The Need for a Performance Map Based Decision Process

This paper presents a mathematical framework to create a unique decision process based on multiple performance maps of varying functional value and volatility to aid in complex decisions for highly nonlinear and coupled systems by involving human visualization and direct algebraic computations. This map-based process is intended to extend other decision tools, such as fuzzy logic and classical optimization (through cost functions) by embedding in-depth measurement certification, lessons learned, and data uncertainty in geometric map representations. As systems become ever more complex with more human intervention, reliable decisions must be made in less and less time (i.e., a few milliseconds). This decision framework is best carried out with selected maps combined into envelopes as functional/proven decision surfaces. Its potential for data-driven decisions of broad applicability is illustrated in several unique domains ranging from an intelligent actuator, a sensored soldier, health care provider management, a venture capital firm, and in human guided design/synthesis processes for parametrically dense systems. We also show that there is a geometrical structure (serial, parallel, or hybrid) to these decision processes which guides the efficient formulation of the most useful forward or inverse computation.