NKRAFA JOURNAL OF SCIENCE AND TECHNOLOGY

Mechanical Properties of Pavement Recycling Materials as Base Layer for Quality Improvement with Cement and Asphalt Emulsion

2025-09-02T15:46:35+07:00

This research was conducted by using pavement recycling materials from rural road structure No.3027 with the reclaimed asphalt pavement thickness of 0.05 m and the reclaimed crushed rock thickness of 0.15 m. Firstly, the basic properties of the materials were tested and then designed on the basic of recycled materials with a cement hydraulic GU type in the ratios of 0, 1, 2, 3, 4, and 5% by total mass weight, Then the material was further improved with an asphalt emulsion of CSS-1 type in the ratio of 0, 1, 2, and 3% by total mass weight, and the results of the curing periods of 7 and 28 days were compared. The test results showed that the aggregate material was distributed close to the standard of crushed rock base layer type A with a maximum dry density of 2.291 g/cm³ and an optimum moisture content of 5.60%. The use of cement was 2.82% by total weight, resulting in an unconfined compressive strength of 17.50 kg/cm². Based on indirect tensile strength, indirect tensile stiffness modulus, and Resilient Modulus, it was found that CSS-1 can improve its mechanical properties of elasticity, permanent deformation resistance, and cracking resistance, where CSS-1 was equal to 1% by total mass weight. It was suitable for use to improve the base course layer material because of the increased amount of CSS-1. As a result, the material has a reduced unconfined compressive strength. In addition, the incubation period at the age of 28 helped to develop the hydration reaction of cement and asphalt emulsions and was also environmentally friendly because it reduced the use of new materials and also the pollution from transportation.

An Automatic OMR System Using Digital Image Processing with Hybrid K-Means and K-Nearest Neighbor Classification

2025-09-23T16:29:03+07:00

This research focused on developing an automated Optical Mark Recognition (OMR) system to overcome the issues of delay and errors associated with manual checking. We utilized Digital Image Processing (DIP) techniques combined with a Hybrid Classification approach employing K-Means and K-Nearest Neighbor (K-NN) algorithms to enhance accuracy and reduce data noise. The prototype processed 15,000 samples from 100 answer sheets (150 marks per sheet), divided into 80% training and 20% testing sets. The methodology included image preprocessing, segmentation, and feature extraction, focusing on Mean Intensity, Variance, and Pixel Density, to classify marks as "Filled" or "Empty." Comparative results across four methods (K-NN, SVM, Decision Tree, and
K-Means + K-NN) showed that the latter technique yielded the highest average Accuracy from multiple repeated trials at 99.30% (SD = 0.23). Furthermore, evaluation using a Confusion Matrix on the 3,000 test samples confirmed a high Accuracy of 99.30%, Precision of 99.11%, Recall of 98.56%, and an F1-score of 98.83%. These findings confirmed that integrating K-Means with K-NN significantly improved the system's correctness and stability, demonstrating its potential for real-world deployment without reliance on specialized paper or high-quality scanners.

Development of An Intelligent Classification of Plastic Waste System

2025-09-08T10:41:47+07:00

Plastic waste is a severe environmental issue in Thailand, accounting for 12% of total waste, yet only 25% is recycled. Manual classification is therefore limited and prone to errors. This study applied YOLOv11, a real-time object detection model with higher accuracy and speed than its predecessors, to the classification of four major types of plastic waste: HDPE, PET, PP, and PS. A web application was developed using Flask to enable users to detect and classify plastic waste conveniently, rapidly, and accurately. The experimental results indicated that the YOLOv11 model can effectively classify four types of plastic waste (HDPE, PET, PP, and PS), achieving an average latency of 22 ms per image and a processing speed of 55 FPS. Evaluating the classification performance by plastic type, PET achieved the highest accuracy of 0.994 and precision of 0.999, followed by HDPE with an accuracy of 0.983 and precision of 0.958. PP and PS recorded accuracies of 0.958 and 0.976, and precisions of 0.943 and 0.971, respectively. In terms of recall, HDPE reached 1.0, PET 0.97, PP 0.898, and PS 0.918. The F1-scores of PET and HDPE were the highest at 0.984 and 0.978, while PS and PP achieved 0.944 and 0.92, respectively. These results demonstrated that the model can reliably and accurately classify PET and HDPE waste. The system was suitable for practical deployment in waste sorting facilities or recycling centers, integrated with IoT devices such as robotic arms or conveyor belts, enabling real-time waste separation, reducing reliance on manual labor, and enhancing overall waste management efficiency.

