{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp11_mechanism_context_external20_20260804","research_id":"eoa_inverse_innovation_exp11_external_scrutiny_20260804","cell_id":"deadweight_loss_reduction__systems_cybernetics","opaque_id":"deadweight_loss_reduction__systems_cybernetics__C","search_lanes":{"direct_problem":{"queries":["industrial process control PID tuning management of change approval parameter changes delay","control loop retuning change management approval workflow plant PID gains","poorly tuned control loops energy consumption actuator wear operator workload industrial study","\"PID tuning\" \"management of change\" process control"],"source_ids":["S1","S2","S3","S4"],"no_result_note":null},"closest_prior_art":{"queries":["adaptive control approved operating regions tuning limits process industry","site:emerson.com DeltaV adaptive control automatically adjust PID tuning constraints limits","streamlined management of change low risk changes process safety industry","risk based management of change tiered approval process chemical plant MOC minor changes"],"source_ids":["S3","S4","S5","S6","S8"],"no_result_note":null},"historical_terminology":{"queries":["adaptive control strategies process control self-tuning regulator operating problems","gain scheduling PID tuning process industry operating regions","online self-tuning PID controller industrial process historical","PID self-tuner save undo controller parameters"],"source_ids":["S6","S7","S8"],"no_result_note":null},"products_practices_standards":{"queries":["site:osha.gov process safety management management of change operating parameters minor well understood change","IEC 61511 management of change controller parameters safety instrumented systems official","site:siemens.com PID autotuning industrial controller automatic tuning parameter limits","site:emerson.com DeltaV Adapt continuous adaptive tuning approved operating regions"],"source_ids":["S2","S3","S4","S5","S6","S8"],"no_result_note":null},"non_english_regional":{"queries":["réglage PID gestion des modifications MOC paramètres prédéfinis industrie","PID-Regler Selbstoptimierung Parametergrenzen Freigabe Prozessleitsystem","自動 PID チューニング パラメータ 制限 産業 制御 変更管理","ajuste PID automático límites parámetros control industrial gestión del cambio"],"source_ids":["S8"],"no_result_note":null},"composition_subproblems":{"queries":["safe automatic PID tuning constraints rollback industrial process control paper","bounded adaptive PID tuning stability constraints automatic rollback paper","\"pre-defined parameters\" \"control system changes\" management of change","control loop performance monitoring retuning workflow plant case study MOC"],"source_ids":["S1","S2","S3","S4","S5","S6","S8"],"no_result_note":null}},"sources":[{"source_id":"S1","title":"The current state of control loop performance monitoring – A survey of application in industry","url":"https://www.sciencedirect.com/science/article/pii/S0959152415002127","publisher":"Journal of Process Control / Elsevier","date_or_year":"2016","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Industrial plants contain hundreds or thousands of PID loops, and surveyed industrial users reported more poor performers than expected.","Poor loop performance can arise from tuning and other faults, while organizational issues, data availability, and access can impede performance-monitoring practice.","The source supports the general stale or poor-loop problem but does not isolate approval latency as its cause."]},{"source_id":"S2","title":"1910.119 App C — Compliance Guidelines and Recommendations for Process Safety Management","url":"https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.119AppC","publisher":"U.S. Occupational Safety and Health Administration","date_or_year":"1992, current web edition","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Changes other than replacement in kind require management, documentation, review, and authorization in covered processes.","For a minor, well-understood change, a checklist reviewed by an authorized person may suffice; complex or significant changes may require multi-department hazard evaluation.","Temporary changes should have time limits and mechanisms for returning equipment and procedures to their original or designed condition.","An employer and its authorized personnel are identifiable authorizers."]},{"source_id":"S3","title":"CIC Management of Change","url":"https://minerva.jrc.ec.europa.eu/en/shorturl/technical_working_group_2_seveso_inspections/cic_management_of_change","publisher":"European Commission Joint Research Centre, MINERVA","date_or_year":"2023 web edition","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["The guidance defines management of change as a review and authorization process intended to prevent new hazards.","It explicitly excepts instrument or control-system changes within predefined parameters from its definition of change while retaining MOC for risk-relevant changes.","It permits preliminary risk screening to scale assessment effort and requires competent roles, authorization, documentation, and pre-startup verification.","This closely anticipates preauthorization inside a validated envelope with escalation outside it."]},{"source_id":"S4","title":"Guidance on meeting expectations of EI Process Safety Management Framework Element 12: Management of Change and Project Management","url":"https://www.energyinst.org/industry/publications/topics/process-safety/guidance-on-meeting-expectations-of-ei-process-safety-management-framework-element-12-management-of-change-and-project-management","publisher":"Energy Institute","date_or_year":"2015","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["The guidance requires changes, including software and control-system changes, to be classified as high or low risk using approved criteria.","Low-risk changes may be reviewed by an appropriate competent line manager with delegated authority, whereas high-risk changes receive independent multidisciplinary or safe-operations-committee review.","It requires defined control measures, documentation, monitoring, delegated authority, and time limitation for temporary changes.","This is established risk-tiered change authorization rather than a uniform approval path."]