{"title":"Comparison between Different Techniques to Predict Municipal Water Consumption in Jeddah","authors":"Manal Alshahrani, S. Mekni","doi":"10.1145/3584202.3584265","DOIUrl":null,"url":null,"abstract":"Since forecasting water demand is very important either for the short or the long-term, many techniques were used to effectively do predictions such as the Moving Average (MA), the Auto Regressive (AR), the Autoregressive Integrated Moving Average (ARIMA) and the Long-Short Term Memory (LSTM) models. The latter model demonstrates its superiority in accuracy when predicting time series that's why in this article; we will use it to forecast the future water consumption in Jeddah City using the historical records collected from the Jeddah water authorities. We will also compare LSTM and ARIMA models. Moreover, in this paper we will use the Mean Square Error (MSE), the Mean Absolute Relative Error (MAPE), the Root mean square (RMSE), and the Mean Absolute Deviation (MAD) to decide and choose the best model. Experiments and interpretation of results obtained led to the conclusion of the superiority of LSTM in forecasting water demand in Jeddah City from 2004 to 2018.","PeriodicalId":438341,"journal":{"name":"Proceedings of the 6th International Conference on Future Networks & Distributed Systems","volume":"404 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 6th International Conference on Future Networks & Distributed Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3584202.3584265","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Since forecasting water demand is very important either for the short or the long-term, many techniques were used to effectively do predictions such as the Moving Average (MA), the Auto Regressive (AR), the Autoregressive Integrated Moving Average (ARIMA) and the Long-Short Term Memory (LSTM) models. The latter model demonstrates its superiority in accuracy when predicting time series that's why in this article; we will use it to forecast the future water consumption in Jeddah City using the historical records collected from the Jeddah water authorities. We will also compare LSTM and ARIMA models. Moreover, in this paper we will use the Mean Square Error (MSE), the Mean Absolute Relative Error (MAPE), the Root mean square (RMSE), and the Mean Absolute Deviation (MAD) to decide and choose the best model. Experiments and interpretation of results obtained led to the conclusion of the superiority of LSTM in forecasting water demand in Jeddah City from 2004 to 2018.