Search for: All records

Abstract This study employs graph mining and spectral clustering to analyze patterns in railway crossing accidents, utilizing a comprehensive dataset from the US Department of Transportation. By constructing a graph of implicit relationships between railway companies based on shared accident localities, we apply spectral clustering to identify distinct clusters of companies with similar accident patterns. This offers nuanced insight into the underlying structure of these incidents. Our results indicate that “Highway User Position” and “Equipment Involved” play pivotal roles in accident clustering, while temporal elements like “Date” and “Time” exert a diminished impact. This research not only sheds light on potential accident causation factors but also sets the stage for subsequent predictive safety analyses. It aims to serve as a cornerstone for future studies that aspire to leverage advanced data-driven techniques for improving railway crossing safety protocols. 

Abstract Tensor decompositions have proven to be effective in analyzing the structure of multidimensional data. However, most of these methods require a key parameter: the number of desired components. In the case of the CANDECOMP/PARAFAC decomposition (CPD), the ideal value for the number of components is known as the canonical rank and greatly affects the quality of the decomposition results. Existing methods use heuristics or Bayesian methods to estimate this value by repeatedly calculating the CPD, making them extremely computationally expensive. In this work, we proposeFRAPPE, the first method to estimate the canonical rank of a tensor without having to compute the CPD. This method is the result of two key ideas. First, it is much cheaper to generate synthetic data with known rank compared to computing the CPD. Second, we can greatly improve the generalization ability and speed of our model by generating synthetic data that matches a given input tensor in terms of size and sparsity. We can then train a specialized single-use regression model on a synthetic set of tensors engineered to match a given input tensor and use that to estimate the canonical rank of the tensor—all without computing the expensive CPD.FRAPPEis over$$24\times $$ $24\times$faster than the best-performing baseline, and exhibits a$$10\%$$ $10\%$improvement in MAPE on a synthetic dataset. It also performs as well as or better than the baselines on real-world datasets. 

Abstract Expanding on the insights from our initial investigation into railway accident patterns, this paper delves deeper into the predictive capabilities of machine learning to forecast potential accident trends in railway crossings. Focusing on critical factors such as “Highway User Position” and “Equipment Involved,” we integrate Kernel Ridge Regression (KRR) models tailored to distinct clusters, as well as a global model for the entire dataset. These models, trained on historical data, discern patterns and correlations that might elude traditional statistical methods. Our findings are compelling: certain clusters, despite limited data points, showcase remarkably Root Mean Squared Error (RMSE) values between predictions and real data, indicating superior model performance. However, certain clusters hint at potential overfitting, given the disparities between model predictions and actual data. Conversely, clusters with vast datasets underperform compared to the global model, suggesting intricate interactions within the data that might challenge the model’s capabilities. The performance nuances across clusters emphasize the value of specialized, cluster-specific models in capturing the intricacies of each dataset segment. This study underscores the efficacy of KRR in predicting future railway crossing incidents, fostering the implementation of data-driven strategies in public safety. 

Abstract In this work, we explore multiplex graph (networks with different types of edges) generation with deep generative models. We discuss some of the challenges associated with multiplex graph generation that make it a more difficult problem than traditional graph generation. We propose TenGAN, the first neural network for multiplex graph generation, which greatly reduces the number of parameters required for multiplex graph generation. We also propose 3 different criteria for evaluating the quality of generated graphs: a graph-attribute-based, a classifier-based, and a tensor-based method. We evaluate its performance on 4 datasets and show that it generally performs better than other existing statistical multiplex graph generative models. We also adapt HGEN, an existing deep generative model for heterogeneous information networks, to work for multiplex graphs and show that our method generally performs better. 

