A Synthesis of Structural Equation Model-Analytical Hierarchy Process, Nonlinear Autoregressive and Backpropagation Neural Network-Sensitivity Analysis for Construction and Demolition Waste Assessment in the Philippines

Abstract

The Philippines' Build-Build-Build initiative has led to an increasing number of infrastructures, which have a great impact on the public, natural resources, life cycle, and ecosystem. Hence, waste is inevitable during the infrastructure's construction or demolition. Thus, the factors of sustainable construction and demolition waste management (SCDWM) must be considered to evaluate. The structural equation model (SEM). nonlinear autoregressive exogeneous (NARX) neural network (NN) and backpropagation (BP) NN were conducted to predict the coefficient of determination of SCDWM's rate impact. Furthermore, the analytical hierarchy process (AHP) and sensitivity analysis (SA) were used to determine the relative importance factors of SCDWM. The assessment results of the synthesis method of SEM-AHP, BP-NN-Garson's Algorithm (GA), and NARX-SA, which are applied in the construction engineering management field provided an interesting insight into the Philippines' initiative. Utilizing SEM as causal analysis and BP-NN and NARX as machine learning would give a new recommendation for the industry 4.0 transition.