Enhancing the Safety and Reliability of Canadian Railways: Analysis of Highway Railroad Grade Crossing Accidents and Railcar Inspection Technology Assessment

• Author / Creator

  Rana, Parth Jayeshkumar

• The railway industry is one of the major contributors for the transportation of goods and the backbone of Canada’s economy. Safe railway operation is vital for public safety, the environment, and property. Concurrent with climbing amounts of rail traffic on the Canadian rail network are increases in the last decade in the annual accident counts for derailment, collision, and highway railroad grade crossings (HRGCs).
  The development of community areas near railway tracks increases the risk of HRGC accidents between highway vehicles and moving trains, resulting in consequences varying from property damage to injuries and fatalities. Also, the 2018 railway safety act showed concern over increasing trend of HRGC accidents and casualties. Thus, authorities have shown concerns about HRGC improving safety in rail network of Canada.
  Various technologies have been used in the railway industry that improves decision-making, reduces errors, lower costs, save time and keep the safety of railway operation. Transport Canada, in the 2018 railway safety act, highlighted the incorporation of new technological innovations or Canadian railway network to enhance operation and reliability. One of such technological solution used for inspection of railcars is Train Inspection Portal System (TIPS). This system uses multiple camera systems with 360° images of railcars, which are then inspected by remote certified car inspectors (CCIs) and flag any defects/potential defects in the railcars of the trains. This technology is faster and better than manual inspections conducted by CCIs at rail yards.
  The first study is focused on improving HRGC safety by identifying major factors that cause HRGC accidents and affect the severity of associated casualties using ExtraTree classifier method. Combining these causal factors and ensemble algorithms, machine learning (ML) models were developed to analyze HRGC accidents and the severity of associated casualties that occurred between 2001 and 2022 in Canada. Furthermore, spatial autocorrelation and optimized hotspot analysis tools from ArcGIS software were used to identify hotspot locations of HRGC accidents on the railway network.
  The second, third, and fourth studies of my research focus on technology assessment of the Train Inspection Portal System (TIPS). The second study employs a fuzzy-FMEA method, which uses machine learning to account for the imprecision and vagueness of real-life language, to conduct a risk assessment of the TIPS system. The study provides recommendations for reducing the risk of failure by addressing high RPN failure modes and enhancing the overall reliability of the TIPS system.
  In the third study, we assess human factors in remote inspection tasks using the Human Factor Analysis and Classification System (HFACS) framework. The study identifies key HFACS elements that contribute to human errors in remote inspection processes and recommends strategies for reducing these errors and improving the overall quality of remote inspections. The fourth study aims to determine the detectability of rare railcar component defects in a TIPS technology environment and examines the response of remote CCIs. We performed simulations of artificial defects and supported the claim of human factors influence remote inspection performance.
  This research is one of the small contributions to the railway network of Canada. The machine learning models developed to identify causal factors for HRGC accidents and severity of casualties can be used with future data to improve safety strategies. The assessment of POI technology using fuzzy-FMEA has identified high-risk causes of system failure and recommended control measures to improve reliability. Additionally, the artificial defect simulation and human factor assessment have highlighted the need to address rare defect capturing and the impact of human error on remote inspection performance. In summary, this research work has contributed to improving the safety of HRGC railway network and evaluating a futuristic technology that brings efficient, faster, and safer railcar inspection tasks. The findings of this study can improve policy and decision-making for railway safety and inspire future research in this field.

• Subjects / Keywords

  • Canadian rail network
  • Highway railroad grade crossing
  • SMOTE
  • Machine learning
  • Fuzzy-FMEA
  • Reliability
  • ArcGIS

• Graduation date

  Fall 2023

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-0t09-d707

• 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.