1. RSUIR at Rangsit University
  2. RSU Digital Collection
  3. บัณฑิตวิทยาลัย Graduate School
  4. Grad-ML-M-Thesis
Please use this identifier to cite or link to this item: https://rsuir-library.rsu.ac.th/handle/123456789/3503
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorBavornwit Rojsuwan-
dc.contributor.authorQuanrui Li-
dc.date.accessioned2026-09-04T03:10:02Z-
dc.date.available2026-09-04T03:10:02Z-
dc.date.issued2025-
dc.identifier.urihttps://rsuir-library.rsu.ac.th/handle/123456789/3503-
dc.descriptionThesis (M.Sc. (Management of Logistics)) -- Rangsit University, 2024en_US
dc.language.isoenen_US
dc.publisherRangsit University. Libraryen_US
dc.subjectSales forecastingen_US
dc.subjectRegression analysisen_US
dc.subjectElectronic industriesen_US
dc.subjectLinear models (statistics)en_US
dc.subjectPrinted circuitsen_US
dc.subjectBusiness forecastingen_US
dc.titlePrediction sales forecasing of plug-in board s product factory by using linear regressionen_US
dc.typeThesisen_US
dc.description.other-abstractThe aim of this study is to construct a sales forecasting model for Yaqi Factory’s plug-in board products to optimize inventory management and improve operational efficiency. By collecting sales data over 31 months, a linear regression method was used to model and analyze the relationship between the sales volume of five best-selling products (B5220, B5330, B5320, UU4222, and UU4320) and key attributes such as price, inventory, and production volume. The results indicate that the linear regression model based on six months of data exhibits high predictive performance, with an R² value of 0.816, explaining approximately 81.6% of sales variations. The study also examined the impact of different time spans (three months vs. six months) on prediction accuracy. Comparisons show that as the data time span increases, the model’s prediction error significantly decreases, demonstrating improved stability and accuracy. Specifically, the absolute error of the model trained on six months of data is 607, with a standard deviation of 646, outperforming the model trained on three months of data, which had an absolute error of 1,528. Additionally, the study visually assessed the model's accuracy by comparing moving average (MOA) sales with predicted sales. It was found that for products with significant fluctuations (such as B5330), the prediction results still require further optimization. This study provides valuable insights for Yaqi Enterprise in establishing reliable sales forecasting models, optimizing inventory management, and enhancing operational efficiencyen_US
dc.description.degree-nameMaster of Scienceen_US
dc.description.degree-levelMaster's Degreeen_US
dc.contributor.degree-disciplineMamagement of Logisticsen_US
Appears in Collections:Grad-ML-M-Thesis

Files in This Item:
File Description SizeFormat 
QUANRUI LI.pdf591.36 kBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.