Cost estimation for the city of Edmonton's water and sewer installation services using an artificial neural network model

The construction industry in Canada has been following a dynamic change over recent years. Cost estimation is integral to the success of the construction industry. Over the last six years, the City of Edmonton's Drainage and Maintenance Department has seen a marginal increase of about 12% in the installation of water and sewer services for residential occupancies in Edmonton.; The current estimating procedure has showed discrepancies between the estimated and actual cost sustained during the course of the projects to the effect of 60%, and in some cases, the excess of 60%.; This research investigates the factors that affect this variation in costs between estimates and actual costs within the existing process. A detailed analysis of all activities involved in the installation of the water and sewer service installations have been carried out.; The proposed methodology is based on the analysis of all the past data that has been obtained from the City of Edmonton's Drainage Section for the period of 1999 to 2004. The proposed methodology has been incorporated into the module, which integrates the Artificial Neural Network (ANN) and the current estimating system used by the City of Edmonton, 'SmartEST'. This thesis will focus on describing the algorithm used in the ANN and will assess the past data obtained for over 800 jobs (cases) performed over the period of the study.