Article
Engineering, Civil
Tomoharu Iwata, Hitoshi Shimizu, Naoki Marumo
Summary: We propose a probabilistic model of pedestrian behavior that estimates population at each road using observed populations and routes. The model incorporates pedestrian dependence on road congestion and derives transition probabilities between roads using a Gaussian distribution. Parameters are estimated through gradient-based optimization methods, and experiments show accurate estimation of road populations.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Mathematics
Weerawat Sudsutad, Nantapat Jarasthitikulchai, Chatthai Thaiprayoon, Jutarat Kongson, Jehad Alzabut
Summary: This study investigates novel estimations involving the expectation, variance, and moment functions of continuous random variables using a generalized proportional fractional integral operator, and presents interesting results of the proportional fractional expectation, variance, and moment functions.
Article
Multidisciplinary Sciences
Gregor Betz, Kyle Richardson
Summary: It is argued that suitably trained neural language models exhibit key properties of epistemic agency and can revise their degrees of belief rationally. Computational experiments with rankers show that pretrained models can gradually obtain a probabilistically coherent belief system aligned with logical constraints through self-training on auto-generated texts. This self-training also plays a pivotal role in rational evidential learning, enabling rankers to adjust individual degrees of belief based on novel evidence. These findings confirm the Rationality Hypothesis and suggest its broader empirical, normative, and conceptual implications beyond the original practical applications of NLMs.
Article
Engineering, Environmental
Tianpeng Wang, Shihao Xiao, Jie Zhang, Baocheng Zuo
Summary: This study validates the depth consistency assumption of V-s data and proposes two probabilistic models for liquefaction potential assessment based on this assumption. The performances of the suggested models are found to be superior to the Chinese code model and comparable to frequently used models.
BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
(2022)
Article
Environmental Sciences
Yifu Gao, Abdullah Sahin, Jasper A. Vrugt
Summary: Variance-based analysis is used to quantify the sensitivity of the output y to the input variables x. This paper focuses on the sensitivity analysis of correlated input variables using high-dimensional model representation (HDMR) to separate the structural and correlative contributions. The results show that HDMR and HDMRext successfully analyze the structural and correlative contributions of the model output and provide an optimal experimental design for parameter correlation.
WATER RESOURCES RESEARCH
(2023)
Article
Green & Sustainable Science & Technology
Andrea Pompigna, Raffaele Mauro
Summary: This paper introduces the methods of measuring traffic quality and congestion level in highway engineering and discusses in depth the application of reliability approach. By simulating and estimating the ARIMA models of speed random processes, combined with the Product Limit Method, it is possible to assess traffic reliability and congestion probability.
Article
Mathematics
Huimin Xiao, Shouwen Wu, Chunsheng Cui
Summary: This paper presents a novel consistency analysis model to check and improve the consistency of probabilistic linguistic preference relations (PLPRs). The proposed method includes a new consistency index and checking method based on similarity measure, as well as an optimization model based on minimum adjustment. A numerical example is provided to validate the feasibility of the proposed method, and a comparative analysis with other approaches is conducted. The conclusion is drawn that the proposed consistency analysis model outperforms previous methods in terms of determining adjustment parameter, adjustment cost, and retention of original preferences.
Article
Computer Science, Artificial Intelligence
Tijin Yan, Tong Zhou, Yufeng Zhan, Yuanqing Xia
Summary: With the development of AIoT, data-driven attack detection methods for cyber-physical systems (CPSs) have gained attention. This paper proposes a general framework called TFDPM which uses energy-based generative models and graph neural networks to address issues in data distribution and correlation modeling. The framework extracts temporal and feature patterns, utilizes a conditional diffusion probabilistic model for prediction, and introduces a conditional noise scheduling network for real-time detection. Experimental results demonstrate its superior performance compared to existing methods.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Information Systems
Adam Marszalek, Tadeusz Burczynski
Summary: This paper explores the concept of ordered fuzzy numbers and fuzzy random variables to develop a method for constructing fuzzy stochastic time series models, which can estimate parameters using classical equations.
INFORMATION SCIENCES
(2021)
Article
Mathematics, Applied
J-C Cortes, A. Navarro-Quiles, J- Romero, M-D Rosello
Summary: This study probabilistically solves random initial value problems for non-homogeneous first-order linear differential equations with complex coefficients by computing the first probability density of the solution. The probability of stability and the density of the equilibrium point are explicitly determined. The Random Variable Transformation technique is extensively utilized for the overall analysis.
