Fulment, Arnold Kabyemala2019-10-262020-01-072019-10-262020-01-072014Fulment, A.K (2014) Forecasting the volume of water sold by DAWASCO using time series techniques: the case Dar es salaam, Master dissertation, University of Dar es Salaam. Dar es Salaam.http://localhost:8080/xmlui/handle/123456789/2605Available in print form, East Africana Collection, Dr. Wilbert Chagula Library, Class mark (THS EAF HD1699.T34F84)Dar es Salaam being the highly developed city in Tanzania, sustainable and adequate water supply is one of the greatest challenges to Dar es Salaam Water and Sewerage Corporation, (DAWASCO). This study specifically dealt with modeling, identification techniques, diagnostic and forecasting of the time series data of volume of water sold by DAWASCO using Box-Jenkins methodology.The procedure used in evaluating models and forecasting was through monthly data on volume of water sold for Kiosk Usage (C1), Domestic Usage (C2), Commercial Usage (C3), Industrial Usage (C4) and Institutional usage (C5) from April, 2007 to December, 2012. The original time series data were plotted; there were trends (upward trends) which suggested that time series data were characterized with non stationarity. Hence the first difference was taken to the original time series data, which was enough to transform time series data to stationary time series. Also seasonal differencing was taken to time series data so as to remove most or all of seasonality effect. In identification, estimation and diagnostic checking stage, the study of sample autocorrelations and partial autocorrelations were done in order to identify the model as well as estimate the associated appropriate parameters. Formal test statistic was applied to verify an appropriate model adequate for time series data, namely, Akaike Information Criterion (AIC). The selected adequate seasonal time series models were used to forecast the volume of water sold by DAWASCO for the year 2013 for each category of subscribers. They included: SARIMA (2, 1, 0) x (0, 0, 2)12 for Kiosk Usage (C1), SARIMA (0, 1, 1) x (1, 0, 2)12 for Domestic Usage (C2), SARIMA (0, 0, 2) x (2, 1, 2)12 for Commercial Usage (C3), SARIMA (1, 1, 1) x (0, 0, 2)12 for Industrial Usage (C4) and SARIMA (2, 1, 1) x (1, 0, 2)12 for Institutional Usage (C5). In concluding, the identified models can be used as guidelines to understand probable future volume of water sold to respective categories of subscribers.enWater resources developmentWater rights, Water supplyEconomic aspectDar es SalaamTanzaniaThesis