{"title":"Time series analysis of wheat moisture content variations for grain storage systems","authors":"","doi":"10.1016/j.jspr.2024.102395","DOIUrl":null,"url":null,"abstract":"<div><p>The Philippines is one of the many countries that consume and produce wheat grains. Despite having agricultural resources, the country still struggles to supply wheat and continues to import tons of wheat from different countries. Moreover, being a tropical country, the weather conditions exacerbate the parameters in defining grain quality, and thus, impact the preservation of grains in bulk storage systems. This study aims to understand the fluctuations of moisture content over time and its impact on grain quality and identify trends to utilize in optimizing storage systems management practices. Furthermore, this study investigated the moisture content variations, analyzed the temporal dynamics, and forecasted future trends using the ARIMA model, which can effectively handle time series data and has been applied in several agricultural applications. From the correlation patterns from lag intervals of the correlogram, there is a total of 24 ARIMA models generated with varying (p, d, q) values. These models were then evaluated using log-likelihood, AIC, RMSE, and MAE to select the best-performing and most appropriate model for forecasting wheat moisture content data. Out of these 24 models, Models (4,0,3), (4,0,4), and (1,0,0) are the top performing models, however, Model (4,0,3) emerged as a strong forecasting model in predicting wheat moisture content trends for the forthcoming two-year horizon. Its evaluation performance resulted in a log-likelihood of −69.38384, AIC of 156.7677, RMSE of 0.4248, and MAE of 0.30427, having the least values in three of the validation methods. When used in forecasting, the trend of the moisture content continues to consistently fall within the lower end of the accepted range. The distribution of the predicted values also shows a perfectly normal distribution, showing its effectiveness of the model for forecasting. These findings will serve as a guide towards enhancing operational efficiencies and sustainable growth within the flour milling landscape.</p></div>","PeriodicalId":17019,"journal":{"name":"Journal of Stored Products Research","volume":null,"pages":null},"PeriodicalIF":2.7000,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Stored Products Research","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0022474X24001528","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENTOMOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
The Philippines is one of the many countries that consume and produce wheat grains. Despite having agricultural resources, the country still struggles to supply wheat and continues to import tons of wheat from different countries. Moreover, being a tropical country, the weather conditions exacerbate the parameters in defining grain quality, and thus, impact the preservation of grains in bulk storage systems. This study aims to understand the fluctuations of moisture content over time and its impact on grain quality and identify trends to utilize in optimizing storage systems management practices. Furthermore, this study investigated the moisture content variations, analyzed the temporal dynamics, and forecasted future trends using the ARIMA model, which can effectively handle time series data and has been applied in several agricultural applications. From the correlation patterns from lag intervals of the correlogram, there is a total of 24 ARIMA models generated with varying (p, d, q) values. These models were then evaluated using log-likelihood, AIC, RMSE, and MAE to select the best-performing and most appropriate model for forecasting wheat moisture content data. Out of these 24 models, Models (4,0,3), (4,0,4), and (1,0,0) are the top performing models, however, Model (4,0,3) emerged as a strong forecasting model in predicting wheat moisture content trends for the forthcoming two-year horizon. Its evaluation performance resulted in a log-likelihood of −69.38384, AIC of 156.7677, RMSE of 0.4248, and MAE of 0.30427, having the least values in three of the validation methods. When used in forecasting, the trend of the moisture content continues to consistently fall within the lower end of the accepted range. The distribution of the predicted values also shows a perfectly normal distribution, showing its effectiveness of the model for forecasting. These findings will serve as a guide towards enhancing operational efficiencies and sustainable growth within the flour milling landscape.
期刊介绍:
The Journal of Stored Products Research provides an international medium for the publication of both reviews and original results from laboratory and field studies on the preservation and safety of stored products, notably food stocks, covering storage-related problems from the producer through the supply chain to the consumer. Stored products are characterised by having relatively low moisture content and include raw and semi-processed foods, animal feedstuffs, and a range of other durable items, including materials such as clothing or museum artefacts.