Operational Research Commons^™

Contextual Detection Of Anomalies In Hyperspectral Images, Adam J. Messer

Theses and Dissertations

The majority of anomaly detectors in Hyperspectral Imaging use only the statistical aspects of the spectral readings in the image. These detectors fail to use the spatial context that is contained in the images. The use of this information can yield detectors that out perform their spatially myopic counterparts. To demonstrate this, we will use an independent component analysis based detector, autonomous global anomaly detector (AutoGAD), developed at AFIT augmented to account for the spatial context of the detected anomalies. Through the use of segmentation algorithms, the anomalies identified are formed into regions. The size and shape of these regions …

Decision Analysis And Validation Of Value Focused Thinking Decision Models Using Multivariate Analysis Techniques, James L. Pruitt

Theses and Dissertations

Decision Analysis (DA) is a useful tool to assist decision makers (DM) with difficult and complex decisions using mathematical models. Value Focused Thinking (VFT) models are a useful DA tool widely employed in the Air Force. However, VFT models are rarely validated. This research will attempt to validate any given VFT model and provide insight into the discriminating attributes of the alternative set. First, a two group discriminant analysis is applied the alternative set given the prior knowledge of the selected alternatives. Next, compromise programming is used attempt to minimize the distance between the posterior probability of an alternative being …

Empirical Characterization Of Ballistic Impact Flash, Thomas P. Talafuse

Theses and Dissertations

Fires onboard aircraft are leading mechanisms for mishaps and losses during peacetime and combat operations. Typical ignition sources causing fires onboard aircraft include electrical sparks and hot surfaces. However, impact and penetration of common threats encountered during combat operations, such as armor piercing incendiary projectiles and missile fragments, generate short-lived, but thermally-intense clouds (flashes) capable of igniting fires. Fire simulations supporting system-level survivability analyses depend on accurate characterization of these flash clouds, however, such accurate representations are not currently available. This research presents the modeling approach developed to estimate the boundary model representations of impact flashes. The research presents generalized …

Fusion Schemes For Ensembles Of Hyperspectral Anomaly Detection Algorithms, Brooks R. Turnquist

Theses and Dissertations

Hyperspectral imaging is playing an ever increasing role in our military's remote sensing operations. The exponential increase in collection operations generates more data than can be evaluated by analysts unassisted. Anomaly detectors attempt to reduce this load on analysts by identifying potential target pixels which appear anomalous when compared to what are determined to be background, or non-target, pixels. However, there is no one individual algorithm that is best suited for all situations and it can be difficult to choose the best algorithm for each individual task. Fusion techniques have been shown to reduce errors and increase generalization, eliminating the …

Tailoring The Statistical Experimental Design Process For Lvc Experiments, Casey L. Haase

Theses and Dissertations

The use of Live, Virtual and Constructive (LVC) Simulation environments are increasingly being examined for potential analytical use particularly in test and evaluation. The LVC simulation environments provide a mechanism for conducting joint mission testing and system of systems testing when scale and resource limitations prevent the accumulation of the necessary density and diversity of assets required for these complex and comprehensive tests. The statistical experimental design process is re-examined for potential application to LVC experiments and several additional considerations are identified to augment the experimental design process for use with LVC. This augmented statistical experimental design process is demonstrated …

A Methodology Using Simulation Results For Test And Evaluation, Jonathon S. Hosket

Theses and Dissertations

Each year the Air Force spends billions of dollars on Test and Evaluation to ensure acquisition programs roll out the best possible products. In 1997, the National Research Council assembled to evaluate the overall procedure used in procuring various platforms with system planning, research, development and engineering (SPRDE) and program management (PM) processes. In their final report, they claimed that the full advantages of statistical practices, simulation, model-test-models, and incorporation of prior test information into current test practices have not been fully utilized. To examine one of the report's recommendations, this thesis defines and explores a methodology using simulation to …

