Task–Taxon–Task Framework for Modeling and Predicting the Cognitive Impact of Collaborative Robots on Worker Performance in Modular Construction

Human-robot collaboration (HRC) is transforming modular construction (MC) by improving productivity, efficiency, and safety. While extensive research has focused on the technical design, programming, and implementation of collaborative robots (cobots) in MC, their cognitive impacts on human workers remain largely unexplored. Neglecting these human-centered factors can lead to tangible operational and ethical risks. This study aims to develop and validate an interpretable and scalable framework for modeling and predicting the cognitive impacts of cobots on human workers. The proposed framework integrates cognitive task analysis, a customized task–taxon–task (T3) methodology, and linear mixed-effects models (LMMs) to systematically quantify and predict the worker performance changes induced by cobots. To validate its applicability, a practical protocol was developed and tested through an experimental case study involving a wooden wall panel manufacturing task. The results demonstrate the developed LMM achieved an average predictive accuracy of 87.5%, confirming the framework’s effectiveness in quantifying cobot-induced cognitive impacts. The findings highlight attention and spatial perception as key cognitive demands in HRC and identify spatial proximity as a significant factor influencing cognitive load. This research contributes to the body of knowledge by introducing an interpretable, cognitively based modeling framework for predicting human performance in HRC settings, demonstrating its potential for generalization across diverse prefabrication and modular tasks. The study offers practical insights for construction professionals through a validated, structured protocol that supports worker-centered task design, performance forecasting, and safe robot integration in MC environments. These findings establish a theoretical foundation for cognition-aware HRC task scheduling and dynamic assessments, supporting the broader adoption of robotics in the construction industry.