21 February 2024 in Journal of Information Technology in Construction
This study investigates the possibility of doing probabilistic forecasting of construction labor productivity metrics for both long-term and short-term estimates. The research aims to evaluate autoregressive forecasting models, which may help decision-makers with information currently unavailable in construction projects. Unlike point forecasts, the proposed method employs probabilistic forecasting, offering additional valuable insights for decision-makers. The distributional information is obtained by updating the moments of the distribution during training. Two datasets are used to evaluate the models: one collected from an entire construction site for long-term forecasting and one from an individual worker for short-term forecasting. The models aim to predict the state of direct work, indirect work, and waste. Several models are trained using different hyperparameters. The models are tuned on the number of trees and the regularization used. The presented method gives estimates of future levels of direct work, indirect work, and waste, which will add value to future processes.
19 June 2025 in CIB Conferences
Work-related musculoskeletal disorders (WMSDs) are among the most frequently reported injuries among industrial workers, primarily due to prolonged exposure to awkward postures. While ergonomists emphasize process and equipment redesign, exoskeletons offer a promising supplementary solution for improving workplace ergonomics. However, the effectiveness of different exoskeleton designs remains understudied. This study employs an experimental approach to evaluate the impact of two passive exoskeletons—(A) HAPO SD and (B) BISKO—on posture and task performance during repetitive lifting activities. Postural assessment was conducted using an AI-driven software that analyzes posture based on the Rapid Upper Limb Assessment (RULA) method. The findings indicate that both exoskeletons influenced posture and performance, albeit with varying effectiveness. Exo A demonstrated more consistent benefits by shifting postures toward moderate-risk categories (RULA scores of 3 to 4) while enabling participants to complete more repetitions and sustain longer task durations compared to both the no-exoskeleton condition and the BISKO exoskeleton. However, statistical analyses did not confirm significant differences in posture improvements between exoskeleton and no-exoskeleton conditions, underscoring the importance of ergonomic training and safe lifting techniques as primary interventions. These findings contribute to the growing body of research on exoskeleton applications in workplace ergonomics, emphasizing the need for careful exoskeleton selection to optimize worker augmentation and mitigate fatigue.
1 May 2023 in Journal of Engineering, Project, and Production Management
The Work Sampling (WS) technique, used worldwide to understand how workers spend their time, represents a time-consuming and costly activity. Therefore, several researchers work on different approaches to automate the data collection using sensor-based and vision-based technologies. The challenge of all the sensor-based approaches is that they do not provide the share of time in different work categories. The lack of knowledge on a possible correlation between Direct Work and, e.g., presence, location, or worker movement represents a gap in the current body of knowledge. Thus, this research aims to understand the correlation between Direct Work as the independent predictor variable; and Movement as the dependent response variable. The authors used the data gathered through the application of WS in five case studies on building renovation projects in Denmark. To explain this correlation. The authors selected a combination of four quantitative techniques: (1) curve estimation; (2) linear regression; (3) ANOVA analysis; and (4) t-test. The correlation of the result is discussed considering three assumptions: (1) the structure of the day; (2) global vs. local; and (3) Movement vs. Transporting and Walking. The result shows a significant correlation between Direct Work and Movement with an average R 2 of 0.328. This is considered acceptable predictability taking the socio-technical system aspect of a construction site into account.