Operational Research Commons^™

A Parametric Regression Of The Cost Of Base Realignment Action (Cobra) Model, Douglas D. Hardman, Michael S. Nelson

Theses and Dissertations

This study develops a parametric model that is capable of generating accurate estimates of the costs to close Air Force installations. The new model is based upon, but much simpler to use than, the Cost of Base Realignment Action (COBRA) model. COBRA is an economic cost analysis model that requires a minimum of 250 inputs and as many as 700 inputs. The new parametric model requires just 10 input variables and was developed using least squares multiple regression. Comparison of the new parametric model to COBRA indicates that it captures 91 percent of the variance in cost estimates generated by …

A Comparison Of Variable Selection Criteria For Multiple Linear Regression: A Second Simulation Study, David P. Woollard

Theses and Dissertations

This thesis implements a variable selection method proposed by Alan J. Miller, and makes an extension of Ross J. Hansen's 1988 thesis research by comparing the methods he examined: Minimum MSE, Minimum Sp, and Minimum Cp with Miller's method. Response Surface methodology is employed with two performance measures: the percentage of correct variables in a model and the Theoretical Mean Squared Error of Prediction (TMSEP). Each technique is applied on generated data with known multicollinearities, variances, random predictors, and sample sizes. Both performance measures are computed for models selected under each technique. A full factorial design using each performance measure …

Modified Anderson-Darling And Cramer-Von Mises Goodness-Of-Fit Tests For The Normal Distribution, David A. Gwinn Sr.

Theses and Dissertations

New techniques for calculating goodness-of-fit statistics for normal distributions with parameters estimated from the sample are investigated. Samples are generated for a Normal(0,1) distribution. Critical values are calculated for five modifications to the Anderson-Darling statistic and five modifications to the Cramer-Von Mises statistic. An extensive power study is done to test the power of the new statistics versus the power of the unmodified statistics. Powers of six of the new statistics show minimal to no improvement, two of the new statistics show a marked decrease in power, and two of the new statistics show an overall increase in power over …

Including Maximum Sustained Wind Speed In A Time Series Model To Forecast Hurricane Movement, Timothy B. Mott

Theses and Dissertations

Techniques for applying time series fundamentals to forecasting hurricane movement are thoroughly examined in this research. The objectives are: (1) to modify Dr. Thomas Curry's threshold autoregressive time series model to improve its ability to forecast hurricane movement, (2) to forecast the maximum sustained wind speed for a hurricane, and (3) to identify if wind speed should be included as an explanatory variable to aid in forecasting hurricane movement. Eleven different models to predict the latitude, longitude and maximum sustained wind speed are compared and contrasted with Curry's bivariate time series model. The results showed the modifications allow significant forecasting …

A Modified Anderson Darling Goodness-Of-Fit Test For The Gamma Distribution With Unknown Scale And Location Parameters, Tamer Ozmen

Theses and Dissertations

A new modified Anderson-Darling goodness-of-fit test is introduced for the three-parameter Gamma distribution when the location parameter is found by minimum distance estimation and scale parameter by maximum likelihood estimation. Monte Carlo simulation studies were performed to calculate the critical values for A-D test when A-D statistic is minimized. These critical values are then used for testing whether a set of observations follows a Gamma distribution when the scale and location parameters axe unspecified and are estimated from the sample. Functional relationship between the critical values of A-D is also examined for each shape parameter by the variables, sample size …

Optimal Silo Attack Plan For The Red Integrated Strategic Offensive Plan (Risop), Richard C. Pace

Theses and Dissertations

The Joint Staff Directorate for Force Structure, Resources, and Assessment (J-8) sought a procedure which could be used to generate an optimal missile allocation for the silo attack portion of the Red Integrated Strategic Offensive Plan (RISOP). Their current solution procedure is a manual heuristic which is time-consuming and is not guaranteed to lead to an optimal solution. J- 8 defines an optimal solution as a feasible solution which minimizes both the flight time of the missile that impacts first and the duration of the attack. J- 8 defined several input rules which limit how missiles may be allocated. A …

A Modified Chi-Squared Goodness-Of-Fit Test For The Three-Parameter Gamma Distribution With Unknown Parameters, Thomas J. Sterle

Theses and Dissertations

A modified chi-squared goodness-of-fit test was created for the gamma distribution in the case where all three parameters must be estimated from the sample. Critical values are generated using a Monte Carlo simulation procedure with 5000 repetitions each. Random samples of 8 different sizes were drawn from gamma distributions with shape parameters 1, 1.5, 2., and 2.5. The shape, scale, and location parameters were then estimated from each sample, using an iterative technique combining the maximum likelihood and minimum distance methods, enabling, computation of the chi-squared statistics and critical values. The same process is used to generate random samples, parameter …

