2016 INFORMS Annual Meeting Program

24 / 561

INFORMS Nashville – 2016
24
SA20
SA20
106C-MCC
Storage and Read-Optimized Data Placement
Structures for High Performance Analysis
Invited: Tutorial
Invited Session
Chair: Edmon Begoli, University of Tennessee-Knoxville,
Joint Institute for Computational Sciences (JICS),, Knoxville, TN,
37831, United States,
1 - Storage And Read-Optimized Data PlacementStructures For High
Performance Analysis
Edmon Begoli, University of Tennessee-Knoxville, Joint Institute
for Computational Sciences (JICS), Knoxville, TN, 37831,
United States, Pragnesh Patel, J. Blair Christian
We present state-of-art structures and methods for efficient data preparation, and
representation for analysis. Our intent is to introduce the data science and
analytics communities to open source data placements, structures, and methods.
These practices can make the foundational processes of data preparation and
access dramatically more efficient than typical raw file or database representations
and use more conservative storage. To illustrate this, we introduce two highly
efficient data placement structures. We then present a tutorial, supported by step-
by-step examples, of how to create, use and access data, structured by Parquet or
ORC, using Apache Spark. Finally, we illustrate the benefits of using these
structures with computational and storage volume benchmarks.
SA21
107A-MCC
Healthcare Operations Research in Practice
Sponsored: Health Applications
Sponsored Session
Chair: Sandra Potthoff, University of Minnesota, 420 Delaware Street
SE, D362 Mayo MMC 729, Minneapolis, MN, 55455, United States,
Co-Chair: David J. Cook, Mayo Clinic, 200 1st St SW, Rochester, MN,
55902, United States,
1 - Optimization-based Scheduling Tools For Hospital Staff
Fei Li, Hennepin County Medical Center,
Clinics and hospital units hope to staff more accurately to demand patterns, but
many have trouble doing so, because of multiple difficulties. We try to understand
the difficulties in real practice and offer optimization-based propose
a solution that is both easier to use and more personalized. Our solution includes
a one-stop web-based access to our CPLEX server for any model, and a
customized user-friendly tool for each team. They gain users’ acceptance, and we
reach better schedules that both the department and staff are happy
with.Continued effort cannot be saved in making the tool fit the team’s need.
Including modeling skill in supporting staff is recommended.
2 - A Target Discharge Strategy For Inpatient Discharge Planning
Nicholas Ballester, PhD Candidate, Wright State University,
1555-A Sudbury Lane, Faiborn, OH, 45324, United States,
, Pratik Parikh, Nan Kong
Ineffective inpatient discharge planning may cause discharge lateness and
upstream patient boarding. Working with a local hospital, we propose a novel
target discharge strategy, the n-by-T strategy. We demonstrate the effectiveness of
this strategy via a simulation study, which suggests nearly 2 hr of advancement in
the mean discharge time and nearly 15% reduction in upstream boarding. The
sensitivity of this strategy for varying unit’s occupancy rates and findings from a
trial implementation at this hospital unit will also be discussed.
3 - Enhancement and Validation Of Outpatient Blood And Marrow
Transplant Practice Using Time Driven Activity Based Costing
(TDABC)
William J Hogan, Mayo Clinic-Rochester, NY, United States,
Mayo Clinic has developed an infrastructure to permit the performance of a
complex procedure, Blood and Marrow Transplant (BMT), as an outpatient. Using
a combination of Standardize to Value (STV) and Time Driven Activity Based
Costing (TDABC) approach a number of strategies were developed to converge
the practice across multiple sites resulting in greater standardization, consistency,
quality of care and cost containment. The value of performing BMT as an
outpatient was interrogated using TDABC, which effectively demonstrated
enhanced resource utilization and cost containment while maintaining excellent
clinical outcomes. These principles are generalizable to other practices involving
complex care.
4 - Anesthesiology Practice Redesign
Michael J Brown, Mayo Clinic-Rochester, NY, United States,
The Department of Anesthesiology led perioperative practice redesign initiative
improved the quality and increased the value of perioperative care. Utilizing an
analytical approach to identify, prioritize, assimilate, and interpret impactful
information, multiple high-value strategies were implemented that optimized
Mayo Clinic’s perioperative facilities, human resources, workflows and clinical
outcomes. The majority of interventions were perioperative specialty specific,
however others were generalizable to the entire practice. “
SA22
107B-MCC
Incentives and Resource Allocation in
Healthcare Settings
Invited: ORinformed Healthcare Policies
Invited Session
Chair: Karthik Natarajan, University of Minnesota, Edina, MN, United
States,
1 - Optimal Patient And Provider Incentives In Funding-constrained
Humanitarian Healthcare Service Settings
Karthik Natarajan, University of Minnesota, Minneapolis, MN,
United States, Mili Mehrotra
We analyze how a budget-constrained humanitarian organization managing a
healthcare service program should design incentives to the provider and patients
to maximize program coverage. We explore how the incentives change with the
service offered and operating environment. We also compare the optimal
incentive scheme to incentive schemes used in practice.
2 - Competing For Donations In The NPO Sector
Milind Sohoni, Indian School of Business,
, Sripad K Devalkar, Neha Sharma
We study the impact of competition under two fund-raising models, commonly
observed in the non-profit sector, when donors and non-profit organizations face
challenges due to information asymmetry and outcome (benefit) uncertainty. We
also analyze implications on benefits delivered from a social planner’s perspective.
3 - Discharge Decision In Emergency Departments: Impact Of
Operational Measures And Pay-for-performance Incentives
Eric Park, University of Hong Kong, 1, Hong Kong,
We study how operational measures in the emergency department such as
number of patients waiting to be seen and physician’s patient load affect patient
discharge decisions. We also analyze the impact of a provincial government level
pay-for-performance incentive scheme on discharge decisions. We empirically
study several major hospitals in the metro Vancouver, Canada area.
SA23
108-MCC
Evaluating Health Policy Decisions
Sponsored: Health Applications
Sponsored Session
Chair: Zhaowei She, Georgia Institute of Technology, 755 Ferst Drive,
NW, Atlanta, GA, 30332, United States,
1 - Positive Externalities In Disease Intervention:
A Study Of Mosquito Borne Viruses
Anneke Claypool, Stanford University,
Jeremy Goldhaber-Fiebert
Health interventions often result in positive externalities that are not captured in
traditional cost-effectiveness analysis. This study uses a dynamic transmission
model to analyze the cost-effectiveness of Chikungunya prevention measures. We
compare a potential Chikungunya vaccine to vector control methods that result in
fewer Chikungunya, Zika and Dengue cases. By using this model, the additional
health benefits of interventions that impact multiple diseases can be captured.