Predicting Future States of Evolving Systems

Abstract

Abstract Systems are characterized by continuous changes that are driven by both the evolving nature of the system itself and external environmental factors. To design systems for the future, it is critical to understand the behavior of these systems, the interactions between variables, and the emergent phenomena that arise in evolving systems. Designers must possess the ability to incorporate this understanding into their designs, to ensure robust solutions for the problems at hand. In response to the capabilities required to design future system states, we propose a method that includes System Dynamics to simulate the system and the compromise Decision Support Problem Construct (cDSP) to obtain robust satisficing solutions. Through this method we envision enabling designers design a system for required system goals by initially simulating the system for the present and past states and using this information of the change in behavior of the system to attain robust satisficing solutions for systems goals of the future state. In our method, we use System Dynamics to simulate the system and obtain the values of the variables for different states. Based on these values, we formulate and solve a cDSP to find robust satisficing solutions. We account for the trade-off between different variables. Based on the information obtained from present and past states we establish a satisficing solution space for the future state. In this paper, we demonstrate the efficacy of our method using a simple problem to gain clarity on the process and the outcome. Our focus in this paper is on describing the method rather than the outcomes associated with the test problem.