Informs Annual Meeting 2017

WB50

INFORMS Houston – 2017

WB50

WB51

361C Autonomous Systems Optimization I Invited: TSL, Urban Transportation Invited Session

361D Disruption Management Contributed Session Chair: Churlzu Lim, UNC-Charlotte, Charlotte, NC, United States, clim2@uncc.edu 1 - A Bi-objective Optimization Model for System Resilience Daniel Hernando Romero, University of South Florida, Tampa, FL, 33613, United States, danielromero@mail.usf.edu, Aritra Pal, Alex Savachkin Most system resilience optimization models have a single objective based on a specific resilience metric. There is a need to expand such approaches to offer solutions based on different metrics or dimensions of resilience. We offer a bi- objective model to analyze two main dimensions of system resilience, the absorptive and restorative capacities. The model provides a Pareto front of non- dominated solutions enabling a decision maker to select suitable hardening and recovery strategies. 2 - Vulnerability Analysis of Critical Infrastructures under Disruptions using Monte Carlo Method Chao Fang, Associate Professor, Wuhan University, Economics and Management School, Wuhan, 430072, China, chao.fang@whu.edu.cn Reliable and safe critical infrastructures are essential for the sustainable development of modern societies. Considerable attention has been paid to analyzing and mitigating the vulnerability of critical infrastructures, especially against the increasing occurrence of disruptive events such as man-made or natural disasters. In this paper, we use Monte Carlo method to simulate the disruptions attacking the modelled infrastructure network. The innovation is that new metrics based on flow information are proposed in order to predict the propagation of consequences. The application to China High-Speed Rail illustrate the approach with respect to enhancing the system’s reliability. 3 - Facility Protection Model for the Supply of Emergency and Perishable Products Girish Chandra Dey, IIT.Kharagpur, Qtr No. EM/G-18, IIT.Khargpur Campus, IIT.Khargpur Campus, Kharagpur, 721302, India, deygirish@gmail.com, Mamata Jenamani We present facility protection model considering the importance of maximum distance limits for the supply of emergency products like medical supplies or perishable products during the disruptions. We formulate it as a tri-level mixed integer problem, which considers capacitated facilities, interdiction, and fortification budget limits simultaneously. Then we reformulate it to a bi-level problem and solve it by implicit enumeration (IE) algorithm performed on a binary tree. We report computational results obtained from 64 randomly generated test instances. 4 - Applying Stackelberg-game Model under Credit Guarantee Mechanism with Information Asymmetry Cheng-Feng Wu, Hubei University of Economics, No.8 Yang Qiaohu Road, Jiangxia, Wuhan, 430205, China, wuchengfeng@hbue.edu.cn The strait of capital shortage in small and medium enterprises(SMEs) causes them difficult to fulfill orders. Furthermore, a buyer is concerned about supplier disruption when placing an order with his SME supplier. Therefore, obtaining and operating financial support for SMEs is a significant issue. The supplier thoughtfully allocates its borrowed money to production and technology after buyer promises the financial institution a commitment in credit guarantee. This study tries to develop the model to examine the issue of sourcing model under Stackelberg-game with credit guarantee scheme. 5 - Minimization of Restoration Time for Disrupted Power Distribution Network Churlzu Lim, Associate Professor, UNC-Charlotte, 9201 University This study considers scheduling repair crews to restore power distribution systems that are affected by a high impact low frequency event. Given a set of information pieces including locations of faults, estimated repair times for faults, available resources, and estimated travel times between fault locations, the goal is to assign repair jobs to each repair crew team in a specific sequence that minimizes the system restoration time. Two alternative mixed integer programming formulations as well as a heuristic method are presented and numerically compared in solution qualities and efforts. City Boulevard, Charlotte, NC, 28223-0001, United States, clim2@uncc.edu, Bhavish G. Golla, Badrul Chowdhury, Andrew N.Kling

Chair: Alireza Talebpour, Northwestern University, 600 Foster Street, Transportation Center 3rd Floor, Evanston, IL, 60208, United States, atalebpour@civil.tamu.edu 1 - Platooning in the Presence of a Speed Drop: A Generalized Control Model Sina Arefizadeh, Graduate Student, Texas A&M.University, College Station, TX, 77843, United States, sinaarefizadeh@tamu.edu, Alireza Talebpour, Igor Zelenko Designing an autonomous vehicle platooning strategy for tracking a reference velocity profile can improve travel time reliability. Accordingly, this study introduces a generalized control model to track a desired velocity profile, while ensuring safety in the platoons of autonomous vehicles. Defining appropriate error terms, we changed the tracking velocity profile problem into a stability problem. Considering the error term in tracking the velocity profile coupled with error term associated with keeping safe distance from the leader, the proposed method controls the maximum of the two error terms for each vehicle. We showed that the error terms remain bounded throughout the time and space. Two scenarios were simulated, with and without initial perturbations, and results confirmed the effectiveness of the proposed control model in tracking the speed drop, while ensuring safety and string stability. 2 - Design and Operations of Autonomous Vehicle Car-Sharing Systems (AVCS) Zhiheng Xu, Ph.D, University at Buffalo, Buffalo, NY, 14228, United States, zhihengx@buffalo.edu We present a time-space optimization model for design and operations of future Autonomous Vehicle Car-Sharing Systems (AVCS). In addition to design and operations decisions, the proposed model explicitly considers empty vehicle relocation and demand shift between AVCS and privately owned autonomous vehicles. 3 - Microscopic Traffic Modeling of Collisions with the Incorporation of Connected and Autonomous Vehicles Claire E. Silverstein, George Washington University, Washington, DC, United States, csilvs@email.gwu.edu, Samer H.Hamdar Interactions between individual drivers and error in the decision-making process have been recognized as major sources of unsafe and unstable traffic conditions. Two main avenues that reduce/remove human error from the driving experience are being pursued. Autonomous vehicles utilize sensing technology to take over the driving task, while connected vehicles incorporate vehicle-to-vehicle and vehicle-to-infrastructure communication to obtain and share information, allowing the driver/vehicle to make improved driving choices. This research tests the traffic safety of different market penetration rates (MPRs) of connected and autonomous vehicles on the microscopic level with collision formation. This leads to the optimization of MPRs of the various vehicle types based on safety. The inclusion of lateral trajectory during the lane-changing process allows for more robustness when analyzing vehicular movements and angled collisions. 4 - Collaborative Platooning of Automated Vehicles using Variable Time-Gaps Aria HasanzadeZonuzy, Texas A&M.University, College Station, TX, United States, azonuzy@tamu.edu, Srinivas Shakkottai, Alireza Talebpour, Sina Arefizadeh, Swaroop Darbha Connected automated vehicles could be coordinated to safely attain the maximum traffic flow, such as those engendered by the merger of two strings of vehicles. Strings of vehicles have to be shaped correctly in terms of the inter- vehicular time-gap and velocity to ensure that such operation is feasible. However, controllers that can achieve such traffic shaping over the multiple dimensions of target time-gap and velocity over a region of space are unknown. The objective of this work is to design such a controller, and to show that we can design candidate time-gap and velocity profiles such that it can stabilize the string of vehicles in attaining the target profiles. Our analysis is based on studying the system in the spatial rather than the time domain, which enables us to study stability as in terms of minimizing errors from the target profile and across vehicles as a function of location.

492