A Conceptual framework for predicting CIDA construction cost indices using big data analytics

Abstract

The Sri Lankan construction industry is significantly affected by financial risks stemming from volatile material costs, which are inadequately addressed by the static and often delayed CIDA price indices. This paper addresses this challenge by exploring the predictability of CIDA indices using Big Data Analytics (BDA). Adopting a mixed-method approach, which included an analysis of historical quantitative data and in-depth expert interviews, the study synthesizes findings to produce its main contribution: the development of a conceptual BDA framework designed to improve forecasting accuracy. This framework integrates diverse, real-time data sources and leverages advanced analytical techniques such as machine learning. The findings suggest that a data-driven approach can transform cost forecasting from a reactive to a proactive process, significantly enhancing the accuracy and responsiveness of CIDA indices and providing a robust tool for financial risk management.