Evolutionary Compass: A Cross-Domain Framework for Evolutionary Problem Solving

Evolutionary progress in complex systems unfolds through organized change across time. Models are refined, policies are realigned, structures are reconfigured, external resources are incorporated, and the outcomes of change are consolidated, diversified, refreshed, or propagated into later rounds or broader contexts. Across disciplines, these processes are described through highly local vocabularies, e.g., fine-tuning, continual adaptation, transfer, policy updating, cumulative innovation, structural redesign, and self-modification. Evolutionary Compass provides a common design grammar for this broader methodological territory. As the paradigm-specific zoom-in of the Evolutionary-Based branch of the Eight Universal Methodology Paradigms (UM8) (DOI: 10.5281/zenodo.18790203), Evolutionary Compass develops evolutionary methodology into a decision-oriented framework for researchers, system designers, and practitioners. Its purpose is to make the architecture of evolutionary routes explicit. The framework offers a structured way to diagnose the evolutionary design challenge presented by a system, determine how change should be organized across time, and reason about the longer-horizon consequences of present design choices. The value of the framework It supports earlier and better design decisions. Evolutionary Compass makes the main commitments of an evolutionary route explicit before implementation locks them into system architecture: the strategic role of the route, the object of change, the regime of update, the mechanism through which change enters, the signal that governs selection, and the logic that shapes what is carried forward. This gives planning and system design a more disciplined starting point and makes it easier to choose an appropriate route before effort is absorbed by the wrong form of change. It makes evolutionary routes easier to diagnose and redesign. Many projects continue to update without producing robust progress because the difficulty lies in the route itself: the wrong role has been chosen, change is directed at the wrong target, the update regime is misaligned with the problem, or the mechanism and signal generate weak or distorted selection pressure. Evolutionary Compass makes those mismatches easier to detect, which supports more precise diagnosis and more disciplined redesign when a route stalls, fragments, or narrows future possibility. It provides a stable cross-domain framework for cumulative methodological knowledge. Evolutionary advances often appear in local terminology, which obscures their broader relevance and keeps them confined within disciplinary boundaries. Evolutionary Compass brings those advances into a shared frame, so theoretical developments, new methods, and higher-level design strategies from different disciplines can accumulate around stable evolutionary logic. Progress made in one field can then be transferred, adapted under new constraints, and recombined with advances from others, turning isolated disciplinary gains into shared methodological resources for cross-domain innovation. Architecture of the framework Evolutionary Role identifies the primary strategic function served by an evolutionary route. Update Architecture specifies how change is organized in operational terms. This major module contains four linked submodules: Update Target, Update Regime, Update Mechanism, and Feedback Signal. Carryover Logic specifies what is carried forward after update, including what is retained, what remains available, what is refreshed, and what moves into later rounds or broader contexts. This architecture separates three levels of reasoning that are often collapsed in practice. Evolutionary Role frames the purpose of the route. Update Architecture determines how the route proceeds. Carryover Logic shapes the history that the route builds. The framework therefore serves as a scaffold for route selection, comparative analysis, and long-horizon system design. 1. Evolutionary Role Evolutionary Role identifies the primary strategic function served by an evolutionary route. This module anchors the framework at the strategic level, where the central design question concerns the purpose of organized change across time. Refine — Quality Gain. Refine is most useful when the primary aim is to strengthen, mature, or improve an existing line of development through iterative enhancement. Its core logic is directional continuity and progressive strengthening along the same developmental line. A refining route may rely on direct adjustment, structural revision, candidate comparison, or selective reuse, so long as these changes serve the maturation of the same route. This role is especially useful in settings where improvement, completeness, reliability, and performance remain the main priorities across repeated cycles of development. Adapt — Context Fit. Adapt becomes central when the primary aim is to maintain or restore fit under changed tasks, environments, operating conditions, or drift. Its key feature is contextual re-alignment under changing conditions. This role is especially important in deployed systems, institutional settings, and long-lived processes whose conditions of success continue to shift over time. The main strategic concern lies in sustaining effectiveness as the surrounding context changes, while preserving continuity for the route to remain workable, legible, and governable. Explore — Space Discovery. Explore is most relevant when the primary aim is to open new regions of possibility and discover alternatives that do not emerge from continued strengthening of the current route. This role is especially valuable in underexplored spaces, early-stage methodological development, and settings where the most important opportunities lie beyond the neighborhood of the current solution. Its strategic orientation favors discovery, novelty, and access to alternative directions whose value may only become visible once the search moves beyond familiar routes. 2. Update Architecture Update Architecture forms the operational core of Evolutionary Compass. It organizes change through four linked submodules: Update Target, Update Regime, Update Mechanism, and Feedback Signal. Their relationship is central to the framework’s value. A route may target one part of the system while operating under a completely different temporal pattern; it may rely on a particular update mechanism while drawing legitimacy from a signal family that comes from elsewhere. This decomposition allows evolutionary systems to be diagnosed with much greater precision and redesigned without conflating where change happens, how it happens, when it happens, and how it is judged. 2.1 Update Target Update Target identifies the locus of leverage. It specifies what part of the system is actually under revision and therefore where evolutionary effort is being invested. Parameters / Traits — Adjustable Values. This target is appropriate when progress depends on tuning magnitudes: coefficients, thresholds, schedules, gains, sensitivities, or other directly adjustable values. It provides a low-friction route to improvement and often supports reversible, tightly controlled revision. Its value is greatest when the broader architecture remains trustworthy and the main design question concerns how far current settings can be pushed before structural limitations begin to dominate. Structures — System Form. This target becomes important when the main source of leverage lies in topology, modular composition, internal arrangement, or higher-level architecture. Structural targets matter when step-change improvements depend on reorganization rather than finer adjustment of existing settings. They open access to new functional spaces and new combinations of capability while introducing heavier integration costs and higher implementation risk. Structural intervention becomes especially compelling once continuous tuning yields diminishing returns and the real leverage has shifted to system form. Strategies / Policies — Action Logic. This target is appropriate when the evolving object is the logic by which the system chooses, routes, controls, or interacts. It matters most in systems whose capability is defined by behavior under uncertainty rather than by static representation alone. Agentic systems, controllers, and decision-heavy processes often live here. The most valuable evolutionary move then concerns how the system acts, not merely how it is parameterized or assembled. Knowledge / Representations — Internal Content. This target becomes central when progress depends on the refinement of internal schemas, memories, rules, prototypes, latent features, or representational organization. It is especially valuable in systems that build internal models which later govern inference, action, or future learning. This option is most appropriate when the main leverage lies in changes to the system's internal content. 2.2 Update Regime Update Regime specifies the temporal organization of change. It clarifies how update unfolds across time and therefore shapes evaluation rhythm, rollback possibilities, responsiveness, and the system’s exposure to drift and interference. Episodic — Isolated Episodes. Episodic update suits bounded interventions and clearly separated rounds of revision. It offers strong support for validation staging, rollback control, auditing, and discrete deployment decisions. This regime is especially useful in systems where safety, governance, or evaluation clarity matter enough to justify slower, more deliberate revision. Its strength lies in containment and in the legibility of each update round as a distinct event. Cyclical — Recurring Cycles. Cyclical update suits recurring loops, periodic batches, and staged rounds of revision. It supports systems that benefit from rhythmic accumulation of evidence and repeated comparison across iterations. This regime offers a practical ba