Advanced Air Mobility (AAM) presents an emerging alternative to traditional car driving for commuting in metropolitan areas. However, its feasibility has not been thoroughly studied nor well understood at the operational level. Given that AAM has not been in place, this study explores the economic, energy, and environmental feasibility of AAM for commuting at an early stage of AAM deployment. We propose a time expanded network model to characterize the dynamics of eVTOL operations between a vertiport pair in different states: in-service flying, relocation flying, charging, and parking, while respecting various operational and commuter time window constraints. By jointly considering eVTOL flying with vertiport access and egress and using real-world data, we demonstrate an application of the model in the Chicago metropolitan area in the US. Different vertiport pairs and eVTOL aircraft models are investigated. We find substantial travel time saving if commuting by AAM. While vehicle operating cost will be higher using eVTOLs than using auto, the generalized travel cost will be less for commuters. On the other hand, with current eVTOL power requirement, the energy consumption and CO2 emissions of AAM will be greater than those of auto driving, with an important contributor being the significance presence of empty flights relocation. These findings, along with sensitivity analysis, shed light on future eVTOL development to enhance the competitiveness of AAM as a viable option for commuting. 

This paper investigates deploying connected and automated vehicle (CAV) lanes in transportation networks with a focus on measuring and preserving equity among travelers. A new metric is proposed to characterize equity based on (1) generalized travel cost per unit origin-destination (OD) distance for travelers on each OD pair and using each vehicle type and (2) maximum deviation of the standardized unit generalized travel cost from system average. A bi-level bi-objective program is developed to simultaneously minimize system travel cost and inequity while deploying CAV lanes. A solution algorithm that combines nondominated sorting genetic algorithm II and variable neighborhood search is designed. Through extensive numerical experiments, we find (1) inequity is more prominent when travel demand is high; (2) human-driven vehicle travelers become more disadvantageous with lower CAV price and higher CAV automation; and (3) subsidy is effective in mitigating inequity, but a fee for using CAV lanes is less promising. 

Electric vertical takeoff and landing aircraft (eVTOLs) are gaining growing interest recently. However, limited attention has been paid to the prospect of using eVTOLs for package delivery. To fill this void, this paper explores the attractiveness of eVTOL-based package delivery in terms of cost, energy consumption, and CO2 emissions. Given that eVTOLs cannot take off/land at customer doorsteps, a two-leg system design is proposed and formulated as an optimization model. To implement the model, we consider multiple plausible eVTOL and ground vehicle types, their cost economics, and energy use and CO2 emission characteristics. Applying the model in the Chicago metro region, we find that the attractiveness of eVTOL-based package delivery depends critically on the eVTOL and ground vehicle types. With an appropriate eVTOL-ground vehicle combination, eVTOL-based delivery can be attractive compared to van-only delivery in terms of total shipping cost, but not necessarily so from the energy and mission perspectives. This highlights the need for future R&D to further enhance the energy efficiency of eVTOLs. When designing eVTOL-based package delivery systems, the importance to account for the potential interactions between eVTOL traffic and commercial air traffic should also be recognized. 

Electronic information and optical properties coupled with the Quantum Theory of Atoms in Molecules (QTAIM) and Electron Localization Function (ELF) analyses are used to elucidate the erbium (Er+3) and praseodymium (Pr+3) intraband f–f transitions in the lithium tantalate (LiTaO3) doped and co-doped configurations and the metal-oxygen bonding. The generalized gradient approximation calculations show that the Er+3- and Pr+3-4f bands appear closer to the conduction band bottom for Er+3 and Pr+3 at the Li sites and to the valance band top for Er+3 at the Ta sites. However, the corresponding hybrid functional calculations for the dopants at the Li site show that the Er+3 and Pr+3-4f bands spread in energy, which agrees with the observed intraband f–f transitions from the optical properties calculations. QTAIM shows that Ta-, Er+3-, and Pr+3-O bonding is incipient covalent for all configurations of this work. The absence of ELF in the metal-O regions aligns with QTAIM on the lack of strong covalent bonding in these compounds. This complementary insight highlights how weakly interacting metal-O atoms lead to delocalized electron density, a feature that influences the physical, electronic, and chemical behavior of the LiTaO3. 

This paper addresses the distributed tracking control of multiple uncertain high-order nonlinear systems with prescribed performance requirements. By introducing a performance function and a nonlinear transformation, the prescribed fixed-time performance tracking control problem is reformulated as a distributed tracking control problem for multiple special nonlinear systems. With the aid of the universal approximation theorem for continuous functions and algebraic graph theory, distributed robust adaptive controllers are designed using the backstepping technique. Simulation results are presented to demonstrate the effectiveness of the proposed algorithms.