Search

Many agencies within the Navy are competing for a portion of the Navy budget to ensure their agency receives the maximum amount of funding possible. With a budget of $255.8 billion for FY24, it is imperative the Navy does everything in its power to make the most informed investment decisions. Consider a department that allocates funds to several subdivisions each fiscal year. We use mechanical design theory to create an investment decision–based game that motivates each subdivision to report their annual budget prediction truthfully. Specifically, the department penalizes subdivisions for inaccurate budget predictions by giving the subdivision a smaller budget next year, and offers a bonus for subdivisions that predict their budgets accurately. We use Monte Carlo simulation to demonstrate our approach with a set of assumptions on how each subdivision behaves. The output of these simulation experiments provides a confidence interval for the budget each subdivision receives. Finally, the department can run stochastic optimization to determine the final budget allocation to each of its subdivisions to maximize the department’s overall performance.

Las armas menos letales (AML) han sido objeto de debate en la aplicación de la ley. Gran parte de la literatura se centra en su legalidad y conveniencia, en lugar de su efectividad, particularmente para los organismos de seguridad. Para abordar esa brecha, el presente estudio utiliza un análisis de riesgos financieros avanzados y simulación de escenarios para modelar el retorno de la inversión (ROI, por sus siglas en inglés) del uso de las AML. Utilizando un tipo de AML (The Alternative) como caso de estudio, se encuentra que, si bien los eventos extremos con AML ocurren relativamente con poca frecuencia, en comparación con las armas de fuego convencionales, su impacto en la sociedad es significativo. Por ejemplo, las agencias policiales pueden obtener un ROI entre el 1564% y el 2728%. Por cada dólar invertido en adquirir, equipar y capacitar a un oficial con AML, el beneficio adicional se sitúa entre 16 y 28 dólares. Por lo tanto, las AML no solo reducen el daño colateral, el riesgo reputacional y legal, sino que también previenen muertes innecesarias que erosionan la legitimidad policial y la confianza pública.

NPS NRP Technical Report, This proposed research has the goal of proposing novel, reusable, extensible, adaptable, and comprehensive advanced analytical process and Integrated Risk Management to help the (DOD) with risk-based capital budgeting, Monte Carlo risk-simulation, predictive analytics, and stochastic optimization of acquisitions and programs portfolios with multiple competing stakeholders while subject to budgetary, risk, schedule, and strategic constraints. The research covers topics of traditional capital budgeting methodologies used in industry, including the market, cost, and income approaches, and explains how some of these traditional methods can be applied in the DOD by using DOD-centric non-economic, logistic, readiness, capabilities, and requirements variables. Stochastic portfolio optimization with dynamic simulations and investment efficient frontiers will be run for the purposes of selecting the best combination of programs and capabilities is also addressed, as are other alternative methods such as average ranking, risk metrics, lexicographic methods, PROMETHEE, ELECTRE, and others. The results include actionable intelligence developed from an analytically robust case study that senior leadership at the DOD may utilize to make optimal decisions. The main deliverables will be a detailed written research report and presentation brief on the approach of capturing risk and uncertainty in capital budgeting analysis. The report will detail the proposed methodology and applications, as well as a summary case study and examples of how the methodology can be applied., N8 - Integration of Capabilities & Resources, This research is supported by funding from the Naval Postgraduate School, Naval Research Program (PE 0605853N/2098). https://nps.edu/nrp, Chief of Naval Operations (CNO), Approved for public release. Distribution is unlimited.

Existing Artificial Intelligence and Machine Learning (AI/ML) technologies can automate the filtering and accuracy of multiple data streams into the Navy's Common Operating Picture/Common Tactical Picture (COP/CTP). However, this Commercial Off The Shelf (COTS) software is not being leveraged effectively by the US Navy, specifically, the Information Warfare (IW) community for data fusion in support of track data management and targeting. Through analysis of current COTS AI/ML technologies, we will be able to posit how to optimize AI/ML software for accurate track data and disparate data source (such as GEOINT, radar data sets, and other imagery) fusion. We expect to find by completing a thorough analysis of track data and data sources input into AI/ML to fuse this informational data quickly and provide the most current/accurate intelligence. We will be able to proffer recommendations on how this all will be successfully accomplished. Additionally, the ROI of newly developed intelligence AI/ML technology (COP/CTP) will be evaluated. Once this research has been accomplished, specific improvements to various AI/ML algorithms for optimization may be examined and integration of any evaluated technology found during ROI research will be added.

