Please use this identifier to cite or link to this item:
https://dair.nps.edu/handle/123456789/4922
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Johnathan Mun | - |
dc.date.accessioned | 2023-05-07T01:07:00Z | - |
dc.date.available | 2023-05-07T01:07:00Z | - |
dc.date.issued | 2023-05-01 | - |
dc.identifier.citation | APA | en_US |
dc.identifier.uri | https://dair.nps.edu/handle/123456789/4922 | - |
dc.description | SYM Presentation | en_US |
dc.description.abstract | How were the decisions made in the past, and what were the drivers, strategies, or rationale? The old adage holds true on how organizations should learn from the past to help make better decisions in the future. This current first-phase research looks at how the Department of Defense (DOD) can inculcate institutional and corporate memory. Specifically, the research tests and develops recommendations about how a transparent Decisions Options Register (DOR) integrated intelligent database system can be developed, where the DOR helps capture all historical decisions (assumptions, data inputs, constraints, limitations, competing objectives, and decision rules) for programs within the Department of Defense (DOD). Information in this DOR will be compatible with meta-semantic searches and data science analytical engines. The DOR is used for modeling future decision options to enable making decisions under uncertainty while leaning on past best practices and allowing senior leadership to make defensible and practical decisions. The current first phase of research uses stylized data and examples to illustrate the recommended methodologies. This research implements industry best-in-class decision analytics using advanced quantitative modeling methods (stochastic simulation, portfolio optimization) coupled with Artificial Intelligence (AI) and Machine Learning (ML) algorithms (data scraping, text mining, sentiment analysis) and Enterprise Risk Management (ERM) procedures. The DOR will be partially based on ERM methods of using risk registers, where different risk elements are subdivided into different GOPAD groups or Goals (military capability, cost savings, novel technology, future weapons capability, public safety, government priorities, command preference, etc.), Organization (Air Force, Army, Navy, Marines), Programs (acquisition, commercial-off-the-shelf, joint-industry, hybrid, etc.), Activity (inventory, replacement, new development, research and development, and so forth), and Domain (air, sea, cyber, etc.) categories. Multiple competing stakeholders (e.g., the Office of the Secretary of Defense, Office of the Chief of Naval Operations, the U.S. Congress, and the civilian population) have their specific objectives (e.g., capability, efficiency, cost-effectiveness, competitiveness, and lethality, as well as alternatives and trade-offs), constraints (e.g., time, budget, schedule, and manpower), and mission-based domain requirements (e.g., balancing the needs of digital transformation in cybersecurity, cyber-counterterrorism, anti-submarine warfare, anti-aircraft warfare, or missile defense). This research takes a multidisciplinary approach where methods from advanced analytics, artificial intelligence, computer science, decision analytics, defense acquisitions, economics, engineering and physics, finance, options theory, project and program management, simulation with stochastic modeling, applied mathematics, and statistics are applied. The ultimate goals are to provide decision-makers actionable intelligence and visibility into future decision options or flexible real options, complete with the assumptions that led to certain comparable decisions. The recommended approaches include the use of supervised and unsupervised AI/ML sentiment text analysis, AI/ML natural language text processing, and AI/ML logistic classification and support vector machine (SVM) algorithms, coupled with more traditional advanced analytics and data science methods such as Monte Carlo simulation, stochastic portfolio optimization and project selection, capital budgeting using financial and economic metrics, and lexicographic rank approaches like PROMETHEE and ELECTRE. Example case applications, code snippets, and mock-up DORs are presented, complete with stylized data to illustrate their capabilities. The current research outcome will provide a stepping stone to the next phase’s multiyear research, where prototypes can be built and actual data can be run through the prescribed analytical engines. | en_US |
dc.description.sponsorship | Acquisition Research Program | en_US |
dc.language.iso | en_US | en_US |
dc.publisher | Acquisition Research Program | en_US |
dc.relation.ispartofseries | Acquisition Management;SYM-AM-23-155 | - |
dc.subject | Artificial Intelligence | en_US |
dc.subject | Machine Learning | en_US |
dc.subject | Data Science | en_US |
dc.subject | Making Better-Informed Decisions | en_US |
dc.title | Management and Business Knowledge Representation for Decision Making: Applying Artificial Intelligence, Machine Learning, Data Science, and Advanced Quantitative Decision Analytics for Making Better-Informed Decisions | en_US |
dc.type | Presentation | en_US |
Appears in Collections: | Annual Acquisition Research Symposium Proceedings & Presentations |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
SYM-AM-23-155.pdf | 4.22 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.