Article
Mathematics
Francisco Javier Diez, Manuel Arias, Jorge Perez-Martin, Manuel Luque
Summary: OpenMarkov is an open-source software tool designed for probabilistic graphical models, primarily in medicine but also used in other fields and education in over 30 countries. This paper explains how OpenMarkov can be used as a pedagogical tool to teach the main concepts of Bayesian networks and influence diagrams, as well as various inference algorithms.
Article
Engineering, Industrial
Rafael Ballester-Ripoll, Manuele Leonelli
Summary: This paper shows how to apply Sobol's method of global sensitivity analysis to measure the influence of a set of nodes' evidence on a quantity of interest in a Bayesian network. The proposed method exploits the network structure to transform the problem of Sobol index estimation and gives exact results when exact inference is used. It also handles correlated inputs efficiently as long as eliminating the inputs' ancestors is computationally affordable.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Review
Thermodynamics
Tingting Li, Yang Zhao, Ke Yan, Kai Zhou, Chaobo Zhang, Xuejun Zhang
Summary: Probabilistic graphical models are effective in addressing various issues in energy systems, with static models handling incomplete or uncertain information and dynamic models accurately predicting energy consumption, occupancy, and failures. A unified framework combining knowledge-driven and data-driven PGMs is suggested for better performance, with the need for universal PGM-based approaches adaptable to different energy systems and hybrid algorithms integrating advanced techniques for improved results.
BUILDING SIMULATION
(2022)
Article
Computer Science, Artificial Intelligence
Manuel Gomez-Olmedo, Rafael Cabanas, Andres Cano, Serafin Moral, Ofelia P. Retamero
Summary: This study introduces a new structure called value-based potentials (VBPs) to efficiently represent quantitative information in probabilistic graphical models. VBPs leverage repeated values to reduce memory requirements and outperform probability trees (PTs) by overcoming certain limitations.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2021)
Article
Mathematics
Pedro Bonilla-Nadal, Andres Cano, Manuel Gomez-Olmedo, Serafin Moral, Ofelia Paula Retamero
Summary: The computerization of tasks generates a large amount of data, leading to the development of machine-learning methods to extract useful information for decision-making processes. Fields like medicine and education are interested in obtaining relevant information from this data, but the complex nature of the problems requires efficient techniques and approximation methods to handle the high degree of interdependency between variables.
Article
Engineering, Electrical & Electronic
Yuxuan Yuan, Kaveh Dehghanpour, Zhaoyu Wang, Fankun Bu
Summary: In this paper, a multi-source data fusion approach is proposed to locate outage events in partially observable distribution systems using Bayesian networks. The approach takes into account diverse evidence and the complex structure of the distribution systems, which can effectively reduce the computational complexity of outage location inference. This method enhances the resilience of power distribution systems by accurately locating outages in high-dimensional spaces.
IEEE TRANSACTIONS ON SMART GRID
(2022)
Article
Computer Science, Information Systems
Isabel Haasler, Rahul Singh, Qinsheng Zhang, Johan Karlsson, Yongxin Chen
Summary: In this study, a connection between multi-marginal optimal transport problems and probabilistic graphical models is pointed out, showing that an entropy regularized multi-marginal optimal transport is equivalent to a Bayesian marginal inference problem. This relation extends both optimal transport and probabilistic graphical model theories while enabling fast algorithms for multi-marginal optimal transport through leveraging Bayesian inference algorithms. Several numerical examples are provided to illustrate the results.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2021)
Article
Computer Science, Theory & Methods
Rico Krueger, Michel Bierlaire, Thomas Gasos, Prateek Bansal
Summary: This study analyzes two robust alternatives to the multinomial probit model and demonstrates their advantages through simulation and case studies.
STATISTICS AND COMPUTING
(2023)
Article
Computer Science, Information Systems
Luigi Riso, Marco Guerzoni
Summary: This paper addresses the issue of concept-drift in machine learning in high dimensional problems. It introduces a method that assesses concept drift as an independent characteristic of dataset evolution, regardless of the specific target variable. This method is particularly useful in business environments where the same dataset is used for multiple classifications. Unlike previous methods, it does not require re-testing of each new model, saving computational resources.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Interdisciplinary Applications
Xiaochen Chou, Luca Maria Gambardella, Roberto Montemanni
Summary: The Orienteering Problem involves selecting a subset of customers to visit within a time budget to maximize revenue. The Probabilistic Orienteering Problem introduces randomness by considering customer visits based on probabilities, making it more challenging. Tabu Search, a method for escaping local optima, is utilized in this study to solve the problem using a Monte Carlo evaluator.
