Toward developing a predictive model for interpersonal communication quality in construction projects: an ensemble artificial intelligence-based approach

Abstract

Purpose: Following the job demands-resources theory, this study investigates the role of female managers in enhancing employee well-being in terms of psychological health via workplace resources. Design/methodology/approach: To accomplish this objective, we conducted a comprehensive literature review to identify key IPS. Subsequently, a fuzzy-based algorithm was employed to prioritize these skills. Following this, we developed an algorithm based on Extreme Gradient Boosting (XGBoost) to predict the quality of workers’ IC. The efficacy of the XGBoost model was assessed by applying it to three real-life construction projects. Findings: Upon application of the model to the case studies, we made the following conclusions: (1) “Leadership Style,” “Listening,” “Team Building” and “Clarifying Expectations” emerged as significant skills and (2) the model accurately predicted workers’ IC quality in over 78% of the cases. This algorithm has the potential to preempt interpersonal conflicts, enhancing job-site productivity, team development and human resources management. Furthermore, it can guide construction managers in designing IPS training programs. Originality/value: This study contributes to the existing knowledge by addressing the crucial connection between IPS and IC quality in construction projects. Additionally, our novel approach, integrating fuzzy logic and XGBoost, provides a valuable tool for IC prediction. By identifying significant IPS and offering predictive insights, this research facilitates improved communication and collaboration in the construction industry, ultimately enhancing project outcomes.