Journal of Transportation Engineering, Part A: Systems

Railway networks are one of the major parts of the transportation infrastructure. Development of this network by the construction of new lines or the capacity improvement of existing lines needs time and capital cost. Decision makers, who are responsible for the infrastructure network, always have a limited budget. They need to select the best package from the long list of new projects. Providing an optimal periodic plan for investment in the railway network is a major task for the government. In this paper, a solution for this task is developed by using a multiperiod network design model for the railway network. The formulation of the model and some of the concepts are new. The model considers development projects (new line construction and existing line improvement), available budget in each period, origin–destination demand matrix for each period, block capacity, and technical capacity. The suggested model is implemented for the Iranian railway network and is solved by an exact method that shows efficiency of the model. In addition, based on the nature of demand, different scenarios are considered in the proposed model. The selected projects by the proposed model and their usage percentage show the ability and efficiency of the model.

Traffic flow data are needed for traffic management and control applications as well as for transportation planning issues. Such data are usually collected from traffic sensors; however, it is not practical or even feasible to deploy traffic sensors on all of a network’s links. Instead, it is necessary to extend the information acquired from a subset of link flows to estimate the entire network’s traffic flow. To this end, this study proposes a robust deep learning architecture based on a stacked sparse autoencoders (SAEs) model for a precise estimation of the whole network’s traffic flow with an already-deployed sensor set. The proposed deep learning architecture has two consequent components: a deep learning model based on the SAEs and a fully connected layer. First, the SAEs model is used to extract traffic flow features and reach a meaningful pattern of the relation between the traffic flow data and network structure. Subsequently, the fully connected layer is used for the traffic flow estimation. Then, the whole architecture is fine-tuned to update its parameters in order to enhance the traffic flow estimation. For training the proposed deep learning architecture, synthetic link flow data are randomly generated from the network’s prior demand information. The performance of the proposed model is evaluated then validated using two real networks. A third medium real-size network is used to measure the robustness of applying the proposed methodology to this specific problem of traffic flow estimation.

With the increases in both the urban rail transit (URT) network scale and complexity, a route choice model that was previously developed may not function properly anymore and therefore must be constantly evaluated if possible and updated whenever necessary. This paper develops and uses a posterior approach that fuses multisource data from both the automatic fare collection (AFC) and automatic train supervision (ATS) systems to provide accurate and intelligent evaluation of route choice models, especially for large-scale complex URT networks. A method to rebuild passengers’ journey one by one is put forward that makes the proposed approach work in a more disaggregate manner. Then, observed travel time (OTT), and simulated travel time (STT), which are deduced by fusing multisource data from AFC and ATS systems, are defined. Instead of using traditional manual-based methods, the evaluation of route choice models is conducted by comparing and testing the distributions of both OTTs and STTs, and two nonparametric statistical techniques (NPSTs) are adopted. Pilot case studies are conducted on the Beijing subway network and the results obtained clearly show that the approach can disaggregately evaluate the route choice model and can also be easily incorporated into an automatic evaluation procedure.

Information on the quality of ballasted track is normally collected from a track measurement vehicle operating on a monthly basis or otherwise periodically. Track deterioration in terms of alignment and vertical levels is normally predicted by time-series data collected up to a certain point, and subsequent maintenance work is undertaken based on the predetermined maintenance level of the track geometry classified according to its importance. In this regard, deterioration of track geometry based on time-series measurement data can be efficiently modeled by an online support vector regression (OSVR) scheme, and detailed investigation has been carried out to improve the previous work on batch-type prediction models of track geometry proposed by the authors. For such purposes, an incremental support vector regression (ISVR) model based on a Bayesian optimization scheme as well as an OSVR model are introduced in this paper, and the prediction results are compared with those obtained by a conventional machine learning model. The results show that the accuracy of the proposed model increases by approximately 20% compared with that of the conventional model, and the outcome can be applied to the optimal scheduling of track maintenance work.

Control delay, the most important measure of effectiveness at signalized intersections, is commonly estimated using analytical models due to the difficulty of field measurements. In analytical models, the uniform component of control delay is derived from an established ratio between control and stopped delays. The Highway Capacity Manual (HCM) delay model utilizes a unique stopped-to-control-delay ($D_s/D_c$) ratio of 0.76. In this paper, the modeling relationship between control and stopped delays is based on empirical data collected on 28 intersection approaches totaling 1,200 individual trajectories with stopped delay. Authors applied analysis on two different levels based on individual vehicle trajectories (microapproach) and on traffic flow (macroapproach). At the micro-level, the authors found a strong linear relationship between control and stopped delays, which exclusively depends on approach speed. It was also found that the average deceleration-acceleration delay of 12.8 s is greater than values cited in the literature. At the macro-level, it was found that the $D_s/D_c$ ratio varies from 0.49 to 0.86 depending on red time and approach speed. The authors concluded that unique $D_s/D_c$ ratio applied in the HCM delay model is not always appropriate for accurate delay estimation and suggested a modified model.

The allowable bending fatigue stress in a rail will determine the maximum allowable speed, which is then checked against the current operating speed. Therefore, some sensible value of allowable bending fatigue stress must be agreed upon. This paper establishes a relationship between reliability and bending fatigue stress. According to the American Railway Engineering and Maintenance-of-Way Association (AREMA), the recommended allowable bending fatigue stress is $124\mathrm{N}/{\mathrm{mm}}^2$ (18,000 psi). However, this recommendation is associated with low reliability and does not match the AREMA-recommended fatigue test load, which implies a stress of $97\mathrm{N}/{\mathrm{mm}}^2$ (14,042 psi). AREMA should review its recommended value of the allowable bending fatigue stress and fatigue test load. This paper explores the basis of the fatigue test span and formulates a fatigue test load. Additionally, the author suggests the upper limit of ${\mathrm{P}}_2$ load in this paper. The concerns associated with the heavy haul operation are discussed with suggestive measures to reduce rail/weld fracture failure rate. Therefore, this paper could be helpful to compute allowable speed, to assess reliability of rail traffic against fatigue failure of rail/weld, to compute fatigue test load and to limit the ${\mathrm{P}}_2$ load.