Informs 2015

195 / 552

INFORMS Philadelphia – 2015
193
3 - A Multistage Stochastic Programming Model for the Optimal
Surveillance & Treatment of Invasive Species
Eyyub Kibis, Graduate Research Assistant, Wichita State
University, 1845 N Fairmount, Wichita, KS, 67260,
United States of America, ,
Esra Buyuktahtakin, Robert Haight
In this study, we develop a multistage stochastic programming model to address
the invasive species surveillance and treatment while minimizing the expected
damages of invasive species. We use a discontinuous discrete decision tree and
incorporate discretized surveillance decisions along with the probabilities of each
scenario into the spatially-explicit model. The model allows policy makers to take
the best surveillance and treatment decisions over time by exploiting various
scenarios.
4 - Import Inspections: Harnessing Enforcement Leverage to Prevent
Invasive Species Introductions
Rebecca Epanchin-Niell, Resources for the Future, Washington
DC, United States of America,
Michael Springborn, Amanda Lindsay
Allocating scarce border inspection resources over a diverse set of imports to
prevent invasive pest entry presents a substantial policy design challenge. We
develop a risk-based inspection system in which sampling intensities vary across
imports based on risk. We determine optimal sampling of imports to minimize
invasive pest introduction accounting for strategic responses of exporters.
MB63
63-Room 112B, CC
Daniel H. Wagner Prize Competition II
Cluster: Daniel H. Wagner Prize Competition
Invited Session
Chair: Allen Butler, President & CEO, Daniel H. Wagner Associates, Inc.
2 Eaton Street, Hampton, VA 23669, United States of America,
1 - Integrated Planning of Multi-type Locomotive Service Facilities
under Location, Routing and Inventory Considerations
Kamalesh Somani, CSX Transportation, 500 Water St,
Jacksonville, FL, 32202, United States of America,
, Xi Chen, Yanfeng Ouyang,
Siyang Xie, Zhaodong Wang, Jing Huang
Long term infrastructure planning of locomotive service facilities is vital to the
efficiency of the railroad. We developed a large-scale optimization model that
integrates decisions on (i) location, capability, and capacity of fixed facilities, (ii)
home location and routing plan of movable facilities, and (iii) assignment of a
variety of service demands. A decomposition-based solution framework was
developed and shown to bring significant economic benefits in full-scale
implementations.
2 - Scheduling Crash Tests at Ford Motor Company
Daniel Reich, Leadership Program, Ford Motor Company,
Dearborn, MI, United States of America, ,
Amy Cohn, Ellen Barnes, Yuhui Shi, Marina Epelman,
Erica Klampfl
We present the problem of scheduling crash tests for new vehicle programs at
Ford. We developed a completely custom-made scheduling system that
transforms a labor-intensive scheduling process relying on high levels of
expertise, to a more automated one that utilizes optimization and institutionalizes
expert knowledge. Our system enables engineers and managers to consider
multiple scheduling scenarios, using efficient interfaces to specify problem
instances and efficient methods to solve them
MB64
64-Room 113A, CC
Joint Session DAS/ENRE: Environmental Decision
Analysis: Theory and Applications
Sponsor: Decision Analysis
Sponsored Session
Chair: Melissa Kenney, Research Assistant Professor, University of
Maryland, 5825 University Research Court, Suite 4001, College Park,
MD, 20740, United States of America,
1 - Decision Analysis for Sustainable Management of the
Yellow River Delta
Liang Chen, Student, Johns Hopkins University, 3900 N Charles
St. Apt. 1302, Baltimore, MD, 21218, United States of America,
, Benjamin Hobbs, Jeff Nittrouer,
Hongbo Ma, Andrew Moodie
We develop a stochastic programming model for channel management and flood
control that can characterize risks and impacts of possible natural avulsions, and
provides solutions for prevention and mitigation. Reflecting the physical
mechanism in coastal delta, our model imbeds a 1D hydrodynamic model to
simulate sediment transport, channel aggradation and flooding. Our case study is
Huanghe (Yellow River) Delta, China, one of the world’s most dynamic and
heavily urbanized coastal landscapes.
2 - Adaptive Stormwater Management with Green Infrastructure
using Two-stage Stochastic Programming
Fengwei Hung, Student, The Johns Hopkins University, 3400 N.
Charles Street, Ames Hall 313, Baltimore, MD, 21218, United
States of America, Benjamin Hobbs,
Arthur Mcgarity
Green Infrastructure manages stormwater with natural processes involving
significant uncertainty. Thus, many cities choose to implement it adaptively to
learn how it works. We define “learning” as updating of distribution parameters
of the stochastic program’s coefficients, representing: automatic learning,
triggering learning, multi-state learning, and multi-stage learning with technology
improvement. Finally, we calculate risk-return tradeoffs for a Philadelphia
stormwater case study.
3 - Framing Effects Created by Ambiguity Aversion in
Static Decisions
Erin Baker, University of Massachusetts, MIE Department,
220 ELAB, Amherst, MA, United States of America,
Eva Regnier
In climate policy-making, many respected economists recommend using
ambiguity-averse decision rules. The vulnerabilities created by ambiguity aversion
in dynamic decision making have been demonstrated previously. We show that
even in static, one-time, decisions, ambiguity-averse decision rules make policy
makers susceptible to bias created by framing effects.
4 - Using Multi-criteria Decision Analysis to Explore Management
Options in the Grand Canyon
Michael C. Runge, USGS Patuxent Wildlife Research Center,
12100 Beech Forest Road, Laurel, MD, 20708, United States of
America, , Kendra Russell, Kirk E. Lagory
The Bureau of Reclamation and the National Park Service are developing a Long-
term Experimental and Management Plan (LTEMP) for managing water releases
from the Glen Canyon Dam and related activities. We conducted multi-criteria
decision analysis to evaluate the proposed alternatives, integrating scientific input
from a dozen modeling teams, and values-focused input from a wide set of
deeply-involved stakeholder groups. We used value-of-information analysis to
inform experimental design.
MB64