Informs 2015

425 / 552

INFORMS Philadelphia – 2015
423
WB69
69-Room 201C, CC
ITS in Public Transportation
Sponsor: TSL/Intelligent Transportation Systems (ITS)
Sponsored Session
Co-Chair: Alireza Khani, Assistant Professor, University of Minnesota,
United States of America,
1 - Vehicle-Sharing Network Design and its Integration with Public
Transportation Services
Tianli Zhou, Massachusetts Institute of Technology, Cambridge,
MA, United States of America, ,
Virot Chiraphadhanakul, Cynthia Barnhart, Carolina Osorio
We consider the design of a one-way vehicle sharing (VS) system, such that it is
complementary with an existing public transportation system. We propose a two-
stage MIP formulation to address a VS network design problem. We consider a
large-scale case study of the metropolitan area of Boston. To solve the model, we
decompose it and develop a cut generation method. We present numerical results
for the Boston case study.
2 - A Location and Scheduling Model for a Flexible Intercity
Transit Service
Andisheh Ranjbari, University of Arizona, Tuscon, AZ, United
States of America, Mark Hickman,
Yi-Chang Chiu
”Flexpress” is a high-speed intercity bus transit service that has several terminal
locations in the urban area and a dynamic schedule. Travelers have the choice to
board/alight the vehicle at one of the designated terminals, or pay a premium cost
and ask for a door-to-door service. This presentation focuses on the mathematical
model for the terminal location and scheduling problems.
3 - Transit Passenger Flow Prediction under Event Occurrences with
Social Media Data
Ming Ni, SUNY Buffalo, 326 Bell Hall, University at Buffalo,
Amherst, NY, 14260, United States of America,
, Jing Gao, Qing He
Frist, we propose a hashtag-based method to identify events near NYC subway
stations based on tweets data. Second, the relationship between social activities
and subway passenger flow is unveiled. Third, a convex optimization based model
is developed to predict subway volume with social media data.
4 - Reliable Routing in Schedule-based Transit Networks with
Stochastic Travel Times
Alireza Khani, Assistant Professor, University of Minnesota,
United States of America, , Stephen Boyles
In schedule-based transit networks where service is time-dependent and
stochastic, risk-averse users try to minimize the expected travel time as well as to
maximize the chance of arriving on-time. The latter objective is modeled by
transfer failure probability and path algorithms are developed to find the most
reliable paths.
WB70
70-Room 202A, CC
Vehicle Routing I
Contributed Session
Chair: Sadegh Mirshekarian, PhD Student, Ohio University,
285 Stocker Center, Athens, OH, 45701, United States of America,
1 - Branch-and-cut Algorithms for the Time Constrained Covering
Salesman Problem
Gizem Ozbaygin, PhD Candidate, Bilkent University, Department
of Industrial Engineering, Ankara, Turkey,
Hande Yaman, Oya E. Karasan
In this study, we consider the time constrained maximal covering salesman
problem which is a generalization of the orienteering problem. We propose a
mathematical formulation, valid inequalities and branch-and-cut algorithms for
the problem. We test our approaches on several instances we generated based on
some existing VRP instances and report the results of our computational study.
2 - A Fuzzy Vehicle Routing Problem with Time Windows under
Driving and Working Time Restrictions
Can Celikbilek, Ohio University, Industrial & Systems
Engineering, Athens, OH, United States of America,
, Gursel Suer
A new Fuzzy mathematical model is developed to maximize the total profit of the
system while minimizing the total traveled distance. The developed mathematical
model is tested considering real driving, service and working time restrictions. A
single depot and multiple customers with their working time windows are
considered with identical set of vehicles. The developed fuzzy mixed integer
mathematical model provided promising solutions to provide insights to real
industry problem.
3 - Time Dependent Vehicle Routing Problem Solving Based on
Speed Profiles and Speed Drop Coefficients
Martina Ravlic, Faculty of Transport and Traffic Sciences,
University of Zagreb, Vukeliceva 4, Zagreb, Croatia,
, Tomislav Erdelic, Tonci Caric
The vehicle routing problem in real cases can be solved by applying a TDVRP
algorithm. We compared two existing methods for determining the minimum
travel time in TDVRP algorithm: the first method uses speed profiles of traversed
links, while the other uses speed drop coefficients on a wider road network area.
We analyzed the relationship between the execution times, depending on the size
of the problem, and the computed minimum travel times with measured times.
4 - A Vehicle Routing Problem with Budget and Time Constraints
Elham Kookhahi, Wichita State University, 1845 Fairmount
Street, Wichita, KS, 67260, United States of America,
Bayram Yildirim
In this paper, a mathematical model is presented for a vehicle routing problem in
which a sub tour of cities can be visited to maximize the number of served
customers with a limited budget and time. The problem is solved using a genetic
algorithm and numerical results are presented.
5 - A Generalized Single-Depot Vehicle Routing Problem with Time
Windows and Non-Identical Vehicles
Sadegh Mirshekarian, PhD Student, Ohio University, 285 Stocker
Center, Athens, OH, 45701, United States of America,
Gursel Suer, Can Celikbilek
A generalized variant of VRP with time windows is studied, considering vehicles
different in capacity, cost and speed. Vehicles can be used for more than one
route, but are subject to driving and working time constraints. A new genetic
algorithm with specialized crossover and mutation operators is developed and
used to solve the problem, and results are compared with a math model and with
state-of-the-art. The developed GA performed well in terms of solution quality
and convergence speed.
WB71
71-Room 202B, CC
Transportation Operations I
Contributed Session
Chair: Yi Liao, Southwestern University of Finance and Economics,
Liutai Ave 555, School of Business Administration, Chengdu, China,
1 - Conflict Prevention and Detection for Autonomous and
Connected Vehicles
Xin Chen, Assistant Professor, Southern Illinois University,
P.O. Box 1805, Edwardsville, IL, 62034, United States of America,
Shimon Y. Nof
An autonomous vehicle is capable of sensing its environment and navigating
without human input. Conflicts between vehicles and between vehicles and
transportation infrastructure are unavoidable. In this research, the authors
develop a common language structure to represent domain knowledge in
autonomous vehicles and apply algorithms and protocols to detect and prevent
conflicts.
2 - Effective Vehicle Identification and Authentication Mechanism
for Intelligent Transportation System
Joonsang Baek, Khalifa University, Al Saada St. and Muroor Rd.,
Abu Dhabi, United Arab Emirates, ,
Young-ji Byon
Inspired by Internet of Things (IoT), it is possible to develop an identification
mechanism for ITS, which is based on the unique vehicle identification number
(VIN). Such identification mechanism would provide a certain level of privacy,
which will assist traffic monitoring without compromising drivers’ identities by
utilizing a strong authentication mechanism based on message authentication
codes.
WB71