WSC 2007 Final Abstracts

Health Care Track

Monday 1:30:00 PM 3:00:00 PM
Applying Simulation to Health Care

Chair: Georgina Mellor (University of Southampton)

Can Health Care Benefit From Modeling and Simulation in the Same Way as Business and Manufacturing Has?
Jasna Kujlis, Ray J. Paul, and Lampros Stergioulas (Brunel University)

It has been increasingly recognized that the application of simulation methods can be instrumental in addressing the multi-faceted challenges health care is facing at present and more importantly in the future. But the application of these methods seems not to be as widespread as in other sectors, where such methods when used as part of their core operation, reap significant benefits. This paper examines the potential use of modeling and simulation in health care, drawing the parallels and marking the mismatches from the business and manufacturing world. Methods from the latter sectors will be reviewed with the intention to assess their potential usefulness to healthcare. To focus this discussion, we propose and discuss seven axes of differentiation: patient fear of death; medical practitioners (for example approach to healing, investigation by experimentation and finance); healthcare support staff; health care managers; political influence and control; ‘society’s view’; and utopia.

Towards a Framework for Healthcare Simulation
Tillal Eldabi and Terry Young (Brunel University)

The changing needs of healthcare provision around the world are forcing service designers and decision makers to adopt new tools in design and evaluation of processes. Apart from the pressure to deliver better services from constrained resources, the increasing use of metrics to monitor and manage care delivery, also means that service providers require a clearer idea of how a service improve-ment will perform prior to implementation. In turn, this opens up an opportunity for a much greater use of simulation and modeling techniques, provided they can be set within an appropriate framework. This paper discusses and describes a research project aimed at conducting pilot work for developing a framework that facilitates joined up thinking and enables integrative modeling. An approach to achieving such an end is described and progress to date is reported. Since this is an ongoing project, some of the latest results are presented at the conference.

Interconnected DES Models of Emergency, Outpatient, and Inpatient Departments of a Hospital
Murat M. Gunal and Michael Pidd (Lancaster University)

National Health Service (NHS) performance targets in England have put pressure on hospital management to reduce waiting times. The stochastic nature of emergency patient arrivals creates problems for capacity planning for elective patients. We present a whole hospital model which can be used at policy level to investigate cause and effect relations, such as effects of increased emergency arrival volumes on elective waiting times. A typical general hospital can be abstracted in three main units; Accident and Emergency (A&E) department, outpatient clinics, and inpatient units. In real life these units are coupled and share hospital resources. We developed three discrete event simulation (DES) models for each unit to form a whole hospital DES model. We present our models conceptually and our main discussion is on the level of detail in these three models.

Monday 3:30:00 PM 5:00:00 PM
Clinical Models

Chair: Michael Pidd (Lancaster University)

A Discrete Event Model of Clinical Trial Enrollment at Eli Lilly and Company
Bernard M. McGarvey, Nancy J. Dynes, Burch C. Lin, Wesley H. Anderson, James P. Kremidas, and James C. Felli (Eli Lilly and Company)

Clinical trials constitute large, complex, and resource intensive activities for pharmaceutical companies. Accurate prediction of patient enrollment would represent a major step forward in optimizing clinical trials. Currently models for patient enrollment that are both accurate and fast are not available. We present a discrete event model of the patient enrollment process that is accurate and uses relatively small CPU times. This model is now being used on a regular basis to predict the enrollment of patients for large trials with around 13,000 patients and has led to significant reduction in the time it takes to make these predictions.

Important Factors in Screening for Colorectal Cancer
Reza Yaesoubi and Stephen Dean Roberts (North Carolina State University)

A complex, stochastic simulation model of Colorectal Cancer (CRC) is examined through factor screening to determine which factors in the model are important. The factor screening employs a Resolution IV 2k Fractional Factorial experimental design. The factors are examined in terms of their impact on cost, quality-adjusted life years (QALY), and cost per QALYs. Out of 72 factors, eight factors were determined to be most important and observed as "driving factors" in the CRC model. Surprisingly these factors were consistently important for all outcomes. However the limitations of the experimental design may have constrained the important factors to factors related only to the natural history of the disease and therefore subject to minimal control.

Roles for Autonomous Physiologic Agents; an Oxygen Supply and Demand Example
Meyer Katzper (SIA)

In the study of physiologic systems control, lumped parameter and differential equation techniques are standard approaches. Application of these techniques to the study of oxygen supply to tissues is discussed. It is then proposed that progress in dealing with heterogeneous physiologic systems is likely to proceed from the techniques of agent based modeling in the form of autonomous physiologic agents.

