department (JPPH), architectural firms, researchers, and building owners who are involved in the adaptive reuse of pre-war shophouse buildings were given the questionnaire. In this work, multiple linear regression and descriptive statistics were used to analyze the data. In order to determine whether the shophouse can be reused rather than demolished and replaced by new structures, this paper applies multiple linear regression to analyze the relationship between the criteria for potential adaptive reuse. In other words, the study will identify the crucial conditions for potential adaptive reuse for a chosen new use of pre-war shophouses. The results show the new shophouse and the criteria for adaptive reuse have strong correlation.

Automatic Counting of Shrimp Larvae using Artificial Intelligence

Wed, 04 Jun 2025 00:00:00 +0700

The research is to revolutionize the shrimp farming industry by developing a computer vision and Artificial Intelligence (AI) system for accurate and efficient shrimp counting. Recently, the aquaculture industry plays important roles in the global demand for seafood. In particular, shrimp farming has become a significant contributor to the industry. However, the process of counting the shrimp larvae is a labour-intensive and time-consuming task that often requires manual effort, leading to inefficiencies, inaccuracies and increased operational costs. Thus, the project was conducted to challenge the shrimp larvae counting with robust and efficient method. The proposed system captures and analyses images of Macrobrachium Rosenbergii shrimp post larvae (PLs) of varying stages and quantities using a high-resolution webcam and a Convolutional Neural Network framework (CNN). The presence of molts, feed and debris in the imaging chamber is considered by the system. The goal is to have a low mean absolute error when counting large and small PLs. This technology's successful implementation will not only improve the accuracy and reliability of shrimp counting but will also clear the way for counting other small aquatic species in their larval stages, such as fish, crabs, oysters and eggs. The methodology of the project entails training the system with a large image dataset and testing its performance with the trained model. The development of a fast and precise shrimp counting AI system, which has the potential to revolutionize the industry and improve customer satisfaction, is one of the significant results and findings. Finally, this study proposes a different approach to automate counting shrimp larvae counting by combining computer vision and AI, providing a more accurate, efficient and reliable solution for the aquaculture industry.

The Effect of Fibre Loadings on the Mechanical and Thermal Properties of Sugar Palm/Waste Tyre Rubber Reinforced Polylactic Acid hybrid Composites via Fused Deposition Modelling

Wed, 04 Jun 2025 00:00:00 +0700

Sugar palm fiber (SPF) is a renewable, biodegradable, and eco-friendly resource, while waste tire rubber (WTR) enhances the properties of these composites. Poly(lactic acid) (PLA) serves as a promising biobased and biodegradable matrix material for these green composites. The mechanical and thermal properties of hybrid composite are crucial characteristics in the development of a hybrid composite Fused Deposition Modeling (FDM) filament since the printing mechanism of FDM strongly depends on the heating and extrusion process. To enhance interfacial adhesion, WTR and SPF were treated with 6% NaOH and 3% silane, resulting in a composite formulation of 97.5% PLA and 2.5% SPF/WTR. Three different fiber loadings were assessed: 75% SPF:25% WTR, 50% SPF:50% WTR, and 25% SPF:75% WTR. The filaments produced using a twin-screw extruder were utilized to 3D print tensile specimens according to ASTM D638, impact specimens according to ASTM D256 and the test of Thermogravimetric Analysis (TGA). The results indicated that the 75% SPF:25% WTR fiber loading achieved the highest mechanical properties for tensile strength of 37.89 MPa and appropriate that 25%SPF:75%WTR fiber loading achieved the highest mechanical properties for impact strength of 4.3 KJ/m2. Scanning electron microscopy (SEM) analysis also has been studied. The TGA test displayed similar thermal degradation patterns, suggesting that the ratio of components does not significantly alter the overall thermal stability of the composites. This enhanced performance is attributed to improved interfacial adhesion between the treated fibers and the PLA matrix, along with a uniform distribution of fibers throughout the composite. These findings suggest that sugar palm and waste tire rubber hybrid composites are viable, high-performance alternatives for the filament extrusion and printability of hybrid composite filaments.

