{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"invariant_mode_decomposition_design__architecture_urban_planning","trajectory_id":"R","attempt_index":0,"archetype_slug":"invariant_mode_decomposition_design","domain_slug":"architecture_urban_planning","decision":"CANDIDATE","problem_id":"network_congestion_mode_blindness","causal_lever_id":"target_amplifying_traffic_modes","proposal":{"problem":"Urban traffic management often treats recurring congestion as separate intersection or corridor failures, although coupled changes in queues, inflows, signal states, and route demand may form network-wide patterns that persist or amplify. This can make local retiming move congestion elsewhere or miss an emerging unstable pattern.","actors_substrate":["travelers, including transit riders, pedestrians, cyclists, and drivers","traffic-management and transit agencies","residents and businesses near affected streets","signalized street network, sensors, signal plans, and route flows"],"observable_state":"Time-indexed vectors of link queues or speeds, turning flows, signal phases, transit delay, pedestrian delay, spillback, and safety-proxy measures across a bounded network.","consequence":"Recurring delay, unreliable travel, spillback, and locally displaced burdens remain poorly controlled despite repeated site-level adjustments.","affected_objective":"Reduce network delay and spillback without worsening safety, access, transit reliability, or neighborhood burden.","structural_mapping":[{"archetype_element":"coupled transformation","domain_realization":"A locally estimated transition operator maps one traffic-state interval to the next under fixed operating conditions.","claim_kind":"HYPOTHESIS"},{"archetype_element":"surface coordinates","domain_realization":"Individual link and intersection metrics used by ordinary operations.","claim_kind":"CORPUS"},{"archetype_element":"invariant directions","domain_realization":"Combinations of queues, flows, and signal states that approximately persist, decay, or grow together.","claim_kind":"HYPOTHESIS"},{"archetype_element":"modal gain","domain_realization":"Estimated interval-to-interval persistence or amplification of each coupled traffic pattern.","claim_kind":"INFERENCE"},{"archetype_element":"modal intervention","domain_realization":"Coordinated signal or transit-priority changes selected for leverage on a harmful mode rather than one site.","claim_kind":"HYPOTHESIS"},{"archetype_element":"residual and drift checks","domain_realization":"Held-out prediction error, structured residuals, basis rotation, and spectral-gap change delimit continued use.","claim_kind":"INFERENCE"}],"component_map":[{"component":"Transformation Scope","status":"adapted","domain_realization":"One bounded district, fixed time step, and comparable demand/incident regime."},{"component":"State-Vector Definition","status":"direct","domain_realization":"Queues, flows, phases, transit and pedestrian delays, spillback, and safety proxies."},{"component":"Invariant Mode Basis","status":"adapted","domain_realization":"Modes of the locally estimated traffic transition operator."},{"component":"Modal Gain Spectrum","status":"direct","domain_realization":"Estimated gain or persistence for every retained mode."},{"component":"Dominant Mode Selection Rule","status":"direct","domain_realization":"Select modes exceeding gain, outcome-sensitivity, and consequence thresholds."},{"component":"Stable/Unstable Mode Partition","status":"adapted","domain_realization":"Classify modes as decaying, marginal, growing, or oscillatory within the fitted regime."},{"component":"Modal Intervention Map","status":"adapted","domain_realization":"Map feasible signal and priority controls to their estimated modal effects."},{"component":"Reconstruction Residual Check","status":"direct","domain_realization":"Held-out state reconstruction error plus inspection for structured omissions."},{"component":"Mode Drift Monitor","status":"direct","domain_realization":"Track mode rotation, reordering, gains, and residual growth."},{"component":"Interpretation Scope Contract","status":"direct","domain_realization":"Modes are local predictive patterns, not automatically causal or valid across regimes."},{"component":"Mode-Coupling Register","status":"direct","domain_realization":"Record cross-mode effects, near-degeneracy, and non-normality."},{"component":"Local Linearization Window","status":"adapted","domain_realization":"Specified demand, weather, event, incident, and control-plan envelope."},{"component":"Spectral Gap Threshold","status":"direct","domain_realization":"Minimum separation required to treat retained modes as identifiable and dominant."}],"mechanism_dispositions":[{"slug":"eigendecomposition_workflow","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Decompose an explicit locally estimated transition matrix; flag ill-conditioning.","counterfactual_removal":"No invariant directions or gains would anchor the intervention logic."},{"slug":"modal_sensitivity_sweep","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Perturb feasible controls in simulation and log outcome and cross-mode responses.","counterfactual_removal":"Dominant modes could not be translated into actionable leverage."},{"slug":"modal_stability_analysis","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Classify growth only inside the declared operating window.","counterfactual_removal":"The proposal could not distinguish harmful amplification from harmless prominence."},{"slug":"mode_shape_testing","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Deliberate excitation of live traffic is unsafe and ambient identification duplicates model validation.","counterfactual_removal":"No material change; held-out observations provide validation."