Weighted Key Player Problem For Social Network Analysis, Ryan M. Mcguire

Theses and Dissertations

Social network analysis is a tool set whose uses range from measuring the impact of marketing campaigns to disrupting clandestine terrorist organizations. Social network analysis tools are primarily focused on the structure of relationships between actors in the network. However, characteristics of the actors, such as importance or status, are generally the output of the social network analysis rather than an input. Characteristics of actors can come from a number of sources to include information gathering, subject matter experts or social network analysis. Further, the strength of relationships between actors in social networks are often assumed to be all equal. …

Utilizing Social Network Analysis In Support Of Nation Building, Brandon J. Bernardoni

Theses and Dissertations

Social network analysis is a powerful set of techniques used by social scientists to study the formal and informal interrelations in a community. Since 9/11 these techniques have been increasingly utilized by the defense and intelligence communities to analyze terrorist networks to aid in thwarting our foes. This study investigates the use of social networks and structural holes to facilitate nation building in failed and failing states. Investigation of the underlying social structure of a community, identifying st ructural holes and gaps within the government, Security Stabilization Transition and Reconstruction Operations (SSTRO) efforts can be focused to strengthen the host …

Determining Pilot Manning For Bomber Longevity, Jason S. Hamilton

Theses and Dissertations

In support of US Air Force efforts to conserve resources without sacrificing capability, this research examines the question of whether the 509th Bomb Wing could continue to provide maximum combat capability with fewer assigned pilots. During peacetime, pilot proficiency training comprises the majority of annual flying hours for the small B-2 bomber fleet. Optimal pilot manning will decrease the accumulation of excess wear on the airframes; helping to extend the viable life of the B-2 fleet and preserve the deterrent and combat capabilities that it provides to the United States. The operations and maintenance activity flows for B-2 aircraft and …

Measure Of Effectiveness For Jstars Ground Moving Target Indicator: A Value Focused Thinking Approach, Gardner J. Joyner

Theses and Dissertations

As the nature of warfare has shifted from a conventional approach to more guerilla type warfare, intelligence has become more important than at any other time in the history of the United States Military. With the stochastic nature of intelligence gathering, it is almost impossible to know with any degree of certainty where and when the next piece of information that could possibly change the course of the battle or war will be obtained. US intelligence gathering assets have long been plagued with using useless measures of performance rather than measures of effectiveness to determine their worth. This research uses …

Multi Criteria Decision Support Model For The Turkish Air Force Personnel Course/Education Planning System, Cem Malyemez

Theses and Dissertations

Many personnel are assigned to different courses and educational opportunities in the Turkish Air Force every year. This daunting process requires a great deal of time and does not currently seek to find an optimal solution. The Turkish Air Force has a database for all personnel information, but the officials use just a few quantitative personnel data points to complete their tasks. Moreover, the matching process is done by hand. Therefore there is a need for a model, which supports the Decision Makers, to cover the some of the other quantitative data points and also add some qualitative data for …

Alternative Active Duty Military Retirement Plan, Amy L. Schmidt

Theses and Dissertations

No abstract provided.

Determining The Value Of Future Information, James J. Springston

Theses and Dissertations

Companies continuously struggle to quantify the value of their information in an attempt to gain a better understanding of the return on investment of their information technology (IT) architecture. One approach companies have taken to place a numeric value on information is to treat it as a traditional economic asset (e.g. equipment, buildings, and vehicles) that is governed by its own unique set of laws. Once an enterprise understands the behavior of information it can incorporate Skyrme's "10 value adding aspects of information" when developing IT architecture, thus maximizing the potential value of their information. Like most enterprises, the Intelligence …

Migration Issues In The Democratic Republic Of Congo, Michael J. Thompson

Theses and Dissertations

Since the end of the Cold War, the number of weak and failing states has increased significantly. The United States (US) military has been deployed in multiple nation states in an effort to prevent these weak states from collapsing into chaos. This thesis explores one of the driving factors of state collapse, net migration, to gauge how US foreign policy might be employed to reduce the flow of people out of a country. To demonstrate the foreign policies and their effects, a pilot model was constructed using a system dynamics methodology. The Democratic Republic of Congo (DRC) was selected as …

Optimizing Ground Times For Amc Aircraft In Afghanistan, Eric W. Bucheit Jr.

