- 7 views
- 7 downloads
Machine Learning and Text Mining: A New Approach to Determine the Weather Effects on Construction Incidents
-
- Author / Creator
- Atsegbua, Joshua
-
In the construction industry, safety standards are not only a priority but a necessity due to the dynamic nature of the field. Workplace safety is a complex issue influenced by a variety of factors that are constantly evolving. Each incident has the potential to impact the industry's intricate structure, causing project delays, and more importantly, direct impact on human lives. The adverse effects of workplace incidents reverberate not only within the construction companies themselves but also on a national and global scale. To enhance safety in the construction industry, this thesis explores two crucial aspects - the influence of weather conditions and the interplay of worker demographics. These factors are integral in fortifying this vital sector and contributing to a safer and more resilient industry
In the first study, there was a comprehensive examination of the influence of weather conditions on the frequency of incidents. By utilizing advanced machine learning methods, predictive models, including Random forest, Decision trees and K-Nearest Neighbors were constructed with high accuracy. Particularly, the Random Forest model demonstrated superior performance with an accuracy of 97%. In addition to model creation, an executable application was developed to enable stakeholders to conduct real-time risk assessment. This innovation has the potential to facilitate proactive incident management in situations where weather conditions are constantly evolving.
In the second study, a detailed analysis of the interaction between worker demographics and incident rates was examined. By employing root cause analysis, significant factors contributing to incidents, such as insufficient training, inadequate hazard identification, and ambiguous operating procedures, were identified. Additionally, the utilization of time series analysis further enhanced the understanding of incident rates by uncovering dynamic fluctuations among different age groups, occupational categories and experience levels. This comprehension of the influence of demographic variables on workplace safety establishes a solid foundation for the development of effective strategies to mitigate risks in areas with high incident rates
Bringing these two studies together, we recognize the pivotal role that comprehensive safety management plays in addressing workplace incidents. While the first study emphasizes the significance of predictive models in managing incident risks related to weather conditions, the second underscores the intricate relationship between worker demographics and safety. Together, they forge a comprehensive approach to proactive safety management in high-incident zones, ultimately aiming to make such environments safer and more resilient.
This revised structure encapsulates both studies within a unified thesis, maintaining the distinctiveness of each while emphasizing the collective pursuit of enhanced workplace safety. -
- Graduation date
- Spring 2024
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.