Search

With the increasing availability of location‐aware devices, passively collected big GPS trajectory data offer new opportunities for analyzing human mobility. Processing big GPS trajectory data, especially extracting information from billions of trajectory points and assigning information to corresponding road segments in road networks, is a challenging but necessary task for researchers to take full advantage of big data. In this research, we propose an Apache Spark and Sedona‐based computing framework that is capable of estimating traffic speeds for statewide road networks from GPS trajectory data. Taking advantage of spatial resilient distributed datasets supported by Sedona, the framework provides high computing efficiency while using affordable computing resources for map matching and waypoint gap filling. Using a mobility dataset of 126 million trajectory points collected in California, and a road network inclusive of all road types, we computed hourly speed estimates for approximately 600,000 segments across the state. Comparing speed estimates for freeway segments with speed limits, our speed estimates showed that speeding on freeways occurred mostly during the nighttime, while analysis of travel on residential roads showed that speeds were relatively stable over the 24‐h period. [ABSTRACT FROM AUTHOR]

Spatial prediction tasks are challenging when observed samples are sparse and prediction samples are abundant. Gaussian processes (GPs) are commonly used in spatial prediction tasks and have the advantage of measuring the uncertainty of the interpolation result. However, as the sample size increases, GPs suffer from significant overhead. Standard neural networks (NNs) provide a powerful and scalable solution for modeling spatial data, but they often overfit small sample data. Based on conditional neural processes (CNPs), which combine the advantages of GPs and NNs, we propose a new framework called Spatial Multi-Attention Conditional Neural Processes (SMACNPs) for spatial small sample prediction tasks. SMACNPs are a modular model that can predict targets by employing different attention mechanisms to extract relevant information from different forms of sample data. The task representation is inferred by measuring the spatial correlation contained in different sample points and the relationship contained in attribute variables, respectively. The distribution of the target variable is predicted by GPs parameterized by NNs. SMACNPs allow us to obtain accurate predictions of the target value while quantifying the prediction uncertainty. Experiments on spatial prediction tasks on simulated and real-world datasets demonstrate that this framework flexibly incorporates spatial context and correlation into the model, achieving state-of-the-art results in spatial small sample prediction tasks in terms of both predictive performance and reliability. For example, on the California housing dataset, our method reduces MAE by 8% and MSE by 7% compared to the second-best method. In addition, a spatiotemporal prediction task to forecast traffic speed further confirms the effectiveness and generality of our method. [ABSTRACT FROM AUTHOR]

A great challenge for intelligent transportation systems (ITS) is missing traffic data. Traffic data are input from various transportation applications. In the past few decades, several methods for traffic temporal data imputation have been proposed. A key issue is that temporal information collected by neighbor detectors can make traffic missing data imputation more accurate. This review analyzes traffic temporal data imputation methods. Research methods, missing patterns, assumptions, imputation styles, application conditions, limitations, and public datasets are reviewed. Then, five representative methods are tested under different missing patterns and missing ratios. California performance measurement system (PeMS) data including traffic volume and speed are selected to conduct the test. Probabilistic principal component analysis performs the best under the most conditions. [ABSTRACT FROM AUTHOR]

To investigate the impact of changes in pavement roughness on speed, this study developed a linear regression model to estimate free-flow speed on freeways in California. The explanatory variables include lane number, total number of lanes, day of the week, region (district), gasoline price, and pavement roughness as measured by the international roughness index (IRI). Data on the California freeway network from 2000 to 2011 were used to build the model. The results showed that pavement roughness has a very small effect on free-flow speed within the range of this study. For the IRI coverage in this study (90% of the records have an IRI of three or lower and 90% of the records have an IRI change of two or lower), a change of IRI of one m/km (63 in./mi) results in a change of average free-flow speed of approximately 0.48-0.64 km/h(0.3-0.4 mph). This result indicated that making a rough segment of pavement smoother through application of a maintenance or rehabilitation treatment will not result in substantially faster vehicle operating speeds, and therefore the energy and emission reductions gained from the reduced rolling resistance will not be offset by increased fuel consumption from faster speeds. However, developing a good model to predict the free-flow speed was not fully successful. The southern California interstate freeway model yielded the best result with an adjusted R-squared of 0.72. For the rest of the regions in the state, the selected explanatory variables can only explain approximately half of the total variance, indicating there are still other variables with more substantial effects on free-flow speed that were not covered in this study. [ABSTRACT FROM AUTHOR]