Article
Engineering, Multidisciplinary
Peiliang Xu, Yun Shi
Summary: This paper proves two major results with regard to full EIV models: first, under certain conditions, a full EIV model is unlikely to be rank deficient; second, even if a full EIV model is theoretically rank deficient, it is unlikely to determine its rank deficiency through measurements.
Article
Computer Science, Interdisciplinary Applications
Matieyendou Lamboni
Summary: This paper derives practical dependency functions for classical multivariate distributions, which are useful for uncertainty quantification, sensitivity analysis, and simulation of random variables. A method for selecting efficient sampling functions using multivariate sensitivity analysis is provided, and the approach is illustrated through numerical simulations.
MATHEMATICS AND COMPUTERS IN SIMULATION
(2022)
Article
Computer Science, Artificial Intelligence
Andrija Petrovic, Sandro Radovanovic, Mladen Nikolic, Boris Delibasic, Milos Jovanovic
Summary: In this paper, a hybrid soft computing model of two Gaussian conditional random field models was proposed for the inference of traffic speed in large-scale networks. The model addresses the sparsity and spatial dependence of traffic state variables by combining a binary classification model and a regression model. The proposed model was tested on real-world networks in Serbia and showed better prediction performance than other models.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Information Systems
Serafin Moral, Andres Cano, Manuel Gomez-Olmedo
Summary: This paper proposes a generalization of the imprecise probability model, allowing probability intervals to be represented in a hierarchy of sets with a tree structure. It also shows how this model can represent other models, such as possibility measures and generalized p-boxes. Additionally, the paper demonstrates that the resulting model is always an order-2 capacity and that operations like coherence checking, computing the natural extension, and conditioning can be performed efficiently.
INFORMATION SCIENCES
(2023)
Article
Environmental Sciences
Feifei Zheng, Junyi Chen, Holger R. Maier, Hoshin Gupta
Summary: In this paper, a novel strategy is proposed to improve the performance of physics-based models of dynamical systems by using continuous simulation and deterministic data allocation. The strategy addresses the challenge of ensuring distributional similarity in partitioning data into independent subsets. The results of testing on rainfall-runoff models demonstrate that the proposed strategy consistently outperforms the traditional approach, especially under conditions of larger runoff skewness.
WATER RESOURCES RESEARCH
(2022)
Correction
Engineering, Mechanical
S. Blason, A. Fernandez-Canteli, C. Rodriguez, E. Castillo
INTERNATIONAL JOURNAL OF FATIGUE
(2021)
Article
Mechanics
A. Alvarez-Vazquez, M. Muniz-Calvente, A. Fernandez-Canteli, M. J. Lamela, E. Castillo
Summary: This work introduces a novel methodology for characterizing the fracture behavior of metallic notched components, taking into account the notch root radius and temperature effects based on the brittle-to-ductile transition curve. Two regression models are derived, one considering temperature as an influencing variable and the other combining temperature with the notch radius effect. The proposed methodology proves to be applicable and suitable through its application to experimental results on a S355J2 steel with varying temperatures and notch root radii conditions.
ENGINEERING FRACTURE MECHANICS
(2021)
Article
Transportation Science & Technology
A. U. Z. Patwary, Wei Huang, Hong K. Lo
Summary: This paper introduces a metamodel-based simulation optimization framework for calibrating an agent-based multimodal traffic microsimulator, which combines a simplified multimodal traffic model with the metamodel to enhance the efficiency and adaptability of the calibration process.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Chemistry, Physical
Angel De La Rosa, Gonzalo Ruiz, Enrique Castillo, Rodrigo Moreno
Summary: This study presents a new probabilistic approach for calculating dynamic viscosity using the Krieger-Dougherty equation. By utilizing Bayesian analysis, parametric-deterministic models are transformed into parametric-probabilistic models, resulting in significantly improved viscosity calculations compared to theoretical values.
Article
Construction & Building Technology
Omar Kammouh, Maria Nogal, Ruud Binnekamp, A. R. M. (Rogier) Wolfert
Summary: Probabilistic Monte Carlo simulations are commonly used to determine project completion times at specific probability levels. However, schedule changes can negatively impact the probability of timely completion, necessitating a manual trial and error approach to find effective mitigation measures.
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT
(2021)
Review
Engineering, Civil
Yue Shang, Maria Nogal, Haoyu Wang, A. R. M. (Rogier) Wolfert
Summary: Performance evaluation and maintenance planning are becoming increasingly important for ageing rail infrastructure and increasing demand for track safety and continuous availability. Discrete railway assets, such as bridges and level crossings, along with extended track sections, comprise the main railway network infrastructure. These assets have significant implications for train safety, riding comfort, operating expenditures, and effective network capacity. The heterogeneity of asset features and operating environments also presents challenges for efficient maintenance planning. This review focuses on level crossings and synthesizes different perspectives on their maintenance management. The integration of a systems thinking approach and two levels of asset management (micro- and macro-level) provide a structured synthesis that is interdependent and synergistic. The review also compares mechanistic and data-driven modeling approaches and identifies limitations in existing studies, with directions for future research aimed at improving the inspection and diagnosis process and moving towards a system-level maintenance approach for multiple level crossings.