COMPUTERS & OPERATIONS RESEARCH
(2021)
Article
Economics
Murat Genc, Stephen Knowles, Trudy Sullivan
Summary: People generally prefer donating to charities helping people in need in their own country rather than in developing countries; many prioritize where the donation will be spent over how effective it will be; donation decisions are often guided by emotions or intuition rather than rational calculation.
Article
Computer Science, Information Systems
Simon Streicher, Johan A. Du Preez
Summary: Probabilistic graphical models (PGMs) are powerful tools for solving complex relationship systems, with tree-structured PGMs providing efficient and exact solutions. However, inference on graph structures may not always find the optimal solutions. To address this, the purge-and-merge algorithm was developed to nudge graph structures towards tree structures iteratively.
Article
Multidisciplinary Sciences
Celia C. Beron, Shay Q. Neufeld, Scott W. Linderman, Bernardo L. Sabatini
Summary: In probabilistic and nonstationary environments, mice use internal and external cues to make decisions. The behavior of mice in a task with time-varying reward probabilities is both deterministic and stochastic. Modeling their behavior through equivalent models reveals that mice achieve near-maximal reward rates.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Article
Economics
Seong Ok Lyu
Summary: The study reveals that South Korean sport tourists have varying degrees of importance for different attributes of Winter Olympic travel products, and they are willing to make trade-offs among these attributes to purchase the optimal product. They place high willingness-to-pay values for attending skating competitions and popular tournaments, while also showing strong interests in sightseeing and recreational experiences.
Article
Economics
Filipe Rodrigues
Summary: This study proposes an amortized variational inference approach that utilizes stochastic backpropagation, automatic differentiation, and GPU-accelerated computation to enable Bayesian inference in mixed multinomial logit models on large datasets. Furthermore, it demonstrates how normalizing flows can enhance the flexibility of variational posterior approximations. Simulation and real data analysis show that this approach achieves significant computational speedups without compromising estimation accuracy on large datasets.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2022)
Article
Transportation
Mohammad Hossein Chinaei, Taha Hossein Rashidi, Travis Waller
Summary: This paper introduces a distributed architecture for a MaaS digital network using blockchain technology, which integrates multiple service providers to deliver multimodal mobility services. The paper proposes crypto-tickets as a means of service ownership and evaluates the efficacy of the blockchain-based ownership scheme against traditional membership plans. Finally, the potentials of blockchain in personalizing service ownership, congestion management, and data trading between stakeholders of a comprehensive MaaS platform are discussed.
TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH
(2023)
Article
Transportation
Chi Xie, Jue Hou, Ti Zhang, Travis Waller, Xiqun Chen
Summary: This paper examines the impact of electricity-charging prices on the routing choice and flow pattern of electric vehicles in a home-workplace commute traffic network. It proposes a convex programming model to characterize commuters' charging behavior under different prices and designs a path-based solution algorithm to solve this problem.
TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH
(2023)
Article
Transportation
Chi Xie, Yanjie Wan, Min Xu, Xiqun Chen, Travis Waller
Summary: This article reinvestigates the mathematical formulations of traffic assignment problems with perception stochasticity and demand elasticity, proposing a pair of dual general formulations. The equilibrium or optimality conditions of these problems can be redefined as a combination of equations, and the solutions of the primal and dual formulations have been proven to be equivalent and unique. Algorithmic analysis and numerical tests suggest that the dual formulation-based algorithm, the Cauchy algorithm, is more suitable for large-scale problems and converges faster than the primal formulation-based Frank-Wolfe algorithm.
TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH
(2023)
Article
Computer Science, Interdisciplinary Applications
Haoning Xi, Yili Tang, S. Travis Waller, Amer Shalaby
Summary: Mobility-as-a-Service (MaaS) is an emerging business model that integrates various travel modes into a single on-demand mobility service. This study proposes a MaaS ecosystem that provides both mobility and instant delivery services by sharing the same multimodal transport system. A bilaterial surcharge-reward scheme (BSRS) is introduced to manage the integrated mobility and delivery demand, and a solution algorithm is developed to optimize the system equilibrium costs.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2023)
Article
Engineering, Multidisciplinary
Hongzhi Lin, Min Xu, Chi Xie
Summary: This research aims to present a methodology for designing a network of preventive healthcare facilities in order to prevent diseases early. By formulating the problem as a bilevel non-linear integer programming model and solving it using a genetic algorithm and a method of successive averages, the study divides the problem into facility location and capacity planning and user choice, and applies it to an illustrative case in the Sioux Falls transport network.
