The Journal of Industrial Technology Thepsatri Rajabhat University

Applying the Long Short-Term Memory Technique to Sales Forecasting for an Auto Parts Vendor

2025-10-01T14:21:55+07:00

In the face of increasingly fluctuating car demand, sales orders for automotive parts from various vendors have highlighted the growing importance of sales forecasting and predictive analytics to prevent overstocking and stock outs. However, most of the existing research continues to rely predominantly on traditional time-series forecasting methods that do not fully capture dynamic demand. To address these issues, this study applies a deep learning time-series forecasting technique to reduce the discrepancy and find a suitable method for forecasting the orders. The objective of this study is to propose an algorithm for forecasting sales based on an LSTM network, using historical sales data from an auto parts vendor. This study was conducted in an automotive vendor located in Samut Prakan, utilising a dataset covering 130 weeks from 2023-2025 to. A total of 70 percent of the sales data were employed for training, whereas 30 percent were allocated to the testing sets. The selected item stems from high-value auto parts consisting of 77443582-441V, 77443582-521V, 55183973-060V, and 55183973-440V. The four auto parts were selected based on their high sales volume and strategic importance to the vendor’s inventory management. These items contribute to over 40% of total sales revenue and are frequently subject to stockouts, making accurate forecasting critical. Compared with actual sales orders, the results show that the MAPE forecast error from the LSTM model is lower than the 6.14% error, and it achieves an R-squared of 90.228%, outperforming the existing forecasting technique used by the purchasing department. These findings suggest that the proposed forecasting model can effectively improve the sales order accuracy of automotive part vendors.

An Investigation of Parameter Effects on Surface Roughness and Kerf Taper in PMMA Cutting Using a CO2 Laser Cutter: A Full Factorial Experimental Design Approach

2025-09-05T08:45:36+07:00

This study aims to investigate the effects of various parameters on surface roughness and kerf taper in the cutting of PMMA material using a CO₂ laser cutter. A full factorial experimental design was employed, and the results were analyzed using Analysis of Variance (ANOVA) to determine statistical significance. The three control factors considered (3 levels per factor) were: cutting speed (3, 5, 7 millimeters per second), laser power (50, 70, 90 percent), and nozzle standoff distance (12, 14, 16 millimeters). The results of the analysis found that cutting speed and nozzle standoff distance significantly affected the mean values of surface roughness and kerf angle. Meanwhile, laser power and the interaction effects among the three control variables did not significantly affect the mean differences. The set of parameters yielding the optimal cutting quality was a cutting speed of 3 millimeters per second, a laser power of 90 percent, and a nozzle standoff distance of 16 millimeters. This was the condition that resulted in the lowest surface roughness and a kerf angle closest to 90 degrees. The findings of this study can be applied in practice to optimize cutting parameters, thereby enhancing the quality and efficiency of PMMA workpiece production processes in industrial laser cutting.

Assessment of Energy-Saving Potential of Air Conditioning Systems Using Low-Enthalpy Outdoor Air in Hot-Humid Climate of Thailand

2025-08-19T08:44:03+07:00

Air conditioning systems are among the highest energy-consuming components in buildings, particularly in hot-humid climates such as Thailand, where cooling demand continues to rise. The economizer system-introducing outdoor air with lower enthalpy to reduce cooling loads - has gained attention as a potential energy-saving measure. However, its application in humid climates remains challenging due to high relative humidity. Therefore, the objective of this study is to assess the feasibility and energy-saving potential of economizer applications in Thailand by analyzing suitable climatic conditions and time periods, and to provide insights for practical design and implementation. This study utilized daily weather data during the winter season between 2024 and 2025, obtained from representative stations in five regions of Thailand: Chiang Rai (North), Nakhon Phanom (Northeast), Lopburi (Central), Nakhon Nayok (East), and Prachuap Khiri Khan (South). Three time periods were analyzed: full day (24 hours), working hours (08:00 - 17:00), and nighttime (20:00 - 06:00). The hourly enthalpy calculations indicated that suitable conditions for economizer operation occur when the enthalpy is below 50.465 kJ/kg and the outdoor air temperature ranges between 20 and 28.7 °C. Based on this criterion, the number of hours in which outdoor air could be utilized varied across regions; for example, Nakhon Phanom recorded 432 h/year (all-day period), Nakhon Nayok 279 h (Office time), and Lopburi 157 h (Night time). Furthermore, the energy-saving potential analysis indicated that Lopburi and Prachuap Khiri Khan achieved the highest absolute reductions, whereas Nakhon Nayok exhibited the highest percentage savings (12.56%) despite the smaller actual reduction. In contrast, the southern region showed the lowest savings proportion (0.27 - 2.5%), even though energy consumption was relatively high during certain periods. In summary, the implementation of economizers in Thailand exhibits varying potential depending on region and time period. Appropriate selection based on local temperature and humidity conditions can enhance system performance and support the development of sustainable energy-efficient buildings.