A Simulation of Directional Antenna System for UAV Detection Using Radio Frequency Signals

2025-07-24T10:59:57+07:00

This research presented a directional antenna system model for the detection and localization of Unmanned Aerial Vehicles (UAVs) using radio frequency signals. The study employed MATLAB-based simulation techniques for Radiation Pattern analysis to construct and evaluate the antenna's reception characteristics. The simulation results demonstrated the system's precise capability in detecting and determining the direction of a UAV. The location of the signal source was confirmed through the analysis of Signal Strength relative to the rotation of the simulated antenna.

The analysis was conducted across four primary simulation scenarios: Static Homogeneous, Dynamic Homogeneous, Static Heterogeneous, and Dynamic Heterogeneous antenna configurations. Simulation results from 100 trials for each model indicated that the Static Homogeneous model achieved the highest average detection efficiency at 34.9755%, while the Dynamic Heterogeneous model performed best among heterogeneous types with an efficiency of 24.5276%. The study concluded that an appropriately configured static arrangement of homogeneous antennas was the most effective approach to enhance the probability of UAV detection under the simulated parameters. This research provided a crucial foundation for the future design and evaluation of anti-drone defense systems and was highly beneficial for developing surveillance or alert systems in critical areas, such as airports or military installations.

A Real-Time Sales Data Management System for Small and Medium-Sized Enterprises (SMEs)

2025-11-06T09:45:48+07:00

This research aimed to design and develop a real-time sales data management system for small and medium-sized enterprises (SMEs) to enhance the efficiency of collecting, processing, and visualizing sales data from multiple branches accurately and in a timely manner. The case study employed in this research was a community-based warehouse-type retail business located in the northeastern region of Thailand, comprising 23 branch stores across four provinces: Ubon Ratchathani, Yasothon, Sisaket, and Amnat Charoen. The developed system was implemented as a web-based application using Visual Studio Code, PHP, and JavaScript, integrated with a MySQL database deployed on a cloud-based infrastructure. The system was developed using an agile development methodology to ensure alignment with actual business operating contexts. System performance evaluation was conducted using a system usability scale questionnaire administered to 70 users from a total population of 257 users, representing 27.24% of the population. The evaluation results indicated that the system achieved an average score of 84.78 out of 100, corresponding to the “highest” level. High ratings were observed in terms of ease of use, data accuracy, and the system’s capability to effectively support actual operational tasks. These findings demonstrated that the integration of a real-time system significantly enhanced the efficiency of strategic data management and encouraged entrepreneurs to adopt digital technologies to strengthen their competitive capabilities. Furthermore, the system provided a foundation for future predictive data analytics to support effective decision-making.

Development of a Web-Based 3D Model Visualization Application to Support Community-Based Product Design Processes

2026-03-17T13:57:26+07:00

This research and development study aims to develop a web-based interactive 3D visualization application to support participatory product design in community contexts. The study is grounded in the limitations of conventional two-dimensional presentations, which are unable to fully communicate product shape, proportions, and details, resulting in misunderstandings and inefficiencies in design-related decision-making. The system was developed using WebGL technology and the Three.js library to enable real-time 3D rendering through web browsers without requiring additional software installation. The scope of the study included the development of eight packaging and product label models in .glb and .obj formats. Model optimization techniques, including polygon reduction and file compression, were applied to enhance system performance. Technical testing results indicated that reducing the polygon count by approximately 30% decreased file sizes from 120–150 MB to 40–60 MB and reduced loading time from 8–10 seconds to 5–8 seconds, while maintaining an average frame rate of 18–20 FPS. The system functioned reliably across desktop computers, smartphones, and tablets without errors. In addition, the system was tested with 51 participants and evaluated by five experts. The overall mean score was 4.74 with a standard deviation of 0.51, indicating a high level of suitability. Users reported improved clarity in understanding product forms, reduced design communication errors, and enhanced collaborative decision-making at the community level. In conclusion, the developed system serves as a digital tool that enhances the product design and community-based product development process, with strong potential for application in educational settings and future research.