Exploring the Utilization of Reinforcement Learning Technique for Channel Accessibility in a Television White Space Database

Armie E. Pakzad, Raine Mattheus Manuel, Abbasali Pakzad, Lawrence Materum — Wed, 04 Jun 2025 00:00:00 +0700

This study aimed to explore integrating the reinforcement learning (RL) technique into a television white space (TVWS) database using MATLAB. An RL implementation was developed to recommend the optimal channel to a secondary user (SU), considering the time of query, the transmitter power of the device, and the SU’s location. Data regarding the allocation of television frequencies within the Very High Frequency (VHF) and Ultra High Frequency (UHF) ranges was collected to compile the TV White Space database (TVWSDB). These entities were constructed using the 2019 edition of the National Telecommunications Commission's frequency allotment table. The database was structured to facilitate the RL utilization of the different data analysis functions in MATLAB after importing the data as workspace data. The database stored the primary user's (PU) location, transmitting power, height above average terrain (HAAT), frequency of operation, and programming schedule. Furthermore, a comparative analysis was conducted between a linear search algorithm and the AI program, revealing a significant performance advantage of the latter. Specifically, the RL-based program exhibited a speed improvement of 2080 times over the linear search algorithm.

Evaluation of Egg Packaging Cushioning Systems Implemented with Lean Design Principles

Wed, 04 Jun 2025 00:00:00 +0700

Proper packaging is critical to ensure safe and efficient distribution in the egg industry. Packaging systems must be designed to meet safety standards and customer requirements. This study focuses on evaluating the performance of cushion packaging for eggs using the lean design approach to select the most suitable packaging method. Given the fragility of eggs, which are highly susceptible to damage from impacts and surfaces, the study compares the effectiveness of three cushioning materials: polyethylene terephthalate (PET), high-impact polystyrene (HIPS), and moulded pulp cartons. Free-fall tests were conducted at various heights and orientations (flat top and flat bottom), with peak acceleration (GP) data recorded using an accelerometer and displayed via TP3 software. The results indicated that moulded pulp cartons outperformed PET and HIPS in vibration resistance and durability, maintaining product integrity even at a maximum height of 5.5 ft. The findings suggest that moulded pulp cartons offer the best cushioning solution, thanks to their excellent energy absorption properties, minimizing the risk of product damage during transport.

An Enhanced Ants Colony Optimization Algorithm for Path Planning of Autonomous Robots in Continuous Environment

Han Heng, Muhammad Aizat, Wan Rahiman — Wed, 04 Jun 2025 00:00:00 +0700

This paper presents a method for mobile robot path planning in complex environments. First, an advanced Ant Colony Optimization (ACO) algorithm is introduced to find viable paths in a discrete grid environment, connecting the starting point to the destination. The proposed ACO algorithm incorporates a probabilistic prediction mechanism to enhance the efficiency of node selection. By integrating ACO's inherent heuristic factors, this approach offers directional intelligence and significantly increases the likelihood of the ant colony finding feasible initial paths. In the improved ACO algorithm, both the pheromone and heuristic factors are integrated into the new probabilistic prediction mechanism, which further improves path planning efficiency. The effectiveness of the proposed algorithm was evaluated using three different planar environment models of varying sizes and complexities. The results show that the algorithm's performance is minimally influenced by its control parameters. Additionally, a comparative analysis was conducted to evaluate the proposed algorithm against A* and traditional ACO. The comparison results indicate that the proposed algorithm outperforms the others in terms of path length, runtime, and success rate.

