Automating the Task-Level Construction Activity Analysis through Fusion of Real Time Location Sensors and Workers' Thoracic Posture Data

Knowledge of workforce productivity and activity is crucial for determining whether a construction project can be accomplished on time and within budget. As a result, significant work has been done on improving and assessing productivity and activity at task, project, and industry levels. Task level productivity and activity analysis are used extensively within the construction industry for various purposes, including cost estimating, claim evaluation, and day-to-day project management. Nevertheless, assessment of task level productivity and activity analyses are mostly performed through visual observations and "after the fact" even though studies have been performed to automatically translate the construction operations data into productivity information and to provide spatial information of construction resources for specific construction operations. This paper presents an original approach to automatically assess construction labor activity. Using data fusion of spatiotemporal and workers' thoracic posture data, the authors have developed a novel framework for identifying and understanding the worker's activity type over time automatically. This information is used to perform automatic work sampling that is expected to facilitate real-time productivity assessment.