},{"slug":"network_spectral_centrality_analysis","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Static node importance does not identify dynamic congestion modes.","counterfactual_removal":"No change to the causal chain."},{"slug":"power_iteration_probe","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"A dominant-only estimate could miss consequential secondary modes and obscure small gaps.","counterfactual_removal":"No change because the bounded operator permits fuller decomposition."},{"slug":"principal_component_analysis","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Variance directions need not represent traffic-state propagation.","counterfactual_removal":"No change to transformation-based inference."},{"slug":"reduced_order_model","disposition":"selected_supporting","contribution_type":"TEST_DESIGN","adaptation_or_rejection":"Use a bounded surrogate for offline control comparisons, never unreviewed live control.","counterfactual_removal":"Testing becomes slower but the causal theory remains intact."},{"slug":"residual_reconstruction_test","disposition":"selected_load_bearing","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Require held-out error and residual-structure budgets.","counterfactual_removal":"Omitted traffic behavior could be hidden while controls are approved."},{"slug":"singular_value_decomposition","disposition":"selected_supporting","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Use to diagnose non-normal transient amplification and conditioning, not as invariant dynamics.","counterfactual_removal":"Transient-risk and conditioning checks would weaken."},{"slug":"spectral_decomposition_report","disposition":"selected_supporting","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Document meanings, non-causal limits, couplings, and valid regimes.","counterfactual_removal":"Decision makers could overread tentative modes as causal facts."},{"slug":"spectral_gap_monitor","disposition":"selected_load_bearing","contribution_type":"OPERATIONAL","adaptation_or_rejection":"Track retained/dropped separation, basis drift, and identity swaps.","counterfactual_removal":"The model could remain deployed after its simplification loses validity."}],"causal_chain":["HYPOTHESIS: repeated traffic states contain locally stable coupled propagation patterns.","Estimating the transition operator exposes patterns with high persistence, amplification, or transient gain.","Sensitivity analysis identifies feasible controls that reduce harmful modal response while checking cross-effects.","A residual-bounded reduced model compares those controls with ordinary and rival approaches.","If validated, coordinated controls should reduce network harm more reliably than site-level reactions.","Drift, gap, safety, and residual thresholds halt use when the local approximation fails."],"baseline":"Operators monitor site and corridor KPIs, then react with isolated retiming, incident response, or complaint-driven adjustments.","nearest_rival":"Full-state network model-predictive control that optimizes forecast outcomes without an explicit modal decomposition.","authority_safety":{"affected_parties":["travelers across all modes","nearby residents and businesses","people with disabilities or limited route alternatives","traffic operators and emergency services"],"decision_authority":"The public road authority retains approval; transit agencies and accessibility and safety reviewers must approve impacts within their mandates.","authorized_first_step":"Run a retrospective and shadow-mode comparison on one district and several recurring peak periods; make no live control changes.","excluded_actions":["autonomous live signal changes","reducing pedestrian clearance or emergency access","rerouting burdens into vulnerable neighborhoods without review","using modal scores as proof of causation"],"halt_rollback":"Halt if held-out error, structured residuals, spectral drift, gap loss, or any safety/equity threshold is breached; discard recommendations and revert to the existing plan."}},"negative_tests":{"strongest_counterevidence":"Observed congestion may be driven mainly by incidents, demand shocks, capacity constraints, or nonlinear route switching, leaving no reusable modal window; advanced adaptive control may already capture useful coupling without decomposition.","analogy_break":"Traffic participants respond strategically to controls, so the transformation can change when acted upon; eigenmodes of a fitted local operator are not enduring physical modes.","failure_condition":"Modes are poorly conditioned, regime-specific, non-reproducible, or require controls that violate safety, access, or equity constraints.","problem_falsifier":"After conditioning on incidents, demand, weather, and events, held-out traffic states show no repeatable coupled lag structure beyond a single observable bottleneck or ordinary corridor model.","intervention_falsifier":"Reproducible modes predict traffic, but modal targeting fails to outperform the nearest rival on held-out delay, spillback, transit reliability, safety, and distributional outcomes.","risks":["false confidence from local linearization","mode identity swaps near a small spectral gap","displacement of congestion or exposure across neighborhoods","sensor bias or missing pedestrian and transit states","optimization of delay at the expense of safety or accessibility"]},"null_rationale":null,"classification":{"candidate_kind":"TESTABLE_CONJECTURE","prior_art_status":"UNSEARCHED","evidence_maturity":"HYPOTHESIS"},"revision_change_log":{"revision_kind":"ORIGINAL","prior_problem_id":null,"prior_causal_lever_id":null,"problem_changed":false,"causal_lever_changed":false,"conceptual_changes":[],"operational_changes":[],"repairs_addressed":[]},"confidence":0.78,"generator_notes":"The fit is structural but conditional on a reproducible local transition model. No claim of novelty or established effectiveness is made."}