2016 INFORMS Annual Meeting Program

90 / 561

INFORMS Nashville – 2016
90
SC61
Cumberland 3- Omni
Fleet Sizing Models
Sponsored: Railway Applications
Sponsored Session
Chair: Dharma Acharya, President, KOSU Services LLC,
241 Auburndale Dr., Ponte Vedra, FL, 32081, United States,
Co-Chair: Michael Gorman, Prof., University of Dayton, Dayton, OH,
12345, United States,
1 - Fleet Sizing Model Overview
Michael Gorman, University of Dayton,
Appropriate fleet sizing is critical for rail asset utilization and service. I will discuss
the basic trade-offs for the fleet sizing decision, discussing both deterministic and
stochastic approaches. I will present rail fleet sizing and other literature, and
discuss approaches taken in practice. Key discussion points surrounding the fleet
sizing decision will be raised.
2 - Fleet Sizing Model Case - Freight Railroad Perspective
Clark Cheng, Senior Director Operations Research, Norfolk
Southern Railway, 1200 Peachtree Street NE, Atlanta, GA, 30309,
United States, Andy Yoon, Xin Zhang
We will present a freight car fleet planning tool that evaluates the long-term
impact of fleet management decisions based on demand forecast, budget and
costs. The tool has been in use at Norfolk Southern for the last 10 years to assist
with decision making in freight car acquisitions to maximize long-term profits in
the future years.
3 - Intermodal Equipment Fleet Sizing
Bruce Patty, Veritec Solutions,
This presentation will review approaches used to determine the number of both
containers and chassis needed to support projected volumes. Models used to
estimate chassis demand by location will be discussed based on the experience of
the author as AVP-Equipment Strategy at Pacer Stacktrain. Offhire and onhire
planning will also be included in the session.
4 - Fleet Sizing Model: Railroad And Shipper Perspective
dharma acharya, KOSU Services LLC,
In this session, we will present how some railroads and shippers perform fleet
sizing of their rail car fleet to move the forecast shipments. We will also discuss
issues associated with uncertain demand forecast and cycle times. Additional
functionality requirements to meet the future business needs will also be
presented.
SC62
Cumberland 4- Omni
Aviation Applications Section: Best Student
Presentation Competition III
Sponsored: Aviation Applications
Sponsored Session
Chair: Lavanya Marla, University of Illinois, 216E Transportation
Building, 104 S. Mathews Ave., Urbana, IL, 61801, United States,
Entrants for the Best Student Paper Presentation competition will be presenting
their papers.
SC63
Cumberland 5- Omni
Network Design and Maintenance in Transportation
Sponsored: TSL, Intelligent Transportation Systems (ITS)
Sponsored Session
Chair: David Rey, UNSW Australia, School of Civil and Environmental
Engineering, UNSW, 2072, Australia,
1 - Scheduling Work Zones In Networks Of Service Vehicles
Dening Peng, Arizona State University, 699 South Mill Avenue,
Tempe, AZ, 85281, United States, ,
Pitu B Mirchandani
A mixed integer linear programming model is developed to schedule work zones
in transportation networks of service vehicles (e.g. trucks). For links with active
work zones, the available capacity of that link is reduced. The origin-destination
flow demands are given, and they route through the network based on available
capacities on the links to achieve the total minimum cost. The goal is to schedule
the work zones so that all maintenance work can be completed before a given
completion date while the total flow cost over the project period is minimized. An
innovative randomized fix-and-optimize heuristic is developed to solve the
problem with much better efficiency than commercial solvers like CPLEX.
3 - Optimal Deployment Of Autonomous Vehicle Lanes With
Endogenous Market Penetration
Zhibin Chen, University of Florida, 1, Gainesville, FL, 32603,
Lihui Zhang, Yafeng Yin, Fang He
This paper first models the evolution of autonomous vehicle (AV) market
penetration on a multimodal transportation network that includes conventional
vehicles (CVs) and AVs. A time-dependent mathematical approach is then
proposed to optimize the deployment plan of AV lanes such that the social cost is
minimized. The AV lanes are exclusive lanes for AVs, and the deployment plan
will dictate when and where, and how many AV lanes to be deployed.
4 - Road Extension Prioritization And Scheduling Problem
Mersedeh Tariverdi, University of Maryland, College Park,
Washington, DC, 20009, United States, ,
David Rey, Saeed Asadi Bagloee
Budget scarcity as well as limited resources in road construction may have a
significant toll on the completion of the projects. Therefore it is of highest
importance to arrive at a knowledge based decision support system for projects
prioritization and scheduling. We represent this problem as a bi-level program
where the objective function maximizes the return of projects’ completion
defined as the total savings derived from traffic improvement. The
interconnections among the projects as well as prerequisites requirements are
imposed using mixed integer constraints. An efficient solution algorithm is
developed based on an outer approximation approach.
5 - A Branch And Price Algorithm For The Work-zone
Scheduling Problem
David Rey, University of New South Wales, School of
Civil and Environmental Engineering, Sydney, Australia,
, Hillel Bar-Gera, Vinayak Dixit,
S. Travis Waller
We address the work-zone scheduling problem which consists in finding the
optimal schedule for the coordination of road occupancy projects in a transport
network over a planning period. Road works and maintenance operations which
require partial or total road closures over a period of time may considerably
impact network performance and result in significant delays. In addition, the
effects of conducting multiple work-zone projects simultaneously may be non-
additive, hence increasing the difficulty to anticipate congestion effects. We
present a new branch and price algorithm for the work-zone scheduling problem
which relies on the enumeration of work-zone project combinations.
SC64
Cumberland 6- Omni
DAAD Special Session on Multiobjective Optimization
for Improved Modeling of Complex Systems
Sponsored: Multiple Criteria Decision Making
Sponsored Session
Chair: Stefan Ruzika, University of Koblenz-Landau, Universitätsstraße
1, Koblenz, 56070, Germany,
1 - Efficient Computation Of The Search Region In
Multi-objective Optimization
Kerstin Daechert, Chair of Management Science and Energy
Economics, University of Duisburg-Essen, Germany,
Multi-objective optimization methods often proceed by producing new solutions
in an iterative way. For this purpose it is important to determine and update the
search region. It can be described by a set of so-called local upper bounds whose
components are defined by already known nondominated points. In this talk we
concentrate on how to design this update efficiently with the help of a specific
neighborhood structure among local upper bounds. Thanks to this structure we
can quickly identify all local upper bounds that have to be updated with respect
to a new nondominated point. Besides, the neighborhood structure provides new
theoretical insight into the search region for more than two objectives.
SC61