Informs 2015

81 / 552

INFORMS Philadelphia – 2015
79
3 - A Hybrid Social Network-system Dynamics Model of
Team Performance
Kyle Lewis, Professor, U. California - Santa Barbara, 1318 Phelps
Hall, Santa Barbara, CA, 93106, United States of America,
Edward Anderson
Social network analysis has explored many aspects of inter-personal interaction,
yet it is limited in its power to describe important features of team behavior. We
present a hybrid system dynamics-agent based methodology that extends the
power of SNA to analyzing emergent team behavior. We present a proof-of-
concept hybrid model and use it to simulate the: differential effects of hierarchy;
impact of overspecialization; role of generalists; and disruption created by
member turnover.
4 - IT Governance Mechanisms and Organizational Performance:
Investigating the Moderating Role of Platform
Hossein Kalantar, PhD Student, University of Colorado Denver,
1475 Lawrence Street, Denver, CO, 80202, United States of
America, , Jiban Khuntia
Information Technology governance plays a key role in creating value through IT
within an enterprise. There are studies that show the positive impacts of IT
governance on organizational performance. However, there are not many studies
that answer “How IT governance improves organizational performance”. In this
study, we investigate the moderating role of platforms, on the relationship
between IT governance and organizational performance. Context of Health IT
organizations was selected to conduct this study.
5 - Organizational Changes to Benefit from Big Data
Amit Das, Associate Professor, Qatar University, P.O. Box 2713,
Doha, Qatar,
While developments in computing have ushered in the age of Big Data, accounts
of Big Data being actually used to improve the management of organizations are
still relatively rare. We ascribe this to the incompatibility of traditional
management practices with the form of evidence-based decision-making enabled
by Big Data. We suggest that the profession of management is not alone in its
struggle to incorporate Big Data into its established routines.
SB41
41-Room 102A, CC
Optimization Methods in Healthcare Scheduling
Sponsor: Manufacturing & Service Oper
Mgmt/Healthcare Operations
Sponsored Session
Chair: Retsef Levi, J. Spencer Standish (1945) Professor of Operations
Management, Sloan School of Management, MIT, 100 Main Street,
BDG E62-562, Cambridge, MA, 02142, United States of America,
1 - Real-time Pooling for Multi-site Imaging Facilities
David Shmoys, Cornell University, School of ORIE, Rhodes Hall,
Ithaca, NY, 14853, United States of America,
, Chaoxu Tong, Shane Henderson
MRI patients waste a great deal of time in the waiting room; this is largely due to
the misalignment between scheduled and actual imaging time. By selecting
among a pool of nearby MRI facilities, we can redirect patients shortly before
their scheduled appointment time. We demonstrate that this improved load-
balancing can decrease patient waiting time. Working with New York Presbyterian
Hospital, we are implementing a trial for our approach in a complicated real-life
environment.
2 - Increasing throughput in a Large Oncology Infusion Unit
Ana Cecilia Zenteno Langle, Massachusetts General Hospital,
55 Fruit Street, White 400, Boston, MA, 02114, United States of
America, Retsef Levi,
Wendi Rieb, Inga Lennes, Mara Bloom, Bethany Daily,
Peter Dunn
We describe a data-driven online scheduling algorithm aimed at generating a
more predictable and balanced intra-day resource utilization in the Infusion Unit
at the Massachusetts General Hospital Cancer Center. The implementation of the
algorithm, which is based on integer optimization and simulation methods, has a
projected impact of reducing by 30% the average peak utilization and its standard
deviation by 35%. The hospital has contracted with an outside vendor to build a
customized IT tool.
3 - Simultaneous Scheduling of Nurses in Multiple Hospital Units
using Stochastic Integer Programming
Sanjay Mehrotra, Northwestern University, Industrial
Engineering and Management, 2145 Sheridan Road,
Evanston, IL, 60208, United States of America,
, Kibaek Kim
We will present theoretical and computational results on simultaneous scheduling
of nurses in multiple hospital units using a two-stage stochastic mixed integer
programming model. The model allows a nurse pool as well as sharing of nurses
from a more specialized unit to a lesser one. We show that the integrality of the
second stage can be convexified in our model, which allows for the solution of
larger scale models.
4 - Logic-Based Benders’ Decomposition Approaches with
Application to Operating Room Scheduling
Vahid Roshanaei, PhD Candidate, University of Toronto, 5 King’s
College Road, Toronto, ON, Canada, ,
Dionne Aleman, David Urbach
We develop three logic-based Benders’ decomposition (LBBD) approaches and a
cut propagation mechanism to solve location-allocation integer programs (IPs).
Each LBBD is implemented in four different ways, yielding 24 distinct LBBD
variants. We illustrate the LBBDs’ performance on the distributed operating room
scheduling problem, where patients and operating rooms are scheduled across
hospitals. Our LBBDs are 10-100x faster than IP+Gurobi and are more successful
at finding optimal solutions.
SB42
42-Room 102B, CC
Healthcare Operations Modeling and Optimization
Sponsor: Manufacturing & Service Oper
Mgmt/Healthcare Operations
Sponsored Session
Chair: David Kaufman, University of Michigan, 1205 Beal Ave.,
1710 IOE Building, Ann Arbor, MI, United States of America,
1 - Allocating Operating Room Time for Elective Surgery
Steven Shechter, Associate Professor, Sauder School of Business,
University of British Columbia, University of British Columbia,
Vancouver, BC, Canada, ,
Mahesh Nagarajan, Stephanie Carew
We examine how to allocate operating room hours to different surgical specialties
at the British Columbia Children’s Hospital. This is a longer-run planning decision
which has major effects on the wait time experience of the patient population. To
evaluate policies, we construct and validate a simulation model of patient arrival
and appointment processes. We then apply optimization and dynamic
programming techniques to recommend improved allocation policies.
2 - Panel Size, Office Visits and Care Coordination Events in
Primary Care
Hari Balasubramanian, University of Massachusetts Amherst, 160
Governors Drive, Amhert, MA, 01002, United States of America,
Michael Rossi
Using the Medical Expenditure Panel Survey (MEPS, AHRQ), we present a
method to estimate office visit and care coordination workload generated by
patients in a primary care panel. The method uses individual patient histories for
a one year period.
3 - Allocating Scarce Resources in a Patient Centered
Medical Home (pcmh)
Jingxing Wang, University of Michigan, 1205 Beal Ave., Ann
Arbor, MI, 48109, United States of America, ,
Romesh Saigal
We consider a two stage stochastic allocation problem to assign the number of
hours of Primary Care Physician to teams in a PCMH. In the first stage, a
preliminary assignment is made. In the second stage, the demand is observed and
the preliminary assignment adjusted to meet it exactly. We use real options
theory and present three ways to achieve a fair and consistent mechanism to price
the disruption caused by adjustment. The assignments are made such that the
price of disruption is the same.
4 - An Outpatient Planning Optimization Model for Integrated Care
and Access Management
David Kaufman, University of Michigan, 1205 Beal Ave.,
1710 IOE Building, Ann Arbor, MI, United States of America,
, Jivan Deglise-hawkinson, Todd Huschka,
Mark Van Oyen, Jonathan Helm
We present a data-driven methodology for outpatient scheduling. Our work is the
result of a practice-based collaboration with a major medical destination center.
Our capacity planning model seeks to meet visit targets on the time delay from
the appointment request to the appointment occurrence by patient type while
managing the patient mix, which is steered by goals such as increasing the
volume of new patient visits. The focus of the talk is on model validation and
insights.
SB42