VMASC Research Clusters
- Military/Homeland Security
- Medical & Health Care
- Transportation
- Social Science
- Virtual Environments
- Game-Based Learning
- Business & Supply Chain
Military/Homeland Security
Cluster Lead: Dr. Barry Ezell
Homeland security and military research at VMASC provides cutting edge developments in the area of modeling and simulation for use by government, industry, and other academic partners in developing applications to support Military and Homeland Security operations.
VMASC's goal is to apply and leverage our knowledge of the art and science of M&S to enable our partners to better serve their customers, by bringing to the forefront the latest developments in modeling, interoperability, visualization, and providing solutions to problems.
VMASC has a deep workbench in providing:
- Quick-Turn Analytic Support
- Methodology Development
- Model Development
- Tier One Group Facilitation
- Quantitative Risk Analysis
- JCIDS Analytic Support
- Support for the 2010 National Bioterrorism Risk Assessment (BTRA).
- Novel approaches to mature risk methodologies.
- Risk analysis support to DHS.
- Wounded Warrior Program M&S support.
- M&S development of evacuation plans.
Currently, the Homeland Security & Military Defense cluster provides Extreme Event Analysis, Probabilistic Risk Analysis, Intelligent Adversary Modeling, and Emergency Management support for the Joint Forces Command (C2 Simulation Interoperability, Project - XBML, Project - JRSG), the U.S. Army (PEO Soldier), as well as the Commonwealth of Virginia (Critical Infrastructure Interaction and Resiliency).
Homeland Security and Military Defense Cluster Meeting - Sept. 28, 2009
Homeland Security and Military Defense Cluster Meeting - June 29, 2009
Homeland Security and Military Defense Cluster Meeting - April 22, 2009
Homeland Security and Military Defense Cluster Meeting - February 10, 2009
Medical & Health Care
Cluster Lead: Dr. Mohammed Ferdjallah
The medical and health care modeling and simulation cluster has identified four areas in which we have expertise, specifically, the use of M&S for 1) training, 2) treatment, 3) disease modeling and 4) management of health care systems. The cluster consists of researchers from VMASC and ODU, area universities and health care systems. We have identified problems effecting medicine and health care, in which we have interests to implement modeling and simulation. The medical and health care M&S cluster has several researchers as members. Each area has highlighted projects.
In the area of training, researchers at ODU have developed a fully immersive Virtual Operating Room outfitted with a simulated patient and both real and simulated instruments. The system is designed to provide training for members of surgical teams using both real and virtual team members. Researchers are also applying M&S knowledge and expertise to reduce errors and improve safety practices in the area of Labor and Delivery. Efforts have been aimed at improving the identification and presentation of critical events embedded in maternal-fetal heart rate tracings. In addition, M&S technology is being used to augment training with standardized patients (i.e., individuals who realistically portray patients used to teach and assess communication and other clinical skills). A stethoscope has been developed that allows the learner to hear abnormal heart and lung sounds when placed on a normal, healthy standardized patient. Also, researchers are working to create a planning and training tool for surgeons who perform procedures to correct pectus excavatum (a severe chest wall deformity usually found in children and young adults).
The treatment focus area of the cluster focuses on using M&S for rehabilitation, for improving the diagnosis and treatment of orthopedic injuries and disorders and to optimize physical performance. Researchers at ODU are developing and validating a patient specific model of the hindfoot that will be used to optimize treatment in foot and ankle injuries. A cadaver based model of subtalar joint instability is being used to develop a clinically useful method to detect this type of ankle joint instability, which is frequently misdiagnosed. Researchers are also using modeling and simulation to develop a physical performance prediction model that will be used to identify when additional physical training is necessary and what type of training is necessary for an individual to perform their job. Current applications to this model will be implemented in the military. Finally, our researchers have developed a virtual reality based rehabilitation system for the treatment of patients with hemipareis as a result of stroke. A pilot study has show improvements in all patients treated with this system.
