A data driven hybrid approach towards developing a circular economy diffusion model for the building construction industry

Despite the growing body of knowledge about circular economy (CE) practices in the building construction industry (BCI), predicting the diffusion of CE in the BCI has not been adequately explored. The paper aims to explore a hybrid approach towards predicting the diffusion of CE practices in the BCI of a developing economy. The paper adopted the technology-organization-environment (TOE) framework to identify the essential factors influencing CE diffusion in the BCI. Based on the survey data from 303 experts, partial least squares structural equation modelling (PLS-SEM) was adopted to test the developed hypothesis on factors influencing CE diffusion. After that, machine learning (ML) algorithms were deployed to develop a CE diffusion prediction model for the BCI. SHapley Additive exPlanation (SHAP) was applied to interpret the contributions of each essential factor to the predictive model. The PLS-SEM results advocated that four major factors, namely technological compatibility, relative technological advantages, top management support, and organizational readiness, significantly and positively affect CE diffusion in the BCI. Also, random forest is the optimal ML algorithm for predicting CE diffusion, with an accuracy of 80.33% and ROC AUC (area under the Curve receiver operating characteristics) of 80.27%. Based on the SHAP results, the three most essential features contributing to the random forest model prediction are organizational readiness, top management support, and the relative technological advantages of CE adoption. The paper contributes to the extant literature on CE diffusion by providing a comprehensive data-driven approach that stakeholders can apply to forecast future trends and patterns in CE practices and to make strategic decisions and pragmatic plans for promoting CE diffusion in the BCI, particularly in the context of developing countries.