An Analysis Of The Generalized Lambda Distribution, Robert J. Bergevin

Theses and Dissertations

The Generalized Lambda Distribution (GLD) is a four parameter function that is capable of mimicking the behavior of a wide range of probability density functions (pdfs). Unfortunately, the GLD presently cannot model every possible type of pdf. Since the reasons for this limitation are unknown, this thesis examines several potential problems in an attempt to expand the range of distributions the GLD can mimic. We first present a discussion of the behavior of the algorithm that is used to search for the appropriate GLD parameter values. In particular, we examine the effect of using an unconstrained search to find the …

A New Goodness-Of-Fit Test For The Weibull Distribution Based On Spacings, Mark C. Coppa

Theses and Dissertations

The critical values for a new goodness-of-fit test based on spacings are generated for the Weibull distribution when the shape parameter is known. The critical values are used for testing whether a set of observations follow a Weibull distribution when the scale and location parameters are unknown. A Monte Carlo simulation with 10,000 iterations is used to generate the critical values for sample sizes 5(5)35 at shape parameters k equal to 0.5(0.5)1.5 and for sample sizes 5(5)20 at shape parameters k = 2.0(1.0)4.0. A Monte Carlo power study of the Z* test statistic using 5000 iterations is accomplished using nine …

A Fuzzy Logic Optimal Control Law Solution To The Cmmca Tracking Problem, Randy E. Nelson

Theses and Dissertations

The Air Force uses the C-18 Cruise Missile Mission Control Aircraft (CMMCA) to radar track cruise missiles (CM) during test flights. Because of the complexity of the CM flight profiles, maintaining radar coverage at all times is very difficult. This thesis attempted to apply optimal control theory to construct a simulation providing 100% radar coverage. The simulation was divided into ten second intervals, and fuzzy logic was used at the start of each interval to determine the set point, i.e., that point in space where the CMMCA should be in ten seconds. The set point calculation's fuzzy logic balanced CMMCA …

Multidimensional Approach To Implementation Of Continuous Process Improvement Utilizing Small Run Analysis In A Shipyard Environment, Paul Ray Hollandsworth

Engineering Management & Systems Engineering Theses & Dissertations

Nationally, both public and private shipyards are experiencing a high rework rate (over 40%) in the fabrication of plug connectors used for electronic systems onboard Navy vessels. The shipyard's survival depends on acquiring the elements necessary to maintain a major industry in a position of continuous improvement that produces a better product than their competitors, at less cost, in a safe manner, and on schedule.

This research addresses the complex problem of how to introduce quality control methodologies within a job shop environment that assists in identifying variables that influence the rework rate of plug fabrication process over a four …

Recurrent Neural Networks For Radar Target Identification, Eric T. Kouba

Theses and Dissertations

A real-time recurrent learning algorithm was applied to a five class radar target identification problem. The wideband radar was assumed to measure both kinematic (tracking information expressed as estimated aspect angles) and high range resolution data from a single, isolated aircraft. The aspect angles (azimuth and elevation) of the aircraft relative to the radar were assumed to be constantly chancing. This created temporal sequences of high range resolution radar signatures that changed as the aspect angles changed. These sequences were used as input features to a recurrent neural network for three radar target identification test cases. The first test case …

Multilayer Perceptrons For Classification, Lisa M. Belue

Theses and Dissertations

Techniques for training, testing, and validating multilayer perceptrons are thoroughly examined. Results obtained using perceptrons are compared and contrasted with two multivariate discriminant analysis techniques- logistic regression and k neighbor. Methods for determining significant input features are investigated and a procedure for examining the confidence to place in the significance of these features is developed. Procedures to evaluate the applicability of high-order feature inputs are examined. These methods and procedures are applied to two very different applications. The first application concerns the prediction of Air Force pilot retention/separation rates for input to force projection models. The second application concerns the …

Transformation Of Combat Data In Support Of Battle Trace, Hyoung-Kyu Choi

Theses and Dissertations

The purpose of this study was to find an appropriate way of characterizing battle data as a function of time. This study was initially designed to remove the numerical instability problem cited in Barr's battle trace methodology as suggested by TRAC-MTRY. The research focused on the instability problem and identifying/recommending a technique that improved the efficiency of computation, and enhanced an analyst's ability to meaningfully interpret the battle trace. Two Lanchester's Square Law based methodologies are introduced and analyzed. The results of this analysis indicate that the battle trace of a constant value generated by Lanchester's Square Law integration seemed …

Causal Univariate Spatial-Temporal Autoregressive Moving Averages (Starma) Modelling Of Target Information To Generate Tasking Of A World-Wide Sensor System, Kelly A. Greene

Theses and Dissertations

The Department of Defense employs a resource limited world-wide sensor system to detect certain events of interest. The purpose of this research was to establish a methodology using a univariate causal STARMA model for forecasting the relative probability of an event occurring in a geographical location during a time block of the day. These relative probabilities are used as input for a tasking model that assigns the scarce sensor resources so as to optimize the detection of these events. The STARMA model is appropriate for forecasting the relative probabilities because a definite temporal relationship and a definite spatial relationship exists …