This research has the explicit goal of proposing a reusable, extensible, adaptable, and comprehensive advanced analytical modeling process to help the U.S. Navy in quantifying, modeling, valuing, and optimizing a set of nascent Artificial Intelligence and Machine Learning (AI/ML) applications in the aerospace, automotive and transportation industries and developing a framework with a hierarchy of functions by technology category and developing a unique-to-Navy-ship construct that, based on weighted criteria, scores the return on investment of developing naval AI/ML applications that enhance warfighting capabilities. This current research proposes to create a business case for making strategic decisions under uncertainty. Specifically, we will look at a portfolio of nascent artificial intelligence and machine learning applications, both at the PEO-SHIPS and extensible to the Navy Fleet. This portfolio of options approach to business case justification will provide tools to allow decision-makers to decide on the optimal flexible options to implement and allocate in different types of artificial intelligence and machine learning applications, subject to budget constraints, across multiple types of ships. The concept of the impact of innovative technology on productivity has applicability beyond the Department of Defense (DoD). Private industry can greatly benefit from the concepts and methodologies developed in this research to apply to the hiring and talent management of scientists, programmers, engineers, analysts, and senior executives in the workforce to increase innovation productivity.

In this research, we answer the following primary question: Would an advanced analytical model be a more effective metric to estimate total ownership cost (TOC) with life-cycle cost under uncertainty and risk than the current method of life-cycle cost estimates for surface electro-optical infrared (EO/IR) sensors? To accomplish this, we developed and analyzed a computational model for Total Ownership with Life-Cycle Cost Model Under Uncertainty for Surface Electro-Optical Infrared Sensors. During the development of the model, we identified the required data and examined the current Department of Defense (DoD) method for determining system life-cycle costs for defense systems and determined that the proposed model is a useful alternative to the current method of determining the life-cycle costs for EO/IR sensors on surface ships. Finally, we concluded that the developed model can be applied to cost estimating in other sectors of DoD cost projections.

This study examines five advanced decision support methodologies—Lean Six Sigma (LSS), Balanced Score Card (BSC), Integrated Risk Management (IRM), Knowledge Value Added (KVA), and Earned Value Management (EVM)—in terms of how each can support the information technology (IT) acquisition process. In addition, the study provides guidance on when each methodology should be applied during the acquisition life cycle of IT projects. This research includes an in-depth review of each methodology in the context of the acquisition life cycle. All acquisition projects within the Department of Defense must go through the acquisition life cycle. While each acquisition project is unique, all must pass a series of common hurdles to succeed. Understanding how and when the methodologies can be applied to an IT acquisition is fundamental to its success. The study concludes with a set of recommendations for the use of each methodology in the acquisition life cycle of IT projects.

Autonomous systems provide the military with advanced capabilities permitting the execution of increasingly dangerous and difficult missions. A human in the loop is still required to decide how and when to deploy these technologies. The research problem this dissertation addresses is a user’s rejection of new technology in high-risk applications due to a lack of trust in the use of the technology. This is a problem because users’ lack of trust in new technology, designed to support users in high-risk situations, will prevent the use of the technology. The purpose of this dissertation is to investigate the effect of information presented about a new technology as well as the effect of the level of automation offered by the new technology on the perceived trust (i.e., as measured by the perceived usefulness, ease of use, and intent to use). An experiment was conducted that manipulated the level of automation and the presentation of information in three systems. The results indicated that, in high-risk scenarios, it is not possible to develop trust in technology without the system presentation of operational information. The study results also indicated that the level of automation was not a factor in developing technology trust in high-risk scenarios., Civilian, Department of the Navy, Approved for public release. distribution is unlimited