Tuesday 8:30:00 AM 10:00:00 AM
Health Services

Chair: Stephen Roberts (North Carolina State University)

Targeted Strategies for Tuberculosis in Areas of High HIV Prevalence: A Simulation Study
Georgina Mellor and Christine S. M. Currie (University of Southampton), Elizabeth Corbett (London School of Hygiene and Tropical Medicine) and Russell C. H. Cheng (University of Southampton)

We describe the analysis of a discrete event simulation model of tuberculosis (TB) and HIV disease, parameterized to describe the dual epidemics in Harare, Zimbabwe. The HIV epidemic in Sub-Saharan Africa is particularly severe and has led to a significant rise in TB cases. We use the model to evaluate new strategies for improved detection of TB cases in a high HIV prevalence setting. The structure of the model and its validation will be discussed, but the paper will focus on the analysis of the model output.

Improving Primary Care Access Using Simulation Optimization
Hari Balasubramanian, Ritesh Banerjee, and Melissa Gregg (Mayo Clinic) and Brian T. Denton (North Carolina State University)

Primary care providers (PCPs) provide the majority of care patients receive during their lifetime. We consider the problem of determining the size and composition of physician panels in primary care. A physician's panel consists of a set of patients and each patient belongs to one of many different health-related categories. Using real data collected at the Mayo Clinic at Rochester, we propose a multi-period metaheuristic simulation optimization model for determining the panel design of a set of physicians working in a primary care environment. The model seeks to maximize patient visits to their own providers, reduce waiting times, and minimize overage.

An Approach to Hospital Planning and Design Using Discrete Event Simulation
Ian William Gibson (Bovis Lend Lease)

Recent reports have established the need for change in the US health system. Building projects can play an important role in enabling change to support organizational objectives. The current major investment in hospital construction in the US provides an opportunity to improve health service. Planning and design of hospitals generally uses benchmarks and experience without rigorous analysis of processes, resources and facility requirements. This paper considers an improved approach to planning and design of hospitals by using Discrete Event Simulation (DES) to enable improvement in the quality and productivity of health services and an improved workplace for staff

Tuesday 10:30:00 AM 12:00:00 PM
Outpatient and ED Models

Chair: Murat Gunal (Lancaster University)

Bi-criteria Evaluation of an Outpatient Surgery Procedure Center Via Simulation
Todd R. Huschka (Mayo Clinic), Brian T. Denton (North Carolina State University) and Serhat Gul and John W. Fowler (Arizona State University)

Surgical services require the coordination of many activities, including patient check-in and surgical preparation, surgery, and recovery after surgery. Each of these activities requires the availability of resources including staff, operating rooms, and intake and recovery beds. Furthermore, each of these activities has substantial uncertainty in their duration. The combination of a complex resource constrained environment, and uncertainty in the duration of activities, creates challenging scheduling problems. In this study we report on a discrete event simulation model of an outpatient surgical suite, and investigate the impact of several sequencing and scheduling heuristics on competing performance criteria.

"See and Treat" or "See" and "Treat" in an Emergency Department
Ruth M. Davies (University of Warwick)

"See and Treat" in an Emergency Department combines the process of patient assessment with treatment in the expectation that it will increase patient throughput and decrease queuing. This paper describes an evaluation of the flow of minor emergencies in an Emergency Department in the UK that had partially implemented "See and Treat" and was planning to reorganize the department yet again to reseparate the activities of assessment and treatment. A discrete event simulation indicated that the proposed system in which "See" and "Treat" were separated improved patient throughput and was likely to be more cost-effective. There were difficulties in obtaining credible data for the analysis, though this was mitigated by using the same distributions, for the analysis of both of the systems. With increasing pressure to introduce industrial concepts, such as Lean, to the health sector, simulation provides a means of assessing expensive and disruptive changes before implementation.

Modeling of Patient Flows in a Large-scale Outpatient Hospital Ward by Making Use of Electronic Medical Records
Soemon Takakuwa (Nagoya University) and Daisuke Katagiri (Daifuku Co., Ltd.)

All departments of an outpatient hospital ward of Nagoya University hospital were simulated to examine patient flows and congestion. The method of gathering the required data on times for all outpatients and their routes is described in the performing simulation, especially by making use of the electronic medical records. An outpatient visits one or more clinical departments and/or one or more test/inspection rooms, the reception area, and the payment department. In this procedure, a series of data of terminal units and of test/inspection terminals was used to obtain the required input data for performing simulation as well as the electronic medical records. It was found that the proposed procedure was quite effective to perform a simulation of a large-scale hospital to examine patient flows by applying an actual case.