Confidence Interval Estimation For Output Of Discrete-Event Simulations Using The Kalman Filter, Randall B. Howard

Theses and Dissertations

Discrete-event simulation is computer modeling of stochastic, dynamic systems. The Kalman filter is a Bayesian stochastic estimation algorithm. Because of the correlated nature of simulation output, it is difficult to apply the methods of classical statistics directly when constructing confidence intervals of discrete-event simulation parameters. Through the determination of a dynamics equation and application of the Kalman filter to simulation output data, three new confidence interval construction techniques have been developed. One technique obtains an estimate of the mean value and its associated variance from an estimated Kalman filter. The second technique utilizes Multiple Model Adaptive Estimation (MMAE) techniques to …

A Sensitivity Analysis Of Circular Error Probable Approximation Techniques, Peter Puhek

Theses and Dissertations

Several algebraic CEP estimates models were examined in this study. Each assumes that the crossrange and downrange miss distances of the sample data follow a bivariate normal distribution. The analysis determined the sensitivities of these models to changes in the parameters of sample size, bias, correlation, and ellipticity. The accuracy of each model is expressed in terms of relative error, and the parameter regions in which a certain method dominated as the most accurate were noted. In general, it was found that bias was the most significant parameter in determining the best CEP method. A simple method, based on the …

Radar System Classification Using Neural Networks, David M. Cameron

Theses and Dissertations

This study investigated methods of improving the accuracy of neural networks in the classification of large numbers of classes. A literature search revealed that neural networks have been successful in the radar classification problem, and that many complex problems have been solved using systems of multiple neural networks. The experiments conducted were based on 32 classes of radar system data. The neural networks were modelled using a program called the Neural Graphics Analysis System. It was found that the accuracy of the individual neural networks could be increased by controlling the number of hidden nodes, the relative numbers of training …

A Network Flow And Goal Programming Approach To Modeling The Impact Of Pre-Accession Training To The Trained Personnel Requirements Process, William J. Beveridge

Theses and Dissertations

HQ ATC was tasked to analyze the impact of pre-accession training to the Trained Personnel Required (TPR) and training production process. Pre- accession training is a policy of providing contracted initial skills training to enlistees prior to entering basic military training. The purpose of this study was to develop a method to model the impact of pre-accession training. A network modeling and goal programming approach was used. Comparisons between the current training policy and a pre-accession training policy were made. Sensitivity was conducted on the impact of a balk rate on the new policy. The balk rate is the percent …

Mapping Efficient Numerical Methods To The Solution Of Multiple Objective Linear Programs, Michael A. Shields

Theses and Dissertations

This investigation was initiated to increase the speed, accuracy and capacity of m-simplex algorithms for solving multiple objective linear programming problems. Specifically, improvements were sought through the application of general numerical techniques. It soon became apparent that the m- simplex algorithm, like the simplex algorithm, is heavily dependent upon the technology of solving related systems of linear equations. The numerical arguments for the application of LU triangular matrix factorization techniques to simplex computations are well known. OF special significance to m-simplex performance is the case of rank-k updates to basis factorizations. A stable and efficient LU approach to the rank-k …

New Modified Anderson-Darling And Cramer-Von Mises Goodness-Of-Fit Tests For A Normal Distribution With Specified Parameters, Goksel Kahya

Theses and Dissertations

This thesis improves the powers of the Anderson-Darling and Cramer-von Mises goodness-of-fit tests using the mean rank and the median rank plotting positions. A Monte Carlo simulation of 5000 repetitions is used to generate the critical values from n =4 through n =80 for certain significance levels. An extensive power study is performed to compare the powers of the modified tests to the known tests for the uniform, the exponential, the double exponential, the lognormal, and the Weibull distributions. The power of the A-D tests modified by the median rank plotting and the mean rank plotting position are approximately 0.01% …

A Goodness-Of-Fit Test For A Family Of Two Parameter Weibulls With Known Shape Using Minimum Distance Estimation Of Parameters, John S. Crown

Theses and Dissertations

This research is to produce a modified Anderson-Darling goodness of fit test for the Weibull distribution when the location parameter is found by minimum distance estimation, the shape parameter is assumed known, and the MLE is used for the scale parameter. The critical values for the Anderson-Darling test are generated via Monte Carlo simulation when both the Anderson-Darling and Cramer-Von Mises distance statistics are minimized. These critical values are then used for a power study. The Monte Carlo simulation used 5000 repetitions for sample sizes of 5,8,12,15,16,20 and 25 with the Weibull shape parameter of . 5(.5)4.0. The power study …

Linear Programming By G.V. Shenoy, New Delhi: Wiley Eastern Limited, 1989., A K Rao

IIMB Management Review

No abstract provided.