STRUCTURE AND INFRASTRUCTURE ENGINEERING
(2023)
Article
Transportation
Gege Jiang, Shuling Wang, Hong K. Lo, Zheng Liang
Summary: This study introduces the importance of valuing travel time reliability and defines the concept of variation cost. The proposed single-step tolling scheme aims to reduce cost variability and queuing cost. Through the framework, the study explores the combined impact of cost variability, capacity degradation, and road pricing on travelers' departure profile and the overall system generalized cost.
TRANSPORTMETRICA B-TRANSPORT DYNAMICS
(2022)
Article
Engineering, Industrial
M. Nogal, A. Nogal
Summary: This paper introduces a simple method for conducting sensitivity analyses for extreme-based engineering problems without the need to define the probability distribution of input factors, making it convenient for practitioners to use.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Operations Research & Management Science
Enoch Lee, Xuekai Cen, Hong K. Lo, Ka Fai Ng
Summary: By considering spatial and volume stochastic variations of passenger demands, a zonal-based flexible bus service (ZBFBS) is developed with a two-stage stochastic program to minimize operating costs. Reliability requirements are introduced to ensure service quality and the problem is separated into two phases for efficient solutions.
TRANSPORTATION SCIENCE
(2021)
Article
Chemistry, Multidisciplinary
Javier Panadero, Majsa Ammouriova, Angel A. Juan, Alba Agustin, Maria Nogal, Carles Serrat
Summary: In smart cities, unmanned aerial vehicles and self-driving vehicles need to utilize advanced technologies to plan routes and serve customers, as well as real-time coordination in exceptional cases. This paper proposes using team orienteering problem to plan vehicle routes and introduces an 'agile' optimization algorithm.
APPLIED SCIENCES-BASEL
(2021)
Article
Environmental Sciences
M. N. Legasa, R. Manzanas, A. Calvino, J. M. Gutierrez
Summary: The study demonstrates that AP-RFs can better predict the precipitation intensity and gamma distribution parameters on wet days compared to traditional random forests, showing higher practicality. This methodology proposed in this paper has substantial potential for hydrologists and other impact communities in need of reliable climate information.
WATER RESOURCES RESEARCH
(2022)
Article
Chemistry, Physical
Angel De La Rosa, Gonzalo Ruiz, Enrique Castillo, Rodrigo Moreno
Summary: This article develops a probabilistic approach to calculate the dynamic viscosity in self-compacting steel-fiber reinforced concrete. The Bayesian analysis is used to obtain the probability density functions of the parameters, resulting in improved results compared to traditional methods.
Article
Engineering, Civil
Lubing Li, Wei Huang, Andy H. F. Chow, Hong K. Lo
Summary: This study develops a cell-based two-stage stochastic program to address the dynamic, spatial, and stochastic characteristics of traffic flow for arterial adaptive signal control. By incorporating the concept of Phase Clearance Reliability (PCR) and using a gradient-based solution algorithm, the study enhances solution efficiency and validates findings through VISSIM. Results show the importance of capturing dynamic, spatial, and stochastic features for traffic control in order to avoid delay performance degradation.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Operations Research & Management Science
A. U. Z. Patwary, Shuling Wang, Hong K. Lo
Summary: This paper proposes an efficient gradient estimation method called Iterative Back-propagation (IB) for network optimization problems with an embedded static traffic assignment (TA) model. The method exploits the iterative structure of the TA solution process to calculate the gradients simultaneously, without requiring additional function evaluations, and scales well with higher dimensions. Application of the method to the OD estimation problem in the Hong Kong multimodal network demonstrates its effectiveness in reducing errors and matching link counts.
TRANSPORTATION SCIENCE
(2023)
Article
Engineering, Civil
Enoch Lee, Hong K. Lo, Manzi Li
Summary: This paper optimizes the zoning, scheduling, and pricing of a zonal-based flexible bus service (ZBFBS) by considering dynamic stochastic demand volume, ride request locations, and time window constraints. Two approaches are proposed: the demand-responsive method and the stochastic programming approach. The results show the benefit of the bi-level framework in producing an optimal zonal design and the superiority of the bi-level stochastic zonal design framework over the demand-responsive method in a tight planning time.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
