Multivariate Statistical Model for Predicting Occurrence and Location of Broken Rails

Abstract

Broken rails are the leading cause of major accidents on U.S. railroads and frequently cause delays. A multivariate statistical model was developed to improve the prediction of broken-rail incidences (i.e., service failures). Improving the prediction of conditions that cause broken rails can assist railroads in allocating inspection, detection, and preventive resources more efficiently, to enhance safety, reduce the risk of hazardous materials transportation, improve service quality, and maximize rail assets. The service failure prediction model (SFPM) uses a combination of engineering and traffic data commonly recorded by major railroads. A Burlington Northern Santa Fe Railway database was developed in which the locations of approximately 1,800 service failures over 2 years were recorded. The data on each location were supplemented with information on other engineering and traffic volume parameters. A complementary database with the same parameters was developed for a randomly selected set of locations at which service failures had not occurred. The combined databases were analyzed using multivariate statistical methods to identify the variables and their combinations most strongly correlated with service failures. SFPM accuracy in predicting service failures at specific locations exceeded 85%. Although further validation is necessary, SFPM is promising in the quantitative prediction of broken rails, thereby improving a railroad’s ability to manage its assets and risks.

Characteristics, Cause, and Severity Analysis for Hazmat Transportation Risk Management

“…Many methods for analyzing the relationships among the causes that may lead to accidents and the corresponding possible consequences are available, which can be categorized as statistic regression models (i.e., multivariate regression and multinomial logit) [1,[14][15][16][17][18] and machine learning methods (typical as Bayesian network and decision tree) [19][20][21][22][23]. Desai et al [1] presented a sequence of statistical models (i.e., partial least squares, spline regression, and Box-Cox transformations) to estimate the population affected and the impact of the cost incurred as a result of hazmat accidents.…”

Forecasting Risk of Service Failures Between Successive Rail Inspections: A Data-Driven Approach

“…These incidents are defined as service failures. Broken rails, as a major type of service failure, are a leading cause of derailments (Dick et al 2003). Broken rails cause an average cost of $525,000 per incident due to the damage of track and equipment (Schafer and Barkan 2008b).…”

Automatic Prediction of Maintenance Intervention Types in Roads using Machine Learning and Historical Records

“…road pavement conditions, geometry integrity, etc). During the last decade, an escalating number of literature pieces were published on applying ML to linear infrastructures (27,28,29,30). In all cases, the common factor for this progression lays on the increasing availability of data captured from auscultation/monitoring activities and campaigns; thus, ML techniques have promoted the concept of learning from data, facilitating the extraction of patterns and trends by "let the data speak by themselves".…”