{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp04_retrieval_first_paired20_20260802","cell_id":"layer_decay_and_expiration_management__aviation_aeronautics","arm":"RETRIEVAL_FIRST","round_index":0,"hypotheses":[{"hypothesis_id":"H1","title":"Expiring EFB revision layers","problem":"Superseded charts, procedures, and notices can remain searchable on electronic flight bags after newer revisions arrive.","affected_stakeholder":"Airline pilots and aeronautical-data administrators","workflow_boundary":"From ground publication and device synchronization through preflight document selection","failure_mode":"A pilot selects a locally cached superseded revision that appears current.","unit_of_analysis":"One document revision on one EFB device","causal_lever":"Give each revision an effective interval, hide it at expiry, and leave a marker resolving searches to its successor.","archetype_mapping":"Treat successive document revisions as layers whose active authority expires while historical copies remain auditable.","expected_value":"Fewer obsolete-document selections without destroying revision history.","falsifiable_claim":"In a simulated preflight task, expiry plus successor markers reduces superseded-revision selections by at least 30% versus revision dates alone.","diversity_rationale":"Targets distributed publication synchronization and user selection, not storage cost, restoration, or maintenance compliance.","mechanism_slugs":["time_to_live_ttl_policy","tombstone_or_deletion_marker","stale_layer_detection_dashboard"],"search_questions":["Do EFB platforms already enforce effective intervals and successor resolution across offline devices?","Are superseded aeronautical documents implicated in reported operational errors?","What controlled evidence compares expiry controls with revision labels alone?"]},{"hypothesis_id":"H2","title":"Dependency-gated simulation archive","problem":"Aerodynamic simulation runs accumulate costly meshes, solver states, and derived outputs, yet indiscriminate cleanup can break certification reconstruction.","affected_stakeholder":"Aerodynamics and certification engineers","workflow_boundary":"From design-iteration closure through later certification or anomaly reconstruction","failure_mode":"Teams either retain every artifact hot or delete an input still required to reproduce a cited result.","unit_of_analysis":"One provenance-linked simulation artifact bundle","causal_lever":"Rank aging bundles, trace inbound dependencies before demotion, quarantine deletions, and exercise sampled restores.","archetype_mapping":"Manage simulation generations as reconstructable layers with differentiated hot, archived, and expired states.","expected_value":"Lower active-storage burden while preserving reproducibility of consequential analyses.","falsifiable_claim":"A pilot can reduce hot storage by at least 40% while at least 95% of sampled cited runs remain reproducible within a specified restore time.","diversity_rationale":"Centers on computational provenance, dependency graphs, and restoration rather than stale operational guidance.","mechanism_slugs":["age_weighted_value_score","dependency_safe_delete_check","lifecycle_storage_tiering_policy","archive_restore_test","soft_delete_quarantine_window"],"search_questions":["What simulation-artifact retention practices are required for aircraft certification evidence?","Which artifact classes are actually necessary to reproduce historical aerodynamic results?","Have dependency-aware archival systems been evaluated in aerospace engineering workflows?"]},{"hypothesis_id":"H3","title":"Segment-level flight-data decay","problem":"Post-flight monitoring retains large volumes of redundant data while fixed flight-level deletion may erase low-frequency precursors needed for later safety analysis.","affected_stakeholder":"Airline flight-safety analysts and data custodians","workflow_boundary":"From post-flight ingestion through event detection, investigation hold, and archival disposition","failure_mode":"Uniform retention preserves low-value cruise data but removes segments later needed to reconstruct an emerging hazard.","unit_of_analysis":"One flight-phase and sensor-channel segment","causal_lever":"Score segments by anomaly evidence, reconstruction role, age, and access; hold linked segments and tier the remainder.","archetype_mapping":"Decompose each flight record into layers with distinct decay rates, preservation exceptions, and restore requirements.","expected_value":"Cheaper retention with better protection for investigation-relevant evidence.","falsifiable_claim":"Compared with uniform flight-level retention, segment-level lifecycle rules cut stored bytes by at least 25% without reducing detection of a predefined event set and retain at least 95% sampled restore success.","diversity_rationale":"Uses fine-grained telemetry segments, statistical safety value, and investigation holds rather than whole documents or engineering runs.","mechanism_slugs":["age_weighted_value_score","retention_schedule","lifecycle_storage_tiering_policy","archive_restore_test"],"search_questions":["Do flight-data monitoring systems retain and expire data by flight, channel, phase, or detected event?","Which weak precursor analyses require raw historical data rather than derived features?","What legal, labor, privacy, and investigation rules constrain adaptive retention?"]},{"hypothesis_id":"H4","title":"Supersession-aware clearance stack","problem":"Successive trajectory and clearance amendments can clutter controller handoff views, leaving canceled constraints visually competitive with current instructions.","affected_stakeholder":"Air traffic controllers receiving a sector handoff","workflow_boundary":"From clearance amendment issuance through sector handoff and acknowledgment","failure_mode":"A superseded amendment is mistaken for the active clearance or delays reconstruction of the current state.","unit_of_analysis":"One clearance amendment in one aircraft handoff sequence","causal_lever":"Expire superseded amendments from the active view while retaining compact tombstones that preserve sequence and successor identity.","archetype_mapping":"Treat amendments as rapidly deposited authority layers that must decay from active salience without losing chronology.","expected_value":"Faster, more accurate identification of the controlling clearance during handoff.","falsifiable_claim":"In high-workload simulations, supersession-aware expiry reduces wrong-current-clearance answers or retrieval time by at least 20% without reducing sequence-reconstruction accuracy.","diversity_rationale":"Addresses real-time human-factors salience in a seconds-scale sequential workflow, unlike archival or compliance hypotheses.","mechanism_slugs":["time_to_live_ttl_policy","tombstone_or_deletion_marker","log_rotation_and_cleanup_job"],"search_questions":["How do current controller systems display superseded clearance amendments during handoff?","Are amendment-stack errors represented in incident or human-factors literature?","What display experiments compare full history with active-state-plus-tombstone views?"]},{"hypothesis_id":"H5","title":"Leased temporary-repair authority","problem":"Temporary structural repairs and engineering concessions can persist beyond intended inspection or replacement dates when their continuing authority is implicit.","affected_stakeholder":"Maintenance controllers, structures engineers, and continuing-airworthiness managers","workflow_boundary":"From temporary-repair approval through recurring inspection, renewal, permanent repair, and record closure","failure_mode":"An expired concession remains operationally active, or closure removes evidence needed to trace affected parts and inspections.","unit_of_analysis":"One temporary-repair authorization tied to an aircraft location or serialized part","causal_lever":"Stamp approval with an expiring lease, require evidence-based renewal, block closure while dependencies remain, and quarantine superseded records.","archetype_mapping":"Manage temporary repair approvals as safety-critical layers whose authority expires but whose evidence remains retained.","expected_value":"Fewer silent overruns with preserved maintenance traceability.","falsifiable_claim":"Over 12 months, explicit leases and dependency-gated closure reduce overdue temporary repairs by at least 25% with no decrease in successful audit reconstruction.","diversity_rationale":"Focuses on physical-aircraft configuration, recurring maintenance obligations, and renewal behavior rather than information retrieval or storage optimization.","mechanism_slugs":["time_to_live_ttl_policy","dependency_safe_delete_check","soft_delete_quarantine_window","retention_schedule"],"search_questions":["How are temporary-repair limits and renewals currently represented in continuing-airworthiness systems?","How often do repair authorizations overrun due to record-state or ownership failures?","Do existing systems couple expiry, affected-part dependencies, and immutable closure evidence?"]}]}