In the area of disease modeling, ODU researchers have worked on mathematical models and computer simulations of a variety of diseases. While incidences of tick-born diseases are sporadic in both space and time, there has been a significant increase in human cases over the past twenty years. In an effort to gain understanding into predicting the risk of these diseases and possibly how to control future outbreaks, a mathematical model has been developed. Similarly, in conjunction with researchers at the University of Maryland, School of Medicine, a mathematical model has been developed to explore various interventions to control the spread of methicillin-resistant Staphylococcus aureus (MRSA) on a hospital intensive care unit. This model showed that active surveillance with isolation upon admittance to the ward with release from isolation upon a negative test return would significantly reduce the number of new cases on the ward, but this approach was not found to be cost-effective given 2006 cost estimates. Another model was developed to explore the dynamics of the mosquito-borne disease, Rift Valley fever, which is currently only found in Africa and the Saudi Arabian peninsula. The model results shed light on both predicting outbreaks within Africa as well as highlighting when and where the United States could be susceptible to invasion, accidental or intentional. Finally, electronic patient records provide a rich data source for developing, testing and evaluating rapid disease outbreak detectors. Researchers at ODU have worked with US Veteran Affairs researchers to develop whole-system surveillance tools based on the VA's extensive electronic patient record system.
In the area of management of health care systems, we have used M&S to understand the effects of bioterrorism on the health care system in Hampton Roads with a mass casualty model. The mass casualty model examines the effect of the release of bubonic plague on the hospital systems in Hampton Roads. An additional project is to use M&S to help our hospital systems make staffing decisions during hurricanes by considering the accessibility of the hospitals and the surrounding community.
Transportation
Cluster Lead: Mr. Mike Robinson
Transportation researchers seek ways to increase individual and group mobility and safety through the use of modeling and simulation techniques. Research topics include several aspects of transportation, including traffic modeling, transportation network planning, behavioral influences and constraints, aviation safety, maritime operations, logistics and distribution, and training.
VMASC researchers work closely with Old Dominion University’s Transportation Research Institute, Maritime Institute, and Ship Maintenance, Operations, and Repair Institute as well as other leading experts to maximize the strength of project teams. The Transportation Laboratory at VMASC provides meeting space, dedicated graduate researcher computer use, and leading transportation simulation software.
The ODU/VMASC team brings both ground-breaking research and development and the innovative use of application of existing transportation simulation software to solve problems. Recent and current projects include:
- 2008 Multimodal Transportation Planning Grant
Asad Khattak: Principal Investigator, Mike Robinson : Co–Principal Investigator.
Land use and Transportation Planning for Isle of Wight and Suffolk, VA. Funded via the Virginia Department of Transportation, VA.
(Expected completion late 2010.) - Hampton Roads Transportation Alternatives Traffic Simulation
Mike Robinson: Principal Investigator, Asad Khattak: Co-Principal Investigator.
Simulation forecast the transportation situation that might be anticipated in the year 2030 if various combinations of major transportation projects are constructed. Completed for the Hampton Roads delegation of the Virginia General Assembly. - Nationwide Household Travel Survey
Asad Khattak—Principal Investigator. Virginia Department of Transportation, VA. (ODU Transportation Research Institute). - Hampton Roads Hurricane Evacuation Study
John Sokolowski: Principal Investigator, Asad Khattak: Co-Principal Investigator, Mike Robinson: Technical Lead.
Provided a dynamic model of multiple hurricane evacuation scenarios for Hampton Roads using selectable dynamic vehicle assignments and including the impact of accidents and incidents. Funded by the Virginia Department of Emergency Management. - Primary and secondary incident management: Predicting durations in real-time
Asad Khattak: Principal Investigator.
Virginia Transportation Research Council (VTRC), Virginia Department of Transportation.
VMASC also facilitates a Transportation Modeling and Simulation Cluster sponsored via a Department of Commerce Economic Development Assistance grant to the Hampton Roads region in southeastern Virginia. This grant seeks to increase the use of Modeling and Simulation in the region through increased use of existing products as well as new research and development. Transportation cluster members include faculty from ODU, the College of William and Mary, and Norfolk State University as well as government and industry representatives.
Participants have agreed on three immediate goals:
- Create a cooperative committee of individuals willing to propose and jointly pursue transportation projects involving the use of M&S.
