Machine learning algorithms for constructions cost prediction: A systematic review

Machine learning plays a vital role in construction estimation which could make improve the project's safety, and reliability. Many studies have been proposed to explore the potential opportunities to review this technology in the construction cost in structure and transport fields. However, no comprehensive study to review the global research trends on this area's advancement in construction cost. The goal is to taxonomy, review, and summarize the state-of-the-art knowledge body on this topic in a systematic manner based on machine learning (ML) and deep learning (DL) approaches. To achieve this, this paper considered many studies in construction management related to bibliographic records retrieved from the Scopus database by adopting a quantitative analysis approach. This paper found that from 2017 to 2021, there has been a considerable increase in the number of publications in this domain. We categorized and explained civil projects into structures and transport cost, ML/DL as supervised and unsupervised approaches, and the evaluation metrics proposed to evaluate the performance of ML-Cost estimations in the civil area. The findings will help both professionals and researchers to understand and evolve the recent trend research ML/DL methodologies and their role played in the construction management domain.