Forecasting ENR Construction Cost Index: A Time Series Analysis Approach

Abstract

Every month, Engineering News-Record (ENR) publishes Construction Cost Index (CCI), which is a weighted-aggregate index of the 20-city average prices of construction activities. Although CCI is increasing in the long-term, it is subject to considerable short-term variations, which make it problematic for cost estimators to prepare accurate estimates. The ability to predict CCI can result in more-accurate bids and avoid under- or over-estimation. We conduct time series analysis and develop a CCI forecasting model based on the Seasonal Auto-Regressive Integrated MovingAverage (SARIMA) methodology. SARIMA investigates the underlying characteristics of the CCI data and makes systematic forecasts. The predictability of the developed SARIMA model is better than the predictability of the ENR subject matter experts’ forecasts. The developed forecasting model can be used to prepare more-accurate estimates for contractors and budgets for owners and reduce construction costs by better-timed project execution.