2015 Informs Annual Meeting

WA69

INFORMS Philadelphia – 2015

3 - Constructing Activity-mobility Patterns of Students Base on UB Card Transactions Negin Ebadi, University at Buffalo, 326 Bell Hall, Buffalo, NY, 14226, United States of America, negineba@buffalo.edu, Nicholas Delgobbo, Kwangil Suh, Jee Eun Kang This paper proposed two algorithms to construct activity-mobility patterns of students in University at Buffalo based on their UB card transactions. A combination of two different measures of error has been used to evaluate the performance of the proposed algorithms.

It seems that there has been a rash of serious and horrific railroad crashes and derailments in the US in this year that has claimed the lives of more than 14 people and injured hundreds. Authorities have issued a new rule on tank cars and renewed the call for speeding up the installation of the PTC systems on Class I railroads. However, PTC is designed to prevent a specific group of accidents and over reliance on this technology could potentially impair the overall safety of the system. 2 - Managing Rail Transportation for Hazardous Materials Ginger Ke, Assistant Professor, Memorial University of Newfoundland, Faculty of Business Administration, St. John’s, NL, A1A4J9, Canada, gke@mun.ca, Kan Fang, Manish Verma This paper studies the routing and scheduling of rail shipments of hazardous materials with due dates. In particular, we minimize the weighted sum of earliness and tardiness for each demand plus the holding cost at each yard, while forcing a risk threshold on each service leg at any time instant. A mixed-integer programming model and two heuristic-based solution methods are proposed for preparing the shipment plan. Numerical experiments are conducted to provide managerial insights. 3 - Potential Areas Affected by a Liquid Hazardous Material Release Jose Manuel Martin Ramos, University of Illinois (RailTEC), 205 N. Mathews Ave., Urbana, IL, United States of America, mrtnrms2@illinois.edu, Mohd Rapik Saat The increase of crude oil traffic by rail in the U.S. and several severe release incidents highlight the need to further improve railroad transportation safety. Accurate estimation of the potential consequences of a release incident is a key element in risk assessment. This paper describes a specific methodology for evaluating more precisely the consequences of liquid hazardous releases. The model is developed using GIS software allowing anybody to easily implement it in time-efficient manner. 4 - A General Solution for Rail Yard Simulation with Conflicting Routes Yuan Wang, SWJTU, 111 Erhuan Road Bei Yi Duan, Chengdu, China, yuan_wang2@hotmail.com, Gongyuan Lu The biggest challenge in the yard simulation is to handle route conflicts of trains if their moving has shared tracks. The traditional agent-based simulation is usually very complicated and can hardly describe a lot of details. We will present a general event-based solution which can simplify the modeling process tremendously for all kinds of yards. A successful use case from the largest high- speed train station in Asian will be presented. Chair: Xiaolei Guo, Associate Professor, University of Windsor, 834 Hacienda Court, Odette School of Business, Windsor, Canada, guoxl@uwindsor.ca 1 - Using Cumulative Prospect Theory to Model Unexpected Driver Response to Congestion Pricing Christopher Gaffney, Drexel University, 3141 Chestnut St, Philadelphia, PA, 19104, United States of America, cgaffney@stevens.edu, Michael Pennock Dynamic congestion pricing is commonly used to reduce demand on roads. In practice, stated preference surveys are used to gauge drivers’ value of time and set toll prices, while utility theory is often used for theoretical modeling. Evidence from Minneapolis and San Diego has shown the unexpected result that demand for tolled lanes increases with toll price. To account for this, we argue that an alternative measure of driver response to tolling, such as prospect theory, is needed. 2 - A New Quantitative Measure of Network Resilience Hongcheng Liu, Pennsylvania State University, 240 Leonhard Building, State College, PA, United States of America, hql5143liu@gmail.com, Afrooz Ansaripour, Terry Friesz, Wenjing Song, Haipeng Wang, Yiou Wang, Tao Yao, Zhaohu Fan We propose a new quantitative measure of network resilience, called performance centrality, which generalizes the notion of betweeness centrality to explicitly capture not only topology but also distribution of demand and traffic controls. Numerical experiments indicate that performance centrality is a useful resilience index that provides new insights regarding how well traditional network resilience may be integrated with road network operational characteristics. WA71 71-Room 202B, CC Traffic Management Contributed Session

WA69 69-Room 201C, CC

Connected and Autonomous Vehicles II Sponsor: TSL/Intelligent Transportation Systems (ITS) Sponsored Session Chair: Henry Liu, University of Michigan College of Engineering, Ann Arbor, MI, henryliu@umich.edu 1 - Envy-free Intersection Control with Value of Time Heterogeneity Roger Lloret-Batlle, rlloretb@uci.edu, R. Jayakrishnan Current traffic signal systems do not incorporate heterogeneity in the value of time of drivers. Including this information can improve intersection efficiency. We do this by allocating the green times according to a travel cost minimization rule, and then evaluate and discuss different payment schemes with different incentive and fairness properties. 2 - Optimizing Reservation-based Intersections for System Efficiency Michael Levin, University of Texas at Austin, Ernest Cockrell, Jr. Hall (ECJ) 6.204, 301 E. Dean Keeton St. Stop C1761, Austin, TX, 78712, United States of America, michaellevin@utexas.edu, Stephen Boyles Reservation-based intersection controls for autonomous vehicles (AVs) have been demonstrated to reduce delays beyond optimized traffic signals. The question we address is how to optimize the reservation controls themselves. We present an integer programming (IP) formulation for reservation controls in dynamic traffic assignment and consider delay and energy objectives. We use a heuristic to solve the IP on a city network and obtain significant improvements under user equilibrium behavior. 3 - A Cooperative Active Collision Avoidance (CACA) Model for Connected Vehicles Henry Liu, University of Michigan, Ann Arbor, MI, United States of American, henryliu@umich.edu, Xuan Di, Guizhen Yu The emergence of the connected vehicle technology, which enables real-time V2V/V2I, holds the potential to enhance traffic safety applications such as vehicle collision avoidance. In this paper, we develop a novel cooperative active collision avoidance system for connected vehicles that can either avoid rear-end crashes or reduce rear-end crash severity by applying deceleration to the following vehicle and acceleration to its leading vehicles 4 - Auction-based Ridesharing with Pick-up and Drop-off Time Window Alireza Khani, Assistant Professor, University of Minnesota, St. Paul, MN, United States of America, akhani@umn.edu, Stephen Boyles An auction-based ridesharing system is proposed in which users place bids for guaranteed door-to-door service with time constraints, and service provider selects a subset of bids to maximize the profit given operational cost in a time- dependent network. The problem is formulated as a MIP model and solution methods are investigated. WA70 70-Room 202A, CC Recent Academic Research in Railway Applications Sponsor: Railway Applications Sponsored Session Chair: Jeremiah Dirnberger, Manager-Network Modeling & Analytics, CSX, 500 Water St, J300, Jacksonville, FL, 32202, United States of America, jeremiah_dirnberger@csx.com 1 - Integrated Systems Management Framework to Analyze the Critical Role of PTC in Rail Safety Yalda Khashe, USC, University Park Campus, Los Angeles, CA, 90089-0193, United States of America, khashe@usc.edu, Najmedin Meshkati

396