},{"source_id":"S5","title":"DeltaV Adapt — Continuous Closed Loop Adaptive Control","url":"https://www.emerson.com/is/content/emerson/en/systems-and-software/deltav-distributed-control-system-dcs/product-data-sheets/documents/deltav-adapt-continuous-closed-loop-adaptive-control.pdf","publisher":"Emerson","date_or_year":"2024","source_type":"FIRST_PARTY_PRODUCT","language":"English","claims_supported":["A commercial DCS product continuously adjusts PID tuning as process conditions change and retains tuning by operating region.","Engineers can define region boundaries, approve models for each region, and constrain model-identification and tuning parameters.","Validated-model quality metrics precede automatic tuning.","The product supplies full-adaptive, partial-adaptive using approved regional models, and learn-only advisory modes, closely corresponding to bounded automation and shadow mode."]},{"source_id":"S6","title":"SIMATIC PID Temperature Control","url":"https://cache.industry.siemens.com/dl/files/438/18652438/att_26584/v1/PID_Temp_e.PDF","publisher":"Siemens","date_or_year":"2003","source_type":"FIRST_PARTY_PRODUCT","language":"English","claims_supported":["The industrial controller automatically sets PI/PID parameters through self-tuning.","Usable parameters can be saved before tuning, and the prior saved parameters can be restored with UNDO_PAR.","The manual documents tuning status, abort conditions, manual mode, monitoring, and remedies for oscillation or excessive excitation.","This directly establishes technical feasibility of reversible bounded retuning, although it does not establish the proposed governance workflow."]},{"source_id":"S7","title":"Adaptive control strategies for process control: A survey","url":"https://aiche.onlinelibrary.wiley.com/doi/abs/10.1002/aic.690320602","publisher":"AIChE Journal / Wiley","date_or_year":"1986","source_type":"SECONDARY_RESEARCH","language":"English","claims_supported":["Automatic adjustment of controller settings in response to process or environmental change was established adaptive-control terminology by 1986.","The survey covers experimental process-control applications and operating problems of adaptive schemes.","The causal lever of automatic retuning is therefore decades-old prior art rather than a new control concept."]},{"source_id":"S8","title":"Equipment Modules for Chemical Plants — Application Description, Section 6.3 PID Controller Configuration","url":"https://cache.industry.siemens.com/dl/files/373/53843373/att_10111/v1/53843373_equipmentmodules_de.pdf","publisher":"Siemens","date_or_year":"2014","source_type":"FIRST_PARTY_PRODUCT","language":"German","claims_supported":["The German application guide provides a PID tuner and references gain scheduling with operating-point-dependent parameters.","It instructs users to choose excitation small enough to protect the process and product quality and to monitor the process continuously during optimization.","Planned tuning trials are to be coordinated with responsible operating personnel, and an operator-screen authorization grants the engineering system permission to intervene.","This regional source shows an implemented authorization boundary and bounded tuning procedure."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"Industrial evidence supports widespread poor-performing or mismatched loops and organizational impediments to loop-performance work. Official guidance also recognizes that minor, well-understood, or predefined control-system changes warrant lighter treatment than significant changes. However, no retained source directly documents a plant where a uniform slow approval queue is the measured cause of stale tuning, nor quantifies the proposed effects of that queue on energy, cycling, alarms, or operator intervention.","source_ids":["S1","S2","S3","S4"],"uncertainty":"The approval-latency mechanism remains a site-level hypothesis requiring configuration-change records, timestamps, and historian data. Poor performance may instead reflect valves, sensors, disturbances, interaction among loops, or substantive hazard analysis."},"adopter_evidence":{"status":"SUPPORTED","finding":"An identifiable authorization chain exists: the employer or plant control-system owner, an authorized competent controls or operations line manager for bounded low-risk changes, and multidisciplinary process-safety authority for higher-risk or out-of-envelope changes.","source_ids":["S2","S3","S4","S8"],"uncertainty":"The sources establish role classes, not a named plant or named individual. Local PSM, functional-safety, cybersecurity, labor, and insurance rules could reassign or add approvers."},"implementation_evidence":{"status":"SUPPORTED","finding":"Commercial control systems already implement adaptive or self-tuning PID operation, approved operating regions, constrained tuning parameters, learn-only modes, operator enablement, monitoring, and restoration of saved parameters. These components make a bounded shadow-mode and low-criticality pilot technically feasible.","source_ids":["S5","S6","S8"],"uncertainty":"The sources do not validate the proposal's four-week durations, greater-than-10% halt threshold, cross-loop interaction controls, or suitability for any particular plant. DeltaV and Siemens capabilities are platform-specific."},"prior_art":{"disposition":"ESTABLISHED_PRACTICE","closest_analogues":[{"name":"Proportionate and risk-tiered management of change","source_ids":["S2","S3","S4"],"same_problem":true,"same_causal_lever":true,"overlap":"Official and industry guidance distinguishes minor or low-risk changes from significant or high-risk changes, assigns lighter competent-person approval to the former, escalates the latter, and supports predefined operating-parameter boundaries, time limits, documentation, and restoration.","remaining_difference":"The guidance is generic rather than a measured controller-retuning intervention with the proposal's historian metrics and numerical pilot halt rule."