{"cases":[{"alternative_route_blocks":[{"blocked_by_condition_id":"b025_a01_c3","condition_set_id":"B1","how_blocked":"The skippers’ terms and model outputs remain separately recorded; no coding or translation compresses the skippers’ distinctions."},{"blocked_by_condition_id":"b025_a01_c5","condition_set_id":"B3","how_blocked":"The disagreement concerns how water states are partitioned, not competing meanings assigned to the same fishing behavior."},{"blocked_by_condition_id":"b025_a01_c6","condition_set_id":"B4","how_blocked":"Neither account has been designated official, and no account has displaced the other through institutional authority."}],"case_id":"E14A032__DIRECT","case_type":"DIRECT_POSITIVE","domain":"Marine field research and coastal hydrodynamics","intended_route_id":"B2","omitted_condition_id":null,"remedy_leakage_audit":"The statement reports the representational conflict and source-based rejection without proposing any process, framework, or intervention.","route_evidence":[{"condition_id":"b025_a01_c1","intended_status":"SATISFIED","scenario_evidence":"Skippers’ water-state distinctions cross-cut velocity and turbidity bands in both directions, preventing a one-to-one correspondence."},{"condition_id":"b025_a01_c4","intended_status":"SATISFIED","scenario_evidence":"The modeling subgroup excludes the skippers’ interpretations solely because the skippers are embedded practitioners."}],"scenario_text":"Marine social scientists are studying why small-boat crews cancel trips on days that harbor instruments classify as safe. Veteran skippers describe the water as “working,” “hollow,” “holding,” or “about to turn,” judgments learned through years of handling particular hulls and grounds. Hydrodynamic researchers instead divide the same days by current velocity, wave period, turbidity, and forecast uncertainty. A day called “holding” can fall into several instrument bands, while one velocity band can include both “working” and “hollow” water. The field archive retains the skippers’ original log entries alongside the measurements, so their distinctions have not been compressed into instrument codes. The dispute concerns how water conditions are sorted, not different purposes attributed to any crew action. Nevertheless, the modeling subgroup excludes the log interpretations from its causal memo solely because the skippers are embedded in the practice and therefore deemed incapable of objective observation. No account has been declared the agency’s official explanation, and the project remains at the draft-analysis stage.","scenario_title":"Safe Forecasts and Cancelled Sailings","vocabulary_separation_audit":"The scenario uses concrete nautical and hydrodynamic language rather than generic diagnostic terminology."},{"alternative_route_blocks":[{"blocked_by_condition_id":"b025_a01_c3","condition_set_id":"B1","how_blocked":"Operator-state records and forensic classifications remain intact and separate; neither is translated into a reduced common code."},{"blocked_by_condition_id":"b025_a01_c5","condition_set_id":"B3","how_blocked":"Operators and investigators agree on the purpose and consequences of every observed action; only their grouping schemes differ."},{"blocked_by_condition_id":"b025_a01_c6","condition_set_id":"B4","how_blocked":"The review has not adopted an official account, and no institutional decision has displaced either record."}],"case_id":"E14A032__TRANSFER","case_type":"TRANSFER_POSITIVE","domain":"Semiconductor fabrication incident analysis","intended_route_id":"B2","omitted_condition_id":null,"remedy_leakage_audit":"The statement contains no recommendation, corrective action, or description of a proposed interpretive method.","route_evidence":[{"condition_id":"b025_a01_c1","intended_status":"SATISFIED","scenario_evidence":"Fab operators’ restart states and forensic investigators’ incident classes cross-cut one another rather than corresponding individually."},{"condition_id":"b025_a01_c4","intended_status":"SATISFIED","scenario_evidence":"Plant managers reject the forensic conclusions exclusively because the investigators are external consultants."}],"scenario_text":"A semiconductor fab is investigating intermittent wafer contamination after brief utility interruptions. Console operators distinguish “warm carryover,” “soft recovery,” “tool wake-up,” and “lot rescue,” based on chamber history, crew coordination, and whether a run can resume without requalification. External forensic engineers classify incidents by voltage sag, interlock response, contamination pathway, and estimated yield exposure. Several operator states span multiple forensic classes, and the engineers’ “minor control deviation” class contains episodes that crews regard as operationally distinct. Both datasets remain in their original form; no common code has replaced either vocabulary. The two groups also agree about what each button press and lot movement was intended to accomplish, so the disagreement is about grouping the episodes rather than the meaning of the actions. Plant managers nevertheless discard the forensic team’s conclusions solely because the engineers are consultants who have never worked a production shift. They identify no defect in the sampling, measurements, or calculations. The inquiry is still preliminary, and neither record has been made the company’s official account.","scenario_title":"Contamination After Utility Interruptions","vocabulary_separation_audit":"The transfer case replaces fieldwork language with fab operations, interlocks, chamber history, yield exposure, and forensic engineering terminology."},{"alternative_route_blocks":[{"blocked_by_condition_id":"b025_a01_c1","condition_set_id":"B1","how_blocked":"The independently verified crosswalk establishes exact one-to-one correspondence, so the route’s shared mismatch requirement is contradicted; the scenario also states that translation loses no distinctions."},{"blocked_by_condition_id":"b025_a01_c1","condition_set_id":"B3","how_blocked":"The exact correspondence contradicts the shared mismatch requirement; technicians and engineers also assign the same significance to each event."},{"blocked_by_condition_id":"b025_a01_c1","condition_set_id":"B4","how_blocked":"The exact correspondence contradicts the shared mismatch requirement, and neither account has become official or displaced the other."}],"case_id":"E14A032__NEAR_MISS","case_type":"ONE_LITERAL_NEAR_MISS","domain":"Semiconductor fabrication quality investigation","intended_route_id":"B2","omitted_condition_id":"b025_a01_c1","remedy_leakage_audit":"The statement describes the investigation and rejection without recommending a response or naming any solution pattern.","route_evidence":[{"condition_id":"b025_a01_c1","intended_status":"CONTRADICTED","scenario_evidence":"A blind audit verifies that every technician state has exactly one engineering code and every engineering code maps back to exactly one technician state."},{"condition_id":"b025_a01_c4","intended_status":"SATISFIED","scenario_evidence":"Operations leadership rejects the engineering report solely because its authors have not worked on the fab floor."}],"scenario_text":"A semiconductor fab is reviewing recurring alignment faults after robotic wafer transfers. Shift technicians use six named machine states based on sensor sequence, arm position, and recovery history. Independent equipment engineers also use six fault codes derived from telemetry and mechanical inspection. A blind audit of forty incidents finds an exact pairing: every technician state corresponds to one engineering code, every engineering code maps back to one technician state, and no episode breaks the pairing. The descriptions differ in wording but retain the same distinctions, and both groups assign the same operational significance to each transfer and recovery action. Raw notes, telemetry, and the verified crosswalk show that translation between the two sets loses no information. Even so, operations leadership rejects the engineers’ report solely because its authors have never worked on the fab floor. Leaders cite no error in the inspection procedure, evidence, or conclusions. The review committee has not selected an official explanation, and neither record has replaced the other in maintenance policy or incident reporting.","scenario_title":"Two Matching Accounts of Transfer Faults","vocabulary_separation_audit":"The near miss retains the transfer case’s technical manufacturing register and investigative complexity while replacing cross-cutting classifications with an explicitly verified one-to-one mapping."}],"experiment_id":"eoa_inverse_innovation_exp14_applicability_retrieval40_20260813","sample_id":"E14A032","schema_version":1}