Theses and Dissertations

This research provides insight into the cargo unloading and loading process and times at three airfields in Afghanistan. Using a linear regression model and its prediction capability, a more stabilized environment is produced. This uses the times from seven historical months of continuous operations in Afghanistan and uses this information for the prediction expressions. Using this model, no throughput is lowered and early and late times are lowered by an average of seven minutes per mission. Over the course of a month, this increases planning stability by 74.1 hours. This not only impacts the planning at the downrange location, but …

Ambulatory Healthcare Utilization In The United States: A System Dynamics Approach, Rafael Diaz, Joshua G. Behr, Mandar Tulpule

VMASC Publications

Ambulatory health care needs within the United States are served by a wide range of hospitals, clinics, and private practices. The Emergency Department (ED) functions as an important point of supply for ambulatory healthcare services. Growth in our aging populations as well as changes stemming from broader healthcare reform are expected to continue trend in congestion and increasing demand for ED services. While congestion is, in part, a manifestation of unmatched demand, the state of the alignment between the demand for, and supply of, emergency department services affects quality of care and profitability. The central focus of this research is …

Artificial Neural Networking As A Decision Tool For Natural Gas Investment, Micah D. Denecour

Master's Theses

With the growing interest in the Marcellus Shale and its natural gas deposits, there are opportunities to purchase and hold land for investment purposes. A robust decision tool is needed to help guide investors towards the most profitable properties. Artificial neural networks have many unique benefits that make them an ideal candidate for this purpose.

The artificial neural networks created in this study had nine independent variables. Combinations of these nine variables were created to describe 300 theoretical properties available for purchase. Each of these properties were then evaluated by an expert in the field and given a score from …

A Microsimulation Of Traffic, Parking, And Emissions At California Polytechnic State University – San Luis Obispo, Steven Michael Kilbert

Master's Theses

Traffic and parking congestion are significant issues at many universities nationwide. The delays experienced result in wasted time, money, and fuel for students, faculty and staff, not to mention the negative contributions to the environment. This paper quantifies the amount of vehicle emissions generated during an average morning peak hour in the university environment. Using VISSIM and CMEM microsimulation packages, a model is created for California Polytechnic State University- San Luis Obispo to aggregate the collective transportation behaviors and practices of the campus and recognize the implications these behaviors pose on the transportation network as a whole. Reasonable estimates are …

Introduction: Advances In E-Business Engineering, Ling Li

Information Technology & Decision Sciences Faculty Publications

(First paragraph) E-business is more than just e-commerce. It is one of the most challenging areas for industry and research communities. E-business has evolved from business-to-business, business-to-customer, customer-to-business, customer-to-customer, and business-to-government systems to the integrated and collaborative business services among various information systems and e-marketplaces. In this evolving process, integrated e-business systems and their related supporting platforms have to be rapidly designed and developed in order to meet different requirements. A variety of e-business engineering paradigms and technologies have been developed to tackle these challenges. There are many research issues needed to be addressed. These issues include heterogeneous services integration, …

Self Learning Strategies For Experimental Design And Response Surface Optimization, Adel Alaeddini

Wayne State University Dissertations

Most preset RSM designs offer ease of implementation and good performance over a wide range of process and design optimization applications. These designs often lack the ability to adapt the design based on the characteristics of application and experimental space so as to reduce the number of experiments necessary. Hence, they are not cost effective for applications where the cost of experimentation is high or when the experimentation resources are limited. In this dissertation, we present a number of self-learning strategies for optimization of different types of response surfaces for industrial experiments with noise, high experimentation cost, and requiring high …