Federal Railroad Administration

Ballast serves a critical role in railway track stability by supporting rails and crossties, distributing loads, and facilitating drainage. However, its structural integrity can be compromised due to fouling and moisture infiltration, leading to reduced shear strength and stiffness. While geogrid stabilization has been widely recognized for improving ballast performance, its effectiveness under different moisture conditions remains less understood. The objective of this study is to evaluate the effect of moisture content and geogrid stabilization on the dynamic characteristics of fouled railway ballast using Bender Element (BE) sensor technology. Laboratory box tests were performed on ballast specimens with a fouling index (FI) of 25.3, incorporating gravimetric moisture contents ranging from 0% to 12% by weight of fines smaller than 9.5 mm (3/8 in.). Shear wave velocities (Vs) were measured using BE sensors embedded within the ballast layer to estimate small-strain shear modulus (Gmax) characteristics. Geogrid-stabilized specimens were compared to non-stabilized control cases under both dry and wet states. The test results showed that (Vs) and (Gmax) increased for up to 6% moisture and then decreased at higher moisture levels, indicating a transition from stabilizing to lubricating effects of added moisture on the ballast fines. Under both dry and wet conditions, geogrid-stabilized ballast samples exhibited higher shear wave velocities and stiffness values than the control. These findings demonstrate the value of field-installed BE sensors for evaluating ballast layer stiffness and monitoring the performance improvement provided by geogrid stabilization of track substructure under varying moisture conditions.

This study investigates injury severity at rural intersections using machine learning-based classification models applied to crash data obtained from Tennessee’s Enhanced Roadway Information Management System (ETRIMS). Injury severity was categorized into three levels: Property Damage Only, Possible Injury & Suspected Minor Injury, and Suspected Serious Injury & Fatal. Multiple supervised machine learning models were developed, including Gradient Boosting, Categorical Boost, Decision Tree, Support Vector Machine, and Multi-Layer Perceptron. Class imbalance was addressed using Synthetic Minority Over-sampling Technique (SMOTE). Among all models, XG-Boost demonstrated superior performance, achieving an overall classification accuracy of 66.78% and a macro F1-score of 0.37. SHAP analysis was employed to interpret the contribution of individual features across severity classes. The results consistently identified AADT, DHV, percentage of trucks, manner of collision, weather conditions, and land use as dominant predictors influencing injury severity. Specifically for Class 2 Serious and Fatal injuries, elevated traffic volumes, adverse weather conditions, and head-on or angle collisions were strongly associated with higher severity outcomes. These findings highlight the importance of traffic volume management, truck regulation, intersection geometric improvements, and weather-responsive control strategies in mitigating severe crash consequences at rural intersections. Future work should focus on integrating larger datasets, additional roadway geometric features, hybrid machine learning architectures, and cost-sensitive optimization to improve model sensitivity toward rare but critical injury severity classes.

The Action Plan for Rail Energy and Emissions Innovation proposes actions to reduce and nearly eliminate emissions in the U.S. rail sector, in line with the U.S. economy-wide goal of net-zero greenhouse gas (GHG) emissions by 2050. It also proposes actions to leverage the rail system to reduce emissions from other modes. The national goal of achieving a zero-emission freight system by 2050 draws our attention to the fact that freight transport cannot be addressed simply mode by mode, but it should instead be treated as an interdependent system. This is especially true when pursuing decarbonization. This action plan presents how both rail transport and decarbonization intersect with our national transportation decarbonization blueprint, the decarbonization of the freight system, and national transmission goals. The intended audience of this report is the stakeholders who will advance rail decarbonization in a just and economical way by propelling the suite of actions listed here. This includes government at all levels, rail companies, locomotive manufacturers, labor unions, Amtrak, and more.

The purpose of this study is to examine the association between mental health, fear of COVID-19, and job insecurity among bus service workers during the second wave of the coronavirus disease 2019 (COVID-19) pandemic in Japan. In September 2020, 1,889 employees of three private railway bus companies in the Tokai region were administered an anonymous self-questionnaire to be filled out at their workplace. The survey items were depression, anxiety, fear of COVID-19, job insecurity, demographics, and work characteristics. Logistic regression analysis was used to calculate the odds ratio and 95% confidence interval. Overall, 1,316 valid responses, excluding those from workers with a history of mental illness and others, were included in the analysis (valid response rate: 69.7%). Multiple logistic regression analysis revealed that the prevalence of depression and anxiety were significantly higher in groups with medium and high levels of fear of COVID-19 or job insecurity than in the group with a low level of fear of COVID-19 or job insecurity. Fear of COVID-19 and job insecurity were found to be risk factors for depression and anxiety among bus service workers during the second wave of the COVID-19 pandemic, the latter being the stronger risk factor. Hence, comprehensive COVID-19 prevention measures as well as mental health-related measures to reduce job insecurity should be implemented in the bus service industry.

Strain gauges are often used to measure vertical wheel loads in a railroad track. This approach is based on the concept of differential-shear-strain (DSS) measurement: The difference in vertical shear force between two points along a beam equals the magnitude of the resultant of applied vertical forces in between. With a slight modification to the strain-gauge positions and installation of an additional set of strain gauges, this concept can be extended to measure the vertical rail–tie interface reaction forces, thus assessing the tie support conditions. Although the application of DSS measurements for vertical-wheel-load quantification is widely prevalent in the railroad community, the validity of this approach for tie reaction measurement has been relatively unexplored. Conceptually, the approach is similar to the vertical-wheel-load measurement system, with the only difference being the placement of the strain gauges along the rail. Nevertheless, several questions have been raised about how different track and loading configurations can affect the accuracy of such a measurement system. To address some of these concerns and establish this approach as a viable method for tie support condition assessment, a field-validation effort was undertaken. Under the scope of this research effort, the strain-gauge-based DSS measurement system for rail–tie interface reaction-force measurement was evaluated in the field under static as well as dynamic loading. The study showed that the strain-gauge-based measurement approach is as accurate as other conventional methods of tie reaction-force measurement.