Quantitative Risk Assessment of Hourly Solar Radiation Forecasting using Monte Carlo Method: A Comparative Stability Study of CPR Model in Thailand

2025-11-24T13:38:00+07:00

Efficient management of solar energy systems requires accurate hourly global solar radiation data. This study aimed to analyze quantitative risk and establish confidence intervals for solar radiation forecasting using the Collares-Pereira and Rabl (CPR) model. The study applied Monte Carlo Simulation with 10,000 iterations, based on the Normal Distribution Assumption, utilizing Root Mean Square Difference (RMSD) and Mean Bias Difference (MBD) statistics from standard measurement stations in Chiang Mai and Ubon Ratchathani. The results indicated that while the mean forecast values for both stations were similar, the 95% confidence interval width for Ubon Ratchathani was 49.8% wider than that of Chiang Mai. Furthermore, the probability of error exceeding 10% in Ubon Ratchathani was three times higher. These findings suggested that solar system design in the Northeastern region required a higher safety margin compared to the Northern region to accommodate significantly higher uncertainty fluctuations. However, these results relied on the error distribution assumption, which should be further investigated in future research.

Building Renovation Process Using Building Information Modeling (BIM) and Building Energy Code (BEC) to Reduce Energy Consumption in Buildings: A Case Study of the Thai Public Broscasting Services Television (Thai PBS) Transmitter Building in Ubon Ratchathani Province

2025-10-30T13:33:44+07:00

This research aimed to study the current condition of the transmitter building of Thai Public Broscasting Services Television (Thai PBS), Ubon Ratchathani Province, and to analyze and evaluate the energy efficiency of the building before and after the renovation of the building envelope using Autodesk Revit Program in Building Information Modeling (BIM) and using the Building Energy Conservation (BEC) design standards to reduce energy consumption of the transmitter building of Thai PBS, Ubon Ratchathani Province. This research used a mixed research methodology, which involved studying real data (Empirical data) from building surveys using Autodesk Revit as a tool for Building Information Modelling (BIM) (Simulation modeling) and using Building Energy Code Software version 2.6.0 (BEC V.2.6.0) for energy assessment (BEC) to be used as a guideline for building energy conservation.

The results of the study found that before the building renovation, the total energy consumption of the building was 54,227.76 kW.h/year, not passing the specified criteria, which was higher than 35,779.27 kW.h/year. After the building renovation, the total energy consumption of the building was 35,550.97 kW.h/year, passing the specified criteria, which was lower than 35,779.27 kW.h/year. Most of the energy used in the building came from the lighting system, air conditioning system, and other electrical equipment, for which there were two alternatives for improvement, able to reduce the overall electricity usage of the building by 34.44 percent per year. Improvements were made by replacing the light bulbs with LED T8 fluorescent bulbs (3X18W). The LPD value was 7.72 W/m². The roof material color was double-wave roof tiles, light surface. Fiberglass insulation that received the high-efficiency label Stay Cool, 3 inches thick, Premium, was installed on ordinary gypsum board, 9 mm thick. The RTTV value was 7.86 W/m². The total energy value of the building was 35,550.97 kW.h/year, passing the standard criteria. This could be the best way to improve buildings. It could reduce the annual electricity consumption by 78,629.28 baht, with an investment of approximately 168,830 baht, and a payback period of only 2.15 years.