Healthcare Data System for a Digital Twin of a Chemotherapy Centre

Piya Noeikham, Nikorn Sirivongpaisal, Dollaya Buakum, Phungern Khongthong — Wed, 04 Jun 2025 00:00:00 +0700

The aim of this research is to enhance digital twin research, specifically in healthcare and smart health sectors. The main objective is to bridge the gaps in digital twin research in real-world healthcare units by proposing a new data model for a healthcare service digital twin. The research highlights the importance of secure and efficient database design through a case study at Hat Yai-Na Mom Hospital's chemotherapy centre. This innovation integrates a digital twin into healthcare and fills the gaps in literature on data systems for real healthcare units. The use of SQLite for database development adds uniqueness, making this research a pioneering effort in implementing digital twins in a healthcare context. Overall, the research provides valuable insights into database design considerations and possible applications in service system monitoring.

Preliminary Wind Resource Assessment of Songkhla, Thailand Using WAsP Simulation

Dhasara Bodhibukkana, Juntakan Taweekun — Wed, 04 Jun 2025 00:00:00 +0700

This study presents a preliminary wind resource assessment at specific sites in southwestern Thailand for the deployment of onshore wind power for the future deployment. The primary objective of this research is to estimate wind power potential by using the near-surface wind observations over a period of 10-year (2013-2022) acquired from four weather observation stations. WAsP simulations were conducted to analyse prevailing wind direction, wind speed, and power density. The Bonus 300kW Mk III wind turbine was selected to estimate energy production at each site at 30 meters above ground level (AGL). The findings identify four sites Sadao, Hat Yai, Kho Hong, and Songkhla that demonstrate strong potential for prospective wind farm development. The WAsP analysis indicates mean wind speeds of 2.40 m/s for Sadao, 5.83 m/s for Hat Yai, 3.84 m/s for Kho Hong, and 3.41 m/s for Songkhla. The estimated power densities for Sadao, Hat Yai, Kho Hong, and Songkhla are 86 W/m², 74 W/m², 259 W/m², and 74 W/m², respectively. The mean net annual energy production values calculated for Sadao, Hat Yai, Kho Hong, and Songkhla are 171.401 MWh, 629.079 MWh, 351.292 MWh, and 217.072 MWh, respectively.

A study on Wind-Wind energy for Electric Vehicles Charging Station on the Southern Highway of Thailand

Napass Kangwantrakool, Kittinan Maliwan, Juntakan Taweekun — Wed, 04 Jun 2025 00:00:00 +0700

This comprehensive research delves into the energy landscape of Southern Thailand. Leveraging sophisticated tools such as WAsP, the study conducts an exhaustive examination of the wind energy potential, taking into account the intricate geographical and climatic conditions of the region. The research meticulously fulfils its objectives, encompassing the assessment of wind potential based on multifaceted criteria, the development of a comprehensive wind atlas, and an intricate simulation of annual energy production. Moving beyond the technical realm, the study extends its scope to scrutinize the proposed electric vehicle charging stations from an economic perspective. This facet of the research aligns with global sustainability initiatives, adding a practical and applicable dimension to the findings. A notable outcome of the study is the strategic identification of five charging stations in the southern provinces of Thailand, namely Songkhla, Prachuap Khiri Khan, Surat Thani, Chumphon, and Narathiwat. This strategic placement is poised to play a pivotal role in supporting the escalating prevalence of electric vehicles in the region. The research pioneering in its nature, contributing significantly to the ongoing global transition towards renewable energy sources. The findings furnish foundational insights that contribute to both academic discourse and practical applications, not only in southern Thailand but also as a valuable reference for similar endeavours globally. In terms of specific values, the study conducts a detailed analysis of the average mean wind speed, wind power density, and other critical parameters. For instance, the average mean wind speed values for Chumphon, Narathiwat, Prachuap Khiri Khan, Songkhla, and Surat Thani are reported as 2.88 m/s, 4.18 m/s, 3.21 m/s, 5.43 m/s, and 2.16 m/s, respectively. Similarly, the average wind power density values for the mentioned locations are 77 W/m², 225 W/m², 139 W/m², 370 W/m², and 31 W/m², respectively. These metrics, along with other detailed statistical analyses, contribute to a nuanced understanding of the hybrid energy potential in southern Thailand.