This dissertation explores knowledge friction. The concept knowledge flow is defined by many researchers as the transfer of actionable information between individuals, groups and organizations. Knowledge Flow Theory (KFT) frames and offers tools for conceptualizing, analyzing, visualizing, and measuring knowledge flows. Recently, Nissen conceptualized explicitness-based resistance to knowledge transfer; he referred to it as knowledge friction. This study addresses additional factors that inhibit knowledge transfer through an empirical look at Defense Acquisition University student surveys. These factors are clarity, near and longer-term relevance, certification level, and experience. The data was analyzed through descriptive statistics, multiple-regression analysis models, and a Partial Least Squares Structural Equation Model. The key findings: Clarity and near-term and longer-term relevance are quantitatively the largest contributing knowledge friction factors. These new factors interact with, but are also additive, to explicitness. Also discovered: The certification level and experience factors independently contribute little directly, but do increase the impact of longer-term relevance on knowledge transfer. Near-term relevance's impact is not affected by certification level and experience. These findings significantly contribute to KFT by extending and quantifying the factors that contribute to knowledge friction., Civilian, Department of the Navy, Approved for public release. Distribution is unlimited.

The acquisition of artificial intelligence (AI) systems is a relatively new challenge for the U.S. Department of Defense (DoD). Given the potential for high-risk failures of AI system acquisitions, it is critical for the acquisition community to examine new analytical and decision-making approaches to managing the acquisition of these systems in addition to the existing approaches (i.e., Earned Value Management, or EVM). In addition, many of these systems reside in small start-up or relatively immature system development companies, further clouding the acquisition process due to their unique business processes when compared to the large defense contractors. This can lead to limited access to data, information, and processes that are required in the standard DoD acquisition approach (i.e., the 5000 series). The well-known recurring problems in acquiring information technology automation within the DoD will likely be exacerbated in acquiring complex and risky AI systems. Therefore, more robust, agile, and analytically driven acquisition methodologies will be required to help avoid costly disasters in acquiring these kinds of systems. This research provides a set of analytical tools for acquiring organically developed AI systems through a comparison and contrast of the proposed methodologies that will demonstrate when and how each method can be applied to improve the acquisitions lifecycle for AI systems, as well as provide additional insights and examples of how some of these methods can be applied. This research identifies, reviews, and proposes advanced quantitative, analytically based methods within the integrated risk management (IRM)) and knowledge value added (KVA) methodologies to complement the current EVM approach. This research examines whether the various methodologies—EVM, KVA, and IRM—could be used within the Defense Acquisition System (DAS) to improve the acquisition of AI. While this paper does not recommend one of these methodologies over the other, ce

The U.S. Navy estimates that it will cost 16.7 billion per year for new-ship construction to become a fleet of 306 battle force ships over the next 30 years. It is critical that the Navy capture the full benefits of new technologies such as three-dimensional scanning 3DLS, product lifecycle management PLM, and additive manufacturing AM to reduce costs while still meeting mission needs. This project examines the use of 3DLS, PLM, and AM by non-shipbuilding industries as a basis for estimating potential naval shipbuilding savings. The research was conducted in two phases. In the first phase, secondary research was conducted on the three technologies used by various industries, while in the second phase, a model on the potential cost and efficiency savings that could be derived from the use of those technologies was developed. The ultimate goal is to develop insight into the amount and timing of technology costs and potential savings from use of the technologies in shipbuilding. Also, in phase two, recommendations are provided to Navy planners concerning the most effective and efficient strategy for exploiting these technologies. This preliminary report discusses some of the findings during phase one. It provides an overview of the Navys shipbuilding plans, and a framework for understanding of 3DLS, AM, and CPLM technologies, as well as potential applications to commercial shipbuilding., Naval Postgraduate School Acquisition Research Program

The acquisition of artificial intelligence (AI) systems is a relatively new challenge for the U.S. Department of Defense (DoD). Given the potential for high-risk failures of AI system acquisitions, it is critical for the acquisition community to examine new analytical and decision-making approaches to managing the acquisition of these systems in addition to the existing approaches (i.e., Earned Value Management). In addition, many of these systems reside in small start-up or relatively immature system development companies, further clouding the acquisition process due to their unique business processes when compared to the large defense contractors. This can lead to limited access to data, information, and processes that are required in the standard DoD acquisition approach. The well-known recurring problems in acquiring information technology automation within the DoD will likely be exacerbated in acquiring complex and risky AI systems. Therefore, more robust, agile, and analytically driven acquisition methodologies will be required to help avoid costly disasters in acquiring AI systems. This research provides a set of analytical tools for acquiring organically developed AI systems through a comparison and contrast of the proposed methodologies that will demonstrate when and how each method can be applied to improve the acquisitions life cycle for AI systems.

