Development and Application of a Simulation Environment (NEO) for Integrating Empirical and Computational Investigations of System-Level Complexity

Network Exchange Objects (NEO) is a new software framework designed to facilitate development of complex natural or built distributed system models, where the system model is represented as a graph, through which currencies (e.g., coding information) flux. This paper introduces "system-level hypothesis (SLH) testing" as a form of computational thinking that will drive integration of computational and empirical sciences to promote efficient, self- correcting inquiry into the operations and behavior of complex systems. To demonstrate NEO, we examine the problem of maximizing the productivity of a software development organization by measuring growth in the total lines of code (LOC) contributed by developers. We develop a software framework (NEO) that allows rapid creation of model variants representing alternative SLHs. NEO is designed to investigate systems we describe as "complex adaptive hierarchical networks" (CAHNs - complex systems represented as networks that route and store multiple interactive currencies). Models built atop NEO, are organized collections of individual values (model variables) and algorithms (model logic). Modelers systematically combine algorithms to create alternative model formulations at runtime. Thus, NEO is a simulation framework that can be used in any domain of expertise, where systems are represented as interdependent entities that store and flux multiple currencies.