{"title":"汽车零部件销售预测配套业务资源增值服务","authors":"X. Qin, C. Ren, Y. Yu","doi":"10.1109/ISCTIS51085.2021.00041","DOIUrl":null,"url":null,"abstract":"In view of the nonstationary and fluctuating characteristics of auto parts sales data, an EMD-Prophet model is proposed to forecast the sales of auto parts and provide value-added services for auto parts related enterprises to grasp the market demand of auto parts and reduce inventory cost. Empirical mode decomposition (EMD) was used to denoise and stabilize the original data, and the intrinsic mode function components represented different frequency scales were obtained. By calculating the sample entropy of each component, the components with similar sample entropy were recombined to avoid error accumulation; After considering the impacts of seasonality and major events on sales, especially the fluctuations caused by COVID-19, an event window of holidays was established and the Prophet model was applied to predict each component. Every result was accumulated to form the final one. Finally, the model was validated by the data of the automobile industry chain platform and compared with ARIMA, Prophet and EMD-ARIMA models. The experimental result shows that the prediction accuracy of EMD-Prophet model is higher than other models, which verifies the effectiveness of the model.","PeriodicalId":403102,"journal":{"name":"2021 International Symposium on Computer Technology and Information Science (ISCTIS)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Auto parts sales forecast supporting business resource value-added service\",\"authors\":\"X. Qin, C. Ren, Y. Yu\",\"doi\":\"10.1109/ISCTIS51085.2021.00041\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In view of the nonstationary and fluctuating characteristics of auto parts sales data, an EMD-Prophet model is proposed to forecast the sales of auto parts and provide value-added services for auto parts related enterprises to grasp the market demand of auto parts and reduce inventory cost. Empirical mode decomposition (EMD) was used to denoise and stabilize the original data, and the intrinsic mode function components represented different frequency scales were obtained. By calculating the sample entropy of each component, the components with similar sample entropy were recombined to avoid error accumulation; After considering the impacts of seasonality and major events on sales, especially the fluctuations caused by COVID-19, an event window of holidays was established and the Prophet model was applied to predict each component. Every result was accumulated to form the final one. Finally, the model was validated by the data of the automobile industry chain platform and compared with ARIMA, Prophet and EMD-ARIMA models. The experimental result shows that the prediction accuracy of EMD-Prophet model is higher than other models, which verifies the effectiveness of the model.\",\"PeriodicalId\":403102,\"journal\":{\"name\":\"2021 International Symposium on Computer Technology and Information Science (ISCTIS)\",\"volume\":\"16 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 International Symposium on Computer Technology and Information Science (ISCTIS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISCTIS51085.2021.00041\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Symposium on Computer Technology and Information Science (ISCTIS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISCTIS51085.2021.00041","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Auto parts sales forecast supporting business resource value-added service
In view of the nonstationary and fluctuating characteristics of auto parts sales data, an EMD-Prophet model is proposed to forecast the sales of auto parts and provide value-added services for auto parts related enterprises to grasp the market demand of auto parts and reduce inventory cost. Empirical mode decomposition (EMD) was used to denoise and stabilize the original data, and the intrinsic mode function components represented different frequency scales were obtained. By calculating the sample entropy of each component, the components with similar sample entropy were recombined to avoid error accumulation; After considering the impacts of seasonality and major events on sales, especially the fluctuations caused by COVID-19, an event window of holidays was established and the Prophet model was applied to predict each component. Every result was accumulated to form the final one. Finally, the model was validated by the data of the automobile industry chain platform and compared with ARIMA, Prophet and EMD-ARIMA models. The experimental result shows that the prediction accuracy of EMD-Prophet model is higher than other models, which verifies the effectiveness of the model.