The Study of the Influence of Logistics Service Quality and Customer Relationship Management on Customer Engagement in International Freight Forwarding Business

2025-10-16T10:02:21+07:00

This study aimed to examine the influence of logistics service quality and customer relationship management on customer engagement in the international freight forwarding business in Thailand. A quantitative research methodology was employed, collecting data through questionnaires from a sample of 400 business customers who used international freight forwarding services. The data were subsequently analyzed using multiple regression analysis. The findings revealed that logistics service quality and customer relationship management could explain 49.3% of the variance in customer engagement. Specifically, the variables with statistically significant influence included Process Quality, Outcome Quality, Customer Orientation, and Information Technology, respectively. Meanwhile, Capability Quality, Organizational Capability, and Customer Knowledge Management showed no significant influence. The results suggested that freight forwarding operators should prioritize developing service processes, creating quality outcomes, establishing a customer-centric organizational culture, and implementing information technology to enhance efficiency. These efforts would lead to building customer engagement and competitive advantage in the international freight forwarding business. This research contributed to the development of logistics service strategies in Thailand by emphasizing the quality of interactions and service experiences for long-term success.

Development of an Augmented Reality (AR) Application for Cultural Tourism in Nonthaburi Province: A Case Study of Five Cultural Sites

2026-02-09T10:37:59+07:00

This research aimed to 1) study, analyze, and design requirements for presenting cultural tourism sites in Nonthaburi Province through five case study locations; and 2) develop an augmented reality (AR) application to support cultural tourism in these areas. This study employed a research and development (R&D) approach based on the System Development Life Cycle (SDLC). The process began with analyzing user requirements from tourists and the general public, followed by problem analysis and content design. Data were collected through field surveys, including photographs and video recordings of cultural tourism sites, which were used to design presentation formats and develop three-dimensional (3D) models integrated into the application. The application was designed to present information using augmented reality (AR) by integrating multimedia elements, including images, text, video, and three-dimensional (3D) models, in order to create an interactive learning experience for users. The development tools included Unity 3D in conjunction with the EasyAR SDK, and the application was implemented on the Android platform using Android Studio to support multimedia and AR-based content presentation. The application was developed using Unity 3D in conjunction with the EasyAR SDK and implemented on the Android platform to support multimedia and AR-based content presentation.

The results indicated that data collected from 120 participants through random sampling, categorized into four age groups (below 20, 21–30, 31–40, and above 41 years), showed an overall high level of satisfaction (Mean = 4.25, S.D. = 0.49). Satisfaction was evaluated across three dimensions: content, design, and usability. The findings revealed that participants aged over 41 reported the highest overall satisfaction. In terms of individual dimensions, participants aged 21–30 showed the highest satisfaction in content, while those aged over 41 demonstrated the highest satisfaction across all dimensions. Participants under 20 reported the lowest satisfaction in design.

The findings demonstrated that the AR application effectively enhanced the presentation of cultural tourism sites, resulting in high user satisfaction and improved understanding of cultural information. Furthermore, the development of the application enhanced the attractiveness of tourism destinations and positively influenced tourists’ travel decision-making.

Analysis of an SVIR Epidemic Model with Vaccination and Nonlinear Incidence

2026-03-13T09:32:25+07:00

This paper studies an SVIR epidemic model that incorporates vaccination and a nonlinear incidence term to represent behavioral responses and saturation in disease transmission.
The population was divided into four classes—susceptible, vaccinated, infectious, and recovered states —and both vaccine‑induced and naturally acquired immunity were allowed to wane, so that individuals could return to the susceptible class. The original model is an SIR model with nonmonotone incidence. Then we added a vaccinated compartment and assumed imperfect vaccination, meaning that vaccinated individuals may still become infected at a reduced rate.
We derived the disease‑free and endemic equilibria and computed the basic reproduction number by the next‑generation method, obtaining a threshold that separated disease elimination from persistence. Stabilities of the equilibria were investigated under appropriate parameter conditions using Jacobian and Routh–Hurwitz criteria. Numerical simulations then illustrated how changes in the vaccination rate and other key parameters shaped the long‑term dynamics, including convergence to a disease‑free equilibrium or approach to an endemic equilibrium with a nonzero infectious population. The findings indicated that maintaining vaccination coverage above a critical level was essential to keep and prevented sustained endemic transmission in the presence of nonlinear incidence and waning immunity.