Modelling Sewer Systems Costs with Multiple Linear Regression

The aim of this paper is the establishment and validation of cost functions for the various assets of sewer systems, namely gravity and raising pipes, manholes and pumping-stations. Costs are defined as a function of the main physical characteristics of the assets, such as, the pipe material and diameter, excavation depth and percentage of pavement (for sewer pipes), the manhole depth (for manholes) and flow rate, pump head and pump power (for pumping stations,). A four-step methodology was followed: 1) data collection, processing and analysis, 2) present value calculation, 3) key parameters identification and cost functions estimation, and 4) cost functions validation. Cost and infrastructure data for construction contracts of sewer systems managed by Aguas de Portugal (AdP) were analyzed. Cost functions were estimated based on multiple linear regression analysis and compared with the ones obtained in previous studies.