Abstract for: Evolutionary Systems Thinking: From Equilibrium Models to Open-Ended Adaptive Dynamics

We routinely describe markets, institutions, and technologies as "evolving," yet most system dynamics models remain confined to fixed state spaces and equilibrium-seeking behavior. This paper introduces Stability-Driven Assembly (SDA), a minimal framework in which selection emerges from persistence-weighted feedback—without genes, replication, or fitness functions. We examine why equilibrium models cannot exhibit open-ended evolution, and propose two methods for extending current practice. SDA adds just two operations to random assembly: remove expired patterns and assign persistence based on stability. This closes a feedback loop between persistence, population composition, and future interaction. The dynamics can be formalized as a Fokker-Planck equation with population-dependent drift, where accumulated structure reshapes the effective selection landscape. Simulations confirm that persistence-weighted feedback produces selection-like behavior. Compared to unconstrained controls, SDA systems exhibit sustained entropy reduction and concentration of population mass on stable configurations. The dynamics show punctuated structure: quasi-stable regimes persist until a higher-stability motif emerges and reorganizes population flows. Fitness-proportional sampling arises as an emergent property, realizing a "natural genetic algorithm" driven by persistence rather than explicit fitness functions. On a practical level, we propose two methods for introducing evolutionary dynamics into system dynamics practice. First, build SDA simulations from scratch: define base entities, interaction rules, and stability criteria, then observe what structures persist under realistic churn. Second, stress-test existing model equilibria by injecting candidate disruptors with varying stability values and mapping which trigger reorganization. Both extend system dynamics toward open-ended adaptive modeling. Editing and English