A SYSTEM OF SYSTEMS APPROACH TO ENTERPRISE ANALYTICS DESIGN: ACQUISITION SUPPORT IN THE AGE OF MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE
Abstract
This research proposal by Dr. Daniel DeLaurentis at Purdue University aims to improve the way that decision-makers deploy state-of-the-art analytic technologies across relevant parts of an enterprise, in support of its acquisition activities. Literature backed evidence suggests the need to go beyond factors that are traditionally considered (e.g. data inputs to an algorithm or software requirements for the algorithm) in the deployment of these technologies and that a lack of such considerations can lead to lost opportunities and undesirable outcomes. Furthermore, the complex relationships that exist among and between such factors (e.g. stakeholder incentives, informational contexts of data, technical dependencies, and system behaviors) makes the task of selecting analytic deployments in this proposed context, a non-trivial undertaking. The objective of the proposed research is to develop a practical and systematic way of identifying optimal enterprise-level analytic deployments, when considering the broader spectrum of factors that contribute to their effectiveness in supporting acquisition activities - this includes establishing effective descriptors of what constitutes an enterprise analytic deployment, how the disparate types of dependencies between parts of the enterprise are related, and, how to bring to bear relevant methods of analysis on such descriptions to enable objective selection of an optimal deployment. To accomplish this, the researchers will identify the most pertinent factors and associated variables of interest that contribute to the effectiveness of analytic deployments across an enterprise. The researchers will also conduct an extensive literature review and leverage theories from fields of study that include systems engineering (theories and methodologies on the development of systems), system of systems engineering (theories on developing collections of systems that work together to achieve something greater than the sum of their parts) and operations research (strategies on how to optimize the performance of systems - includes use of analytics). The team will adopt relevant perspectives from each field to synthesize a practically motivated framework, that helps decision-makers to objectively determine optimal deployments of enterprise level analytics, in support of acquisition activities. The research will produce artifacts that other researchers can leverage and use as well as demonstrate value to practitioners through a relevant conceptual acquisition problem. Specific outcomes of the work include the synthesis of new knowledge on integrating theories from multiple disciplines of research towards developing an impactful foundation for the design of enterprise analytic deployments. Additional outcomes include demonstrative examples of application of the research to illustrate utility in the approach. Our proposed research is domain agnostic, data agnostic, and has a very wide range of possible applications across public, private and government enterprises.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Oct 29, 2021
- Source ID
- HQ00341910007
Entities
People
- Daniel Delaurentis
Organizations
- Office of the Secretary of Defense
- Purdue University
- Washington Headquarters Services