Search results

The increasing urgency to address climate change in construction has made green construction (GC) and sustainability critical topics for academia and industry professionals. However, the volume of literature in this field has made it impractical to rely solely on traditional systematic evidence mapping methodologies.

This study employs machine learning (ML) techniques to analyze the extensive evidence-base on GC. Using both supervised and unsupervised ML, 5,462 relevant papers were filtered from 10,739 studies published from 2010 to 2022, retrieved from the Scopus and Web of Science databases.

Key themes in GC encompass green building materials, construction techniques, assessment methodologies and management practices. GC assessment and techniques were prominent, while management requires more research. The results from prevalence of topics and heatmaps revealed important patterns and interconnections, emphasizing the prominent role of materials as major contributors to the construction sector. Consistency of the results with VOSviewer analysis further validated the findings, demonstrating the robustness of the review approach.

Unlike other reviews focusing only on specific aspects of GC, use of ML techniques to review a large pool of literature provided a holistic understanding of the research landscape. It sets a precedent by demonstrating the effectiveness of ML techniques in addressing the challenge of analyzing a large body of literature. By showcasing the connections between various facets of GC and identifying research gaps, this research aids in guiding future initiatives in the field.