In this research, we look at answering the following primary question: Would an advanced analytical model be a more effective metric to estimate total ownership cost (TOC) with life-cycle cost under uncertainty and risk than the current method of life-cycle cost estimates for Surface EO/IR Sensors? To accomplish this, the research developed and analyzed a computational model for Total Ownership with Life-Cycle Cost Model Under Uncertainty for Surface Electro-Optical Infrared Sensors. During the development of the model, we identified the required data and examined the current Department of Defense (DoD) method for determining system life-cycle costs for defense systems and determined that the proposed model is a useful alternative to the current method of determining the life-cycle costs for EO/IR Sensors on surface ships. Finally, we concluded that the developed model can be applied to cost estimating in other sectors of DoD cost projections.

As autonomous systems become more capable, end users must make decisions about how and when to deploy such technology. The use and adoption of a technology to replace a human actor depends on its ability to perform a desired task and on the user's experience-based trust that it will do so. The development of experience-based trust in autonomous systems is expensive and high risk. This work focuses on identifying a methodology for technology discovery that reduces the need for experience-based trust and contributes to increased adoption of autonomous systems. Initial research reveals two problems associated with the adoption of high-risk technologies; 1) end user's refusal to accept new systems without high levels of initial trust and 2) lost or uncollected experience-based trust data. The main research hypothesis is that a trust score, or trust metric, can influence the initial formation of trust by functioning as a surrogate for experience-based trust, and that trust in technology can be measured through a probability-based prediction of risk.

As autonomous systems become more capable, end users must make decisions about how and when to deploy such technology. The use and adoption of a technology to replace a human actor depends on its ability to perform a desired task and on the user's experience-based trust that it will do so. The development of experience-based trust in autonomous systems is expensive and high risk. This work focuses on identifying a methodology for technology discovery that reduces the need for experience-based trust and contributes to increased adoption of autonomous systems. Initial research reveals two problems associated with the adoption of high-risk technologies; 1) end user's refusal to accept new systems without high levels of initial trust and 2) lost or uncollected experience-based trust data. The main research hypothesis is that a trust score, or trust metric, can influence the initial formation of trust by functioning as a surrogate for experience-based trust, and that trust in technology can be measured through a probability-based prediction of risk.

Logic regression (LR), not to be confused with logistic regression, is a well-known alternative tree-based method and powerful statistical learning technique that can be used to classify a binary response using Boolean combinations of binary predictors. In our case, given the binary states of the components of a network and its corresponding operating or failed status, LR can quantify the importance of the interactions of components according to their predictive capabilities (strength for classification). Meaning that, unlike traditional approaches in the reliability field, a completely different assumption is used. This paper shows the application of logic regression in six networks. Each example is characterized by a matrix representing the status of each component and a vector showing the corresponding network status. These data are analytically derived or using simulation procedures. The results show that LR could be considered as an additional assessment tool, where the most important effects (single or interactions) of components emerge naturally as a result of an optimization problem. As a byproduct, LR is also able to detect possible minimal cut/path sets.

In this paper, a novel application on how uncertainties in a manufacturing flow line − MFL (e.g., times required to perform an action) could be analyzed and what the benefits are of such analysis. The approach proposed investigates three main goals: i) Uncertainty analysis, ii) Stochastic dominance, and iii) Sensitivity analysis. In particular, this paper extends the application of max-plus algebra to model MFL with different flow configura- tions and buffer capacities and provides the approximated probability density functions (PDFs) of selected performance indicators (e.g., the total idle time in the whole line, output rates, throughputs, among others). As a result, it is possible to quantify the variability of the selected output, compare different possible configurations among MFL, choose the best one, and identify critical variables and risk drivers (e.g., the processing times that affect the most a KPI − key performance indicator). The approach, illustrated by analyzing a case study of the literature, emphasizes the benefits for a decision-maker in charge of the design or managing of the manufacturing system.