JOURNAL OF INDUSTRIAL AND MANAGEMENT OPTIMIZATION
(2023)
Article
Transportation Science & Technology
Ali Najmi, Taha H. Rashidi, Travis Waller
Summary: This paper proposes a generalized multi-modal multi-provider market equilibrium model to evaluate the operation of transport systems. The model includes various modes of transportation such as private vehicles, walking, public transport, ride-sourcing, and ridesharing. The economic behaviors of service providers and a network operator are modeled using optimization problems and user equilibrium conditions, forming a complementarity formulation for the market at equilibrium. Extensive computational experiments demonstrate the model's applicability in handling various market responses to technological improvements, demand changes, emissions restrictions, and cultural barriers.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Construction & Building Technology
Zixuan Chen, Ahmed W. A. Hammad, Steven Travis Waller, Assed Naked Haddad
Summary: This paper presents a novel multi-objective mixed integer linear programming model that considers the selection of suitable suppliers, inventory management practices, and order quantities to optimize the trade-off between procurement cost and material delay impacts. The proposed model treats material prices, supplier capacities, and delays as fuzzy scenario-based parameters and is validated through numerical testing. The sensitivity analysis shows the importance of accurate estimation for uncertain parameters. The paper also demonstrates the higher performance of the proposed model compared to deterministic market conditions.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Economics
Haoning Xi, Wei Liu, S. Travis Waller, David A. Hensher, Philip Kilby, David Rey
Summary: In the context of Mobility-as-a-Service (MaaS), the transportation sector is shifting towards user-centric business models that prioritize user experience and customized mobility solutions. This study proposes an auction-based mechanism and optimization models for the demand-side management of MaaS systems. The mechanism allows users to bid for mobility services based on their willingness to pay and experience-related preferences, and the optimization models aim to maximize social welfare by optimally allocating mobility resources in real-time. Extensive simulations using realistic mobility data demonstrate the benefits of the proposed mechanism.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2023)
Article
Environmental Studies
Bangyang Wei, Bo Du, Meead Saberi, S. Travis Waller, Wei Liu
Summary: This study investigates the strategy of allocating road space as parking for electric ride-sourcing vehicles (ERVs) to reduce cruising. The results show that providing parking increases ride-sourcing demand, reduces charging demand, and increases profit and social welfare.
TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT
(2023)
Article
Engineering, Civil
S. Travis Waller, Moeid Qurashi, Anna Sotnikova, Lavina Karva, Sai Chand
Summary: This paper examines the impact of the ongoing invasion in Ukraine on traffic behavior using analytics and zonal-based network models. The data-driven approach utilizes obtained travel-time conditions within an evolutionary algorithm framework to infer origin-destination demand values in an automated process. The novelty lies in the analysis to ensure the appropriateness of obtained data, the direct linkage of the analysis to the timeline of disruptions, and the identification of specific network metrics for conceptualizing the impact of conflict disruptions on traffic network conditions.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Engineering, Civil
Navid Kalantari, Hamid Mirzahossein, Pooyan Najafi, Travis Waller, Xiang Zhang
Summary: Although demand management has been proven to be effective in managing congestion, many organizations still focus on network modification and capacity increase as congestion relief measures. The network design problem plays a significant role in shaping urban transportation networks. This study proposes an efficient solution algorithm for the continuous network design problem and shows that it can solve the problem in a shorter time with high accuracy.
TRANSPORTATION RESEARCH RECORD
(2023)
Review
Transportation
Can Li, Lei Bai, Lina Yao, S. Travis Waller, Wei Liu
Summary: Transportation is crucial for the economy and urban development, but it faces challenges in terms of efficiency, sustainability, resilience, and intelligence. Reinforcement Learning (RL) has emerged as a useful approach for smart transportation applications, allowing autonomous decision-makers to learn from experiences and make optimal actions in complex environments. This paper conducts a bibliometric analysis to understand the development of RL-based methods in transportation applications and provides a comprehensive literature review on the specific topics. Future research directions for RL applications and developments are also discussed.
TRANSPORTMETRICA B-TRANSPORT DYNAMICS
(2023)
Article
Economics
Peng-Cheng Xu, Qing-Chang Lu, Chi Xie, Taesu Cheong
Summary: This study investigates the resilience evaluation of interdependent networks. A model is developed to quantify the impacts of network interdependency on the resilience of interdependent transit networks, considering interdependency relations, network topology, flow characteristics, and demand distribution. The model is applied to the metro and bus networks of Xi'an, China. Results show that node degree heterogeneity in topology, bidirectional function dependency among networks, and flow matching between networks are important factors influencing network resilience.
TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE
(2024)