- Identify venture capital or other funding sources.
- Identify research or application projects for pursuit as a group. Projects should have commercial viability and meet the needs of a large segment of the cluster.
New cluster members are always welcome. If interested in participating, please contact Mike Robinson at rmrobins@odu.edu.
Social Science
Cluster Lead: Dr. John Sokolowski
Social Sciences Modeling & Simulation
Social scientists work closely with traditional methods of modeling such as statistical modeling--a method for the discovery and interpretation of patterns in large numbers of events; formal modeling--a method that provides a rigorous analytic specification of the choices actors can make and how those choices interact to produce outcomes, and agent-based modeling, a method allowing for the observation of aggregate behaviors that emerge from the interactions of large numbers of autonomous actors. The VMASC Social Sciences Cluster facilitates research opportunities between M&S researchers at VMASC and liberal arts faculty as a way to enhance social sciences traditional modeling and analysis capacity. This is done by integrating modeling, simulation, and visualization as a tool to expand and communicate qualitative data.
Engaging modeling and simulation into an empirical analysis allows one to better understand the "what happened" and to explore the "what if." Social Scientists are integral to solving complex problems as more and more emphasis is being placed on modeling qualitative analysis alongside the quantitative data. This is important to fully grasp the history, culture, politics, economics, and social mores of a society. To do this, one must integrate various applications of modeling, simulation, and visualization into the research as a way to expand and communicate the qualitative analysis of the subject area and provide a much denser schematic for the model.
Modeling and Simulation has a number of modeling domains in its toolbox. Social Scientists are now engaging those tools in collaboration with engineers by employing variations of System Dynamics, Game Theory, Agent-based Modeling, and Social Network Modeling. Social Scientists can find common ground with producers of Modeling and Simulation and that common ground encourages multidisciplinary approaches to research.
Social Science Research
At present there are a number of outstanding projects and proposals that require social science expertise. Some include country case studies focused on assessing the political, military, economic, social, information, and infrastructure elements of society which are necessary to enhance foreign policy and decision making at various levels. Current work assessing the insurgency in the Niger Delta could mitigate a regional and global upheaval relative to the critical need for oil. The Social Science team is developing a study on measuring global warming to determine the effects of humans on the environment from a historical, social, and political perspective. A grant proposal has been submitted to The National Endowment for the Humanities (NEH) to begin creating a digital game, and educational tool that focuses on failed state in Africa.
Virtual Environments
Cluster Lead: Dr. Yiannis Papelis
Intelligent agents are entities, or actors, that exhibit rudimentary intelligence by observing their surrounding environment and then acting according to various goals. Intelligent agents utilize perception to asses the environment and exhibit goal-directed behavior when pursuing their goals. Agent-based-modeling is a technique that employs interacting intelligent agents to model a system. Research has shown that even relatively simple agent models will yield aggregate complex behaviors while interacting with each other and with their environment. When the complexity of individual agents is increased, it is possible to create models of systems whose complexity far exceeds the capabilities of traditional modeling techniques.
Intelligent agents can be purely virtual or physical. A purely virtual agent operates completely within a virtual environment. Perception is implemented in software by creating interrogations of the virtual environment that reflect physical (i.e., line of sight) as well as cognitive constraints (i.e., cognitive overload). Acting is implemented in software by using traditional continuous system modeling approaches to simulate a physical system whose excitation is provided by the agent (i.e., steering input to a car model), or by directly setting the values of virtual parameters. A physical agent operates in the actual world. Perception is implemented by utilizing sensors such as cameras and lasers that provide information about the surrounding environment to the agent model. Acting is implemented through actuators that directly interact with the environment. Examples of physical agents include unmanned aerial or ground vehicles, and autonomous robots.
Research in the cluster is focused on the techniques associated with developing both virtual and physical intelligent agents, including modeling approaches for developing the virtual environments within which virtual agents operate. Domain specific knowledge is incorporated into both the agent and the environment, yielding highly complex yet realistic simulations that can be used for exploration of new concepts and system approaches. The cluster involves focus areas such as the following: Virtual Intelligent Agents, Physical Intelligent Agents, Virtual Environments, and Immersive Simulations. Cluster researchers are currently involved in the following projects: Agent Based Crowd Modeling, AirPORTS (An Agent-Based End-to-End Class B Terminal Area Operations Simulation Tool), Automated Texture Synthesis from Multi-Spectral Satellite Imagery, and Critical Infrastructure Protection.