Thermal Efficiency Enhancement of a Solar-Powered Dryer Equipped with Cylindrical Heat Fins for Turmeric Drying

2025-08-28T10:37:14+07:00

This research aimed to enhance the thermal efficiency of a solar-powered dryer by integrating cylindrical heat fins for turmeric drying. The prototype dryer was designed with a drying area of 1×1 m² and a parabolic-shaped roof, equipped with cylindrical heat fins of varying lengths (10, 20, 30, 40, and 50 mm) to investigate their effects on internal temperature, heat transfer rate, energy efficiency, and thermal efficiency of the solar dryer. Experimental studies were conducted under simulated solar radiation conditions using a 1,500 watt halogen lamp. The results indicated that the heat fin with a length of 40 mm achieved the highest internal temperature of 54°C and the highest heat transfer rate of 37 W. Additionally, it provided the highest average energy efficiency, ranging between 4.7% and 6.3%, at an air mass flow rate of 2.5×10^-3–7.0×10^-3 kg/s, which outperformed fins of other lengths. When comparing the thermal efficiency between dryers with and without heat fins, the fin-equipped dryer exhibited an average thermal efficiency approximately 2% higher throughout the drying process, with a peak thermal efficiency of 36% during the initial drying stage (10:30–12:00 hrs). Moreover, the fin-equipped dryer demonstrated a faster reduction in moisture content during the initial drying phase without compromising the quality attributes of the product, including moisture content, texture, color, and aroma. From practical application at the community level, the developed solar-powered dryer was able to reduce electricity costs by approximately 5 baht/kilogram/day compared to conventional electric dryers. The evaluation of user satisfaction revealed an overall satisfaction rating at a "very good" level, with an average score of 91.55%, reflecting the suitability of this technology for further application in community-based agricultural product processing.

Metaverse for learning resources and agricultural tourism in Lopburi Province

2025-06-10T10:34:28+07:00

The purpose of this study was to (1) develop a metaverse for agricultural learning and tourism in Lopburi Province, (2) evaluate the effectiveness of the metaverse for agricultural learning and tourism in Lopburi Province, and (3) assess the satisfaction of users regarding the use of metaverse technology for agricultural learning and tourism in Lopburi Province. The sample group for this research consisted of 150 individuals, selected through purposive sampling. The research tools included (1) the metaverse for agricultural learning and tourism in Lopburi Province consist Baan Suan Kwan, Bee Conservation Center, Kajonwit Mushroom Farm, Oasis Farm and Gold Medal Garden, (2) an evaluation form for the effectiveness of the metaverse for agricultural learning and tourism in Lopburi Province, and (3) an evaluation form for user satisfaction with the use of metaverse technology for agricultural learning and tourism in Lopburi Province. The statistical methods used for data analysis were mean () and standard deviation (S.D.). The research findings revealed that the effectiveness of the metaverse, as evaluated by experts, was of high quality ( = 4.37, S.D. = 0.75), indicating the effective management of content and high-quality experiences in agricultural learning and tourism through the metaverse technology in Lopburi Province. The metaverse was found to effectively raise awareness and attract the target audience. Furthermore, the evaluation of user satisfaction from the sample group indicated a high level of satisfaction ( = 4.42, S.D. = 0.77), reflecting positive feedback from users who experienced the metaverse. The use of this technology was shown to enhance knowledge and understanding of agriculture and agricultural tourism while also making tourism more engaging.

Development of a Prototype for an Electricity Billing System Based on Actual Consumption for Vendors at the Khon Doi Walking Street Market in Mae Chan District, Chiang Rai Province

2025-10-03T09:27:40+07:00

This research presents the development of a prototype for an electricity billing system based on actual consumption for vendors at the Khon Doi Walking Street Market in Mae Chan District, Chiang Rai Province. This system aims to facilitate accurate electricity fee collection from market vendors based on their actual usage, as well as to improve efficient market management. The system design primarily utilizes Internet of Things (IoT) technology. For the hardware aspect, the MQTT protocol is used to transmit data between devices and the database/web application. The system can display various parameters in real-time and process them automatically. It also records data in a database, which allows for viewing historical data online. After the development of the prototype was completed, its performance was tested in each component to evaluate its efficiency. The average operational efficiency of the prototype system was found to be 97.45%. Subsequently, the researchers had 30 sample participants evaluate their satisfaction with each aspect of the prototype system. The overall average satisfaction score was found to be 4.48, which is a high level of satisfaction.