A Multilayer Perception for Estimating the Overall Risk of Residential Projects in the Conceptual Stage

Abstract

The ability to foresee hazards early plays a critical role in estimating the entire cost of a project. Although several studies have established models to predict the total cost of a project at a conceptual stage, there remains a research vacuum in measuring the overall risk at this stage. Using artificial neural networks, this research provides a strategy for estimating the overall risk in residential projects at the conceptual stage. There are eight important components in the suggested paradigm. The model was created using data from 149 projects. In the first hidden layer in the model, there are five neurons, and in the second hidden layer, there are three neurons. The suggested model’s mean absolute error rate was 11.7%. In the conceptual stage of residential projects, the number of floors, the type of interior finishes, and the implementation of risk management processes are the significant aspects that influence the overall risk. The proposed model assists project managers in precisely estimating the project’s overall risk, which leads to a more accurate estimation of the contract’s entire worth at the conceptual stage, allowing the stakeholders to decide whether or not to proceed with the project.