Abstract Freight rail networks serve a key role in transporting bulk goods to accommodate changing market demands and to serve public needs. Network analyses of such systems can provide important insights into enhancing transportation efficiency and system resilience. This paper develops and investigates a topological analysis model for network efficiency, which is associated with the connectedness of a network's nodes by its links and their corresponding network attributes. This model allows analyzing network topologies with or without assigned weights to their nodes and links based on different attributes. Key attributes may include physical length of links, commodity types and volume of goods transported through links, and dwell-time at nodes, and origination termination volumes of goods, types of goods moved, and origins and destination of goods. The model presented here enables (1) defining distinctions that may be employed for the assignment of node and link weights, (2) gaining an understanding of node and link criticality, and (3) providing methods for objectively maintaining and enhancing network performance. Such analyses can inform rail managers and executives in planning expansions, route or freight changes, or preparations for potential node or link failures. A case study of an aggregate U.S. freight rail network along with other example topologies is presented to demonstrate the use of selected network attributes and their influence on connectedness efficiency and the impacts of node and link failures on the overall transport efficiency.

This report provides evaluations of the NPP site infrastructure and near-site transportation infrastructure for removing spent nuclear from 16 (NPP) sites. The material to be removed from the NPP sites includes both the spent nuclear fuel (SNF) and the greater-than-Class C low level radioactive waste (GTCC waste) that is stored, or will be stored, at the sites.

Abstract Cut spikes have been used to hold rails directly to wood ties and thus provide lateral restraint to rails and maintain rail gage. For tightly curved and/or steeply graded wood tie tracks, particularly those undergoing heavy loads, elastic clips have been increasingly used to fasten rails securely to tie plates, which in turn are held down by cut spikes or lag screws to the crossties. However, several recent derailment accidents were attributed to the breakage of significant numbers of spikes used in association with the elastic fastening systems. A commonly observed failure mode is the development of fatigue cracking across spike surfaces located approximately 1.5 inches below the top surface of the tie. Previous Finite Element (FE) studies have shown that the presence of longitudinal force transmitted into the tie-spike interface is significant to accelerate wood damage and cause spike failure initiation as the strength of wood along the track longitudinal direction is weaker than the other directions. In the field, the magnitude of longitudinal load between a tie and fastener is very difficult to measure relying on the current sensor technology. A literature review conducted so far indicates that longitudinal-load-distribution related computational models are very limited. Therefore, the distribution mechanism of longitudinal load has not been well-established. To better understand how the longitudinal load is transferred from wheels to ties and fasteners, this paper developed an analytical model to investigate the longitudinal tie forces using the finite difference method. In this paper an analytical model is firstly introduced to calculate the longitudinal track response under a single static wheel load. The solution calculates diverse sets of responses such as: (1) rail axial displacements; (2) rail axial forces; (3) force distribution to the tie; and (4) tie forces. The effect of multiple axle loads from a train is then computed using superposition theory. To validate the model, a field test was performed at the Horseshoe Curve in Altoona, Pennsylvania to investigate longitudinal rail forces and longitudinal displacements of rail and ties under train passages. The model predictions were compared and validated with the test results. This model applies to all train types on various track layouts. Key factors, e.g. train braking versus train traction, fastener stiffness, etc., that have a significant effect on longitudinal force distributions are presented in the paper. The model estimates the longitudinal forces during uniform train braking and traction events, and shows that locomotives in traction lead to more significant longitudinal forces distributed to ties. The results indicate the longitudinal force was distributed over a large span (or influence zone) of over one hundred ties whereas the vertical tie load was only distributed over three to five ties. Within the influence zone of the longitudinal load, the vertical tie load ranged dramatically from maximum compressive forces (negative forces) to uplifting tensile forces (positive forces), indicating load configuration was very tie-dependent under a train passage. Significant longitudinal forces could be existing on uplifted ties. The results also show that larger longitudinal stiffness of fastener led to higher maximum forces distributed to ties and smaller influence zones.

Recent research has shown a relationship between track geometry defects and track subsurface conditions as measured by Ground Penetrating Radar (GPR). This paper presents the results of a comprehensive study looking at the development of a probabilitic model for the prediction of track geometry defects as a function of key subgrade parameters as measured by GPR. Specifically, several Logistical Regression (LR) analyses were performed, to include conventional LR modeling as well as a hybrid LR modeling approach based on Hierarchical Clustering Analysis with Histogram Data. The result was a higher order polynomial Logistic Regression model for determination of the probability of a track geometry surface defect occurring at locations with measured ballast fouling and measured ballast thickness. The results showed that there was a statistically significant relationship between high rates of geometry degradation and poor subsurface condition as defined by the GPR parameters: Ballast Fouling Index (BFI) and Ballast Layer Thickness (BLT). Furthermore, a predictive model was developed to determine the probability of a high rate of geometry degradation as a function of these key GPR parameters.