This study develops the first-of-its-kind seismic fragility models for estimation of approach backfill differential settlement for statewide bridges in California. Seismic compression analysis is carried out through a multi-step framework that estimates seismic-induced shear strain profiles for backfills, converts them to volumetric strains, and computes soil settlement through depth integration. As a crucial step, three independent methods are advanced on a uniform basis to estimate soil shear strains under earthquake loading: a simplified one-dimensional (1D) Static approach, a 1D Dynamic method from site response analysis, and a two-dimensional (2D) Dynamic method that accounts for the trapezoidal shape of the bridge embankment. Subsequently, seismic fragility models are derived using the multiple-stripe analysis (MSA) approach that convolves the seismic demands of backfill settlement with capacity models for several broad groupings of approach-slab designs having different abutment and abutment-foundation types, and abutment-connection details. A survey of California's bridge inventory provides stochastic data inputs that quantify various sources of uncertainties in backfill profile, embankment geometry, and both approach slab and abutment designs. Finally, a combined fragility model is developed by considering an equal methodological contribution from each mentioned approach. The final fragility models suggest that the lowest differential-settlement fragility is achieved by combining regular abutments on spread footings with long-length approach slabs connected to the bridge abutment, while the worst case occurs where bridge abutments on deep foundations are not connected with abutting pavements. The proposed seismic fragility models offer a sound basis for first-order estimation of approach-fill differential settlement which may lead to reduced ride quality, traffic speed reductions, and in extreme cases, temporary roadway closure. These represent one of many seismic fragility models for the full range of bridge components needed for prioritizing seismic retrofit measures, facilitating post-hazard inspections and repair actions, as well as assessing the network mobility for resilience quantification. • Develop seismic fragility models of backfill settlements for California bridges. • Integrate three methods to estimate soil shear strains under earthquake loading. • A survey provides stochastic data that quantify various sources of uncertainties. • The fragility models are sensitive to abutment foundation and approach slab design. [ABSTRACT FROM AUTHOR]

• This paper develops SPFs at different aggregation levels for freeways by using microscopic traffic detector data. • The effects of average speed and speed variation on crash frequencies are revealed at different aggregation levels. • The time-of-day operation status of HOV lanes is introduced to the disaggregated SPFs. • The safety performance of HOV lanes at different aggregation levels are evaluated and compared. • The transferability of SPFs at different aggregation levels between Florida, Virginia, and California is evaluated. Safety Performance Functions (SPFs) have been widely used by researchers and practitioners to conduct roadway safety evaluation. Traditional SPFs are usually developed by using annual average daily traffic (AADT) along with geometric characteristics. However, the high level of aggregation may lead to a failure to capture the temporal variation in traffic characteristics (e.g., traffic volume and speed) and crash frequencies. In this study, SPFs at different aggregation levels were developed based on microscopic traffic detector data from California, Florida, and Virginia. More specifically, five aggregation levels were considered: (1) annual average weekday hourly traffic (AAWDHT), (2) annual average weekend hourly traffic (AAWEHT), (3) annual average weekday peak/off-peak traffic (AAWDPT), (4) annual average day of the week traffic (AADOWT), and (5) annual average daily traffic (AADT). Model estimation results showed that the segment length and volume, as exposure variables, are significant across all the aggregation levels. Average speed is significant with a negative coefficient, and the standard deviation of speed was found to be positively associated with the crash frequency. It is noteworthy that the operation of the high occupancy vehicle (HOV) lanes was found to have a positive effect on crash frequency across all the aggregation levels. The model results also showed that the AAWDPT and AADOWT models consistently performed better (the improvements range from 3.14%–16.20%) than the AADT-based SPF, which implies that the differences between the day of the week and peak/off-peak periods should be considered in the development of crash prediction models. The model transferability results indicated that the SPFs between Florida and Virginia are transferrable, while the models between California and the other two states are not transferrable. [ABSTRACT FROM AUTHOR]