A framework for developing executable architecture for aerial intelligence surveillance and reconnaissance systems-of systems: A systems dynamics approach

Abstract

A framework for the development of holistic executable architecture for complex systems-of-systems as a method to analyze means and ways trades across the DOTmLP-P (Doctrine, Organization, Training, materiel, Leadership, Personnel, and Policy) spectrum in support of CBA (Capabilities Based Assessment) was proposed. The Aerial Intelligence, Surveillance, and Reconnaissance (AISR) system-of-systems architecture, complete with reachback for the Processing, Exploitation, and Dissemination (PED) of intelligence products, and utilizing a generic Unmanned Aircraft System in support of long-range precision fires as part of a Decide-Detect-Deliver-Assess (D3A) operation was used a case study to experiment and test the framework. Initial application of systems thinking was utilized reduce system complexity through fundamental understanding of elements, connections, and functions of the systems-of-systems. System Dynamics was selected as an appropriate modeling and simulation paradigm. Static Department of Defense Architecture Framework (DoDAF) models were mapped to key characteristics of a System Dynamics stock-and-flow model with causal loops to create executable architecture to assess feedback loops and interactions over time. A baseline PED system model with standardized inputs was recreated from previously published research and used to demonstrate the ability to evaluate structure, policy, and manning changes to the architecture using Monte Carlo Simulations, sensitivity analysis, and Powell optimization. The PED model was then integrated into the larger executable architecture which included the AISR and D3A fires systems and a parametric analysis was conducted on the variables of interest. Random distribution functions were applied to replicate the effects of combat and assess the system-of-systems against operational measures of performance and effectiveness. Space filling design of experiments with point repetition was applied to sample the input space and accounting for noise variations from random effects. This enabled the creation of surrogate models for statistical assessments of complex interactions between the elements of interest against desired operational outputs. The experiments successfully supported the use of System Dynamics as a means to holistically assess complex systems-of-systems in a rapidly developed, interactive environment that enables trades and could alternatively be used as a gaming and learning tool for stakeholders and decision makers. Findings concerning benefits and limitation are discussed as well as recommendations for future work and improvements.