},{"name":"DeltaV Adapt approved-region adaptive PID tuning","source_ids":["S5"],"same_problem":true,"same_causal_lever":true,"overlap":"The product addresses changing process dynamics through continuous PID retuning, approved operating-region models, constrained tuning parameters, model validation, advisory shadow operation, and bounded partial-adaptive operation.","remaining_difference":"It does not itself classify loops by organizational risk or prove that approval queues cause poor performance at the target plant; explicit automatic restoration on the proposal's event triggers is not stated."},{"name":"SIMATIC self-tuning with authorization and saved-parameter restoration","source_ids":["S6","S8"],"same_problem":true,"same_causal_lever":true,"overlap":"Siemens documentation provides self-tuning, gain scheduling, bounded excitation, operator authorization to intervene, continuous observation, saved prior parameters, and an undo function.","remaining_difference":"The documented workflow still calls for coordination with responsible personnel and does not evaluate a preauthorized low-criticality class against a matched baseline."},{"name":"Historical adaptive and self-tuning process control","source_ids":["S7"],"same_problem":true,"same_causal_lever":true,"overlap":"Automatic controller-setting adjustment for changing processes, including experimental process-industry applications and constrained self-tuning antecedents, was established by the 1970s and surveyed in 1986.","remaining_difference":"The historical literature does not supply the proposal's contemporary MOC routing, cybersecurity controls, distributional review, or pilot governance."}],"contrastive_claim_remaining":"No credible broad claim remains that risk-tiered, bounded, reversible PID retuning is a new intervention package. A narrower site-specific claim remains falsifiable: at a plant that demonstrably routes every low-criticality tuning change through a slow uniform gate, preauthorizing changes within independently validated bounds will materially reduce approval latency and improve predetermined control-error, cycling, energy, alarm, or intervention metrics without degrading protected safety, maintenance, or service-quality metrics.","contrastive_claim_falsifier":"The remaining claim is falsified if procedural review contributes less than 5% of observed tuning latency; delayed eligible requests do not predict degraded performance; no independently defined low-risk class or containing envelope exists; or a matched bounded pilot produces no material latency or performance improvement or worsens any protected safety metric.","confidence":"MODERATE","search_limitations":"The bounded search used exactly eight retained direct sources and covered all six lanes. It found strong standards, product, historical, and non-English analogues but no public plant-level dataset joining approval timestamps to controller outcomes. Commercial prevalence, local procedures, paywalled details, unpublished operating records, and platform-specific configurations remain uncertain."},"researchability_gates":{"externally_supported_problem":{"status":"PASS","rationale":"Poor loop performance and organizational implementation barriers are externally documented, while multiple guidance sources explicitly differentiate minor, predefined, or low-risk changes from significant changes. The exact site's queue mechanism remains for testing rather than being assumed true.","source_ids":["S1","S2","S3","S4"]},"identifiable_adopter_or_authorizer":{"status":"PASS","rationale":"The plant control owner or employer, competent delegated line manager, operating personnel, and process-safety authority are identifiable authorization roles.","source_ids":["S2","S3","S4","S8"]},"distinct_testable_incremental_claim":{"status":"PASS","rationale":"Although the general package is established practice, the site-specific causal claim linking a uniform gate to latency and performance, and the effect of a matched bounded pilot, is distinct and falsifiable.","source_ids":["S1","S4","S5"]},"bounded_next_evidence_step":{"status":"PASS","rationale":"A retrospective log analysis followed, only if supported, by learn-only shadow mode and one redundant low-criticality loop is bounded, reversible, measurable, and compatible with existing product modes.","source_ids":["S2","S5","S6","S8"]},"no_unresolved_safety_or_authority_stop":{"status":"PASS","rationale":"No categorical stop appears if interlocks, trip thresholds, actuator limits, SIS functions, cybersecurity controls, and out-of-envelope changes remain excluded; competent plant and process-safety authorization is still mandatory. Failure to validate containment or independent risk classification would stop the pilot.","source_ids":["S2","S3","S4","S6","S8"]},"adequate_search_evidence":{"status":"PASS","rationale":"All six required lanes were searched adversarially, including historical adaptive-control terminology, official MOC guidance, commercial products, German-language documentation, and combined safety, rollback, authorization, and tuning subproblems. Eight retained direct sources span six independent publishing organizations and include multiple official and first-party sources.","source_ids":["S1","S2","S3","S4","S5","S6","S7","S8"]}},"strict_success":false,"screen_survival":false,"remaining_research_value":"MODERATE","recommended_next_step":"Obtain one candidate plant's change-request and controller-configuration audit logs. Pre-register eligibility independently of outcomes; compare procedural versus substantive latency; link delayed eligible requests to historian metrics; and proceed to learn-only shadow mode and then a time-limited matched pilot only if the retrospective problem threshold is met and the control owner