Tuesday 1:30:00 PM 3:00:00 PM
Epidemic Models

Chair: Douglas Roberts (Research Triangle Institute)

A Hybrid Epidemic Model: Combining the Advantages of Agent-based and Equation-based Approaches
Georgiy V. Bobashev, D. Michael Goedecke, and Feng Yu (RTI International) and Joshua M. Epstein (Brookings Institution)

Agent-based models (ABMs) are powerful in describing structured epidemiological processes involving human behavior and local interaction. The joint behavior of the agents can be very complex and tracking the behavior requires a disciplined approach. At the same time, equation-based models (EBMs) can be more tractable and allow for at least partial analytical insight. However, inadequate representation of the detailed population structure can lead to spurious results, especially when the epidemic process is beginning and individual variation is critical. In this paper, we demonstrate an approach that combines the two modeling paradigms and introduces a hybrid model that starts as agent-based and switches to equation-based after the number of infected individuals is large enough to support a population-averaged approach. This hybrid model can dramatically save computational times and, more fundamentally, allows for the mathematical analysis of emerging structures generated by the ABM.

A Stochastic Equation-Based Model of the Value of International Air-Travel Restrictions for Controlling Pandemic Flu
D. Michael Goedecke, Georgiy V. Bobashev, and Feng Yu (RTI International)

International air travel can be an important contributing factor to the global spread of infectious diseases, as evidenced by the outbreak of Severe Acute Respiratory Syndrome in 2003. Restrictions on air travel may therefore be one response to attempt to control a widespread epidemic of a disease such as influenza. We present results from a stochastic, equation-based, global epidemic model which suggest that air travel restrictions often provide only a slight delay in the epidemic. This delay may give valuable time in which to implement other disease control strategies; however, if other strategies are not implemented, the use of travel restrictions alone may lead to a more severe epidemic than if they had not been imposed. Our results also indicate that the particular network of cities chosen for modeling can have a great influence on the model results.

A Flexible, Large-scale, Distributed Agent Based Epidemic Model
Jon Parker (Brookings Institution)

We describe a distributed agent based epidemic model that is capable of easily simulating several hundred million agents. The model is adaptable to shared-memory and distributed-memory architectures. Several problems are addressed to enable the distributed simulation: allocation of agents to available compute nodes, periodic synchronization of compute nodes, and efficient communication between compute nodes. We assert that our modeling scheme is easily adaptable to different hardware environments and does not require large investments in performance tuning or special case coding.

Tuesday 3:30:00 PM 5:00:00 PM
Spatial Epidemic Models

Chair: Georgiy Bobashev (Research Triangle Insitute)

Simulating Pandemic Influenza Risks of U.S. Cities
Catherine Dibble, Stephen Wendel, and Kristofor Carle (University of Maryland)

We describe the spatial Agent-Based Computational Laboratory that we have developed to study the pandemic influenza risks of US cities. This research presented a series of interesting challenges, from the uncertainty surrounding the future epidemiological characteristics of a human-transmission H5N1 strain of pandemic influenza, to the need to provide timely decision-support despite modeling a country with a population of 300 million individuals. In order to provide time-sensitive policy analyses, we implemented a light-and-fast agent-based model of the spatial and temporal spread of pandemic influenza, which uses a novel compression technique to analyze large numbers of agents. We assessed the impact of parameter uncertainty and of stochastic behavior via very large numbers of simulations. To facilitate this, we developed a parallel job controller that tests combinations of disease scenarios, and a platform-independent job-submission application that harnesses the computational resources of high-performance computing environments ranging from local clusters up through TeraGrid super-computers.

A Teragrid-enabled Distributed Discrete Event Agent-based Epidemiological Simulation
Douglas Roberts and Diglio A. Simoni (RTI International)

We discuss design issues related to the transformation of a mature Agent-Based Model (ABM) for computational epidemiology into a “grid-aware” version. EpiSims is a distributed discrete event ABM that has been in production for nearly a decade. Working under a grant from the National Science Foundation and the NIH (NIGMS) funded MIDAS project, we are reengineeriing EpiSims to run as a single job on multiple Linux clusters on the NSF T

Utilizing Model Characteristics to Obtain Efficient Parallelization in the Context of Agent-based Epidemilogical Models
Steven Naron (Independent Consultant) and Segev Wasserkrug (IBM Research)

There exist many problem agnostic frameworks and algorithms for parallel simulation. However, creating parallel simulation models that take advantage of characteristics specific to either the problem domain or specific model can create significant performance benefits. This article provides an overview of general frameworks and algorithms for paralleling simulation execution, and also demonstrates two ways in which assumptions underlying the implementations of epidemiological models can be used to enable such parallelization in an efficient manner. These examples are based on planning and developing agent-based models activities carried out as part of the NIH's MIDAS (Models of Infectious Disease Agent Study) family of grants.

Wednesday 8:30:00 AM 10:00:00 AM
Hospital Strategic Management

Chair: Hari Balasubramanian (Mayo Clinic)

Simulating the Patient Move: Transitioning to a Replacement Hospital
Marshall Ashby, Martin Miller, David M. Ferrin, and Tanner Flynn (FDI Simulation)

One of the more complex maneuvers a hospital system can perform is moving an entire patient population from an old facility to a replacement facility. All patients must be transported via ambulance or van to a new replacement hospital. This requires massive resources, permits, cooperation of local government, and often assistance from neighboring hospitals. This study utilized simulation to determine optimal resources, routing, and timing for the movement of almost 600 inpatients from two different facilities to a new replacement facility. Potential resource constraints of specialized move teams, ambulances, and other staffing constraints were explored to predict and reduce the likelihood of complications during the two day patient move.

Maximizing Hospital Finanacial Impact and Emergency Department Throughput with Simulation
David M. Ferrin and Martin Miller (FDI Simulation) and Diana McBroom (Carondelet St. Mary’s Hospital)

Carondelet St. Mary's Hospital (Tucson, Arizona), the Ascension Health Operations Resource Group and FDI Simulation team worked collaboratively to improve hospital flow and increase access to care by implementing process improvements based on simulation that reduced the Emergency Center (EC) length of stay by 7%, increased the EC monthly volume by 5%, increased the inpatient daily census by 20% and improved the hospital net operating margin by 1.3% above budget. This paper demonstrates simulation's unique ability to direct improvement efforts for maximum impact operationally, financially and for the best benefit of the patient.

Merging Six Emergency Departments Into One: A Simulation Approach
Martin Miller, David M. Ferrin, Marshall Ashby, and Tanner Flynn (FDI Simulation) and Niloo Shahi (LAC+USC Healthcare Network)

Simulation of existing systems can reinforce a Subject Matter Expert’s gut feelings. However, it is more difficult to develop intuition for proposed systems, particularly when considering the consolidation of multiple systems. This paper discusses the use of simulation to determine the operational ramifications of combining six Emergency Departments into one of the largest in the country. Each of these six existing Emergency Departments serve a different type of patient population and each maintains their own independent processes. This hospital required all Emergency Departments to effectively function using the same floor space, processes and ancillary services, such as testing facilities, waiting rooms, and registration. Healthcare planners need to understand the ramifications of sharing resources among multiple departments and the operational impact of high volume systems. This project explored these challenges to find key bottlenecks and mitigation strategies using simulation.

Wednesday 10:30:00 AM 12:00:00 PM
Improving Health Care Operations

Chair: Andrew Seila (University of Georgia)

Comparing Simulation Alternatives Based on Quality Expectations
Joshua Bosire and Shengyong Wang (Binghamton University (SUNY)), Tejas Gandhi (Virtua Health) and Krishnaswami Srihari (Binghamton University (SUNY))

Computed Tomography (CT) is one of the fastest growing diagnostic imaging procedures. Rapid advances in imaging technologies in conjunction with their widening adoption are some of the issues that are compelling healthcare providers to restructure their systems as they seek to offer a higher quality of care to a growing volume of patients. This paper presents the application of simulation to facilitate the planning of a new CT facility for a hospital. The objective of the study was to evaluate how patient experience would be impacted by proposed design options. Waits for service were utilized as a parameter to quantify the patients’ quality expectations, and hence the satisfaction derived from the healthcare services received. This study was also intended to clarify whether an additional CT-scan unit was a necessity to improving the patients’ experience.

Effect of Coupling between Emergency Department and Inpatient Unit on the Overcrowding in Emergency Department
Erik Michael Wilhelm Kolb, Taesik Lee, and Jordan Peck (MIT Park Center for Complex Systems)

Emergency Department (ED) overcrowding has become a common problem in the United States as well as other developed nations, threatening the safety of patients who rely on timely emergency treatment. Volume of high-acuity patients and the volume of patients that are later admitted to the inpatient unit (IU) are factors reported as major causes of ED overcrowding. These two factors can be interpreted to represent the strength of the interaction between an ED and its associated IU. In addition to confirming the observations reported in previous studies, we were able to use discrete event simulation to characterize the relationship between IU utilization and ED crowding: it was found that the sensitivity of ED overcrowding with respect to IU utilization depends on the degree of coupling between the two units. Our findings have potential implications in guiding a hospital's effort to optimize their system.

[ Return to Top ]