Forecasting Pipeline Construction Costs Using Time Series Methods
Sooin Kim, Bahram Abediniangerabi, Mohsen Shahandashti
Abstract
Pipe and labor costs constitute about seventy percent of pipeline project costs. The accurate prediction of pipe and labor costs is invaluable for cost estimation of capital pipeline projects by helping to eliminate or at least reduce cost under- or over-estimations. The research objective of this paper is to develop and compare various time series methods to forecast pipe and labor costs. The 20-city average pipe and labor costs from 1995 to 2016 published monthly by Engineering News-Record (ENR) were used to develop the time series models. The accuracies of these forecasting models were evaluated using ENR pipe and labor cost data from 2017 to 2019. The results show that predictions with seasonal autoregressive integrated moving average (ARIMA) models are more accurate than those with the other models, such as Holt exponential smoothing. The results contribute to pipeline construction community by helping cost estimators to prepare more accurate bids for pipeline projects.