A Proposed Extension to the Functional Basis for AI/ML-Enabled Cyber-Physical Systems

Abstract Modern engineering projects increasingly require designers to use tools to manage complexity. One such tool, functional decomposition, uses the functional basis to aid designers in developing a structured and consistent description of a product or system in terms of functional requirements and desired behavior. However, cyber-physical systems (CPS) that increasingly integrate artificial intelligence (AI) and machine learning (ML) cannot be represented with the necessary degree of nuance and flexibility through the current functional basis. This is an inconvenience to designers especially in an age where the use of machine learning and artificial intelligence is becoming widespread. This paper proposes an extension to the functional basis that includes new flows and functions to better describe the intricacies of AI/ML systems. While this extension provides a significant advancement in representing AI/ML systems within the functional basis, the rapidly evolving nature of AI/ML technologies presents opportunities for further enhancements and expansions of this framework in the future. By creating a more comprehensive representation of cyber-physical systems with AI/ML capabilities, designers can better conceptualize and design these complex systems, facilitating a more consistent, structured, and descriptive functional model.