Game-Based Learning
Cluster Lead: Dr. Yuzhong Shen
Many people who are younger than 40 grew up on video and computer games and have great affections with electronic games. Even the elder generations start to like games due to the immense user interactions introduced in the latest game consoles, such as Nintendo Wii. The entertainment, challenges, and excitement provided by electronic games make them so engaging. Research has shown that games can be utilized as effective tools to motivate the learners for educational and learning purposes if designed properly. The vision of the game-based learning cluster is to be an important and leading player in research and development of educational games in the United States and in the world. The cluster teaches game development theories and technologies, develops educational games, and promotes the use of games for educational and training purposes. We engage with academic, government, and industry partners, provide forum and facilitation for coordination, and actively seek funding from federal, state, and local agencies as well as industries.
The game-based learning cluster is composed of researchers with backgrounds and expertise in computer science, computer engineering, education, psychology, art and modeling, and user interactions. The cluster is especially interested in and actively developing applications in the following areas:
- K-12 education, such as mathematics, physics, chemistry, history, biology, geography, etc.
- College education, especially STEM education (science, technology, engineering, and mathematics).
- Medical applications, such as surgical training, instrument training, patient recovery, etc.
- Public education, such as historical events and modern events simulation in museums.
- Training, such as driving simulation, mechanic training, business policy and procedures, etc.
- Applications for people with special needs, such as wheelchair training.
- Utilization of a wide variety of computing platforms, including personal computers, game consoles (e.g., Xbox 360), virtual environments, PDAs, multimedia devices such as Apple iPhone and Microsoft Zune player, web-based gaming, etc.
- Utilization of the latest gaming technologies, such as Microsoft XNA Game Studio, Delta3D, OpenSceneGraph, Alice, Game Maker, Flash, etc.
- Micro games that can be deployed and played quickly in order to teach a specific concept or technique.
Business & Supply Chain
Cluster Lead: Dr. Rafael Diaz
A large number of enterprises have successfully employed M&S along with optimization tools to analyze, design, modify, or expand their operations. Most cases have reported substantial benefits from implementing solutions obtained from using these paradigms. In VMASC, the Business and Supply Chain Research Cluster supports the enterprise by providing specialized knowledge in M&S and optimization. Our research cluster employs advanced simulation techniques to model, analyze, and build innovative solutions that increase the leverage of the enterprise. Three recent simulation studies developed at VMASC include enhancing an inventory control system for a retailer and decreasing customer waiting times and improving inventory management in a just-in-time production system.
Optimal storage assignment
This study considers using a technique that efficiently distributes and assigns SKUs inside the warehouse. As a result, operators or robots (selectors) minimize the traveled distance and the service time in the picking process. This technique is based on acknowledging three factors that include 1. probability of SKUs being ordered together (clustering), 2. SKUs throughput, 3. constraints in terms of weights and/or sizes. Results demonstrated that both time and traveled distance are significantly reduced.
JIT logistic and intelligent supply-chain systems
The second study was performed collaboratively with the College of Business and relates to a just-in-time production system in a job shop. This system was tested using a novel assignment priority rules in a job shop environment. The effectiveness of these priority rules was rigorously evaluated employing advanced techniques in M&S. This setting also mimics a JIT logistic system in which workstations represent supply chain nodes (i.e. transportation) in which decisions of what to produce, store, or transport are intelligently prioritized.
In addition, a third project that implied the mathematical description of the demand forecast resulted in a 30-40% improvement in measures of accuracy. Other completed research includes: Developing an economic evaluation model for evaluating transportation alternatives; Using cluster analysis, stochastic simulation, and simulation based optimization to derive optimal policies; Investigating how to increase supply chain flexibility and increase productivity by using regional enablers for Hampton Roads area; Using simulation-based optimization as decision-making tool to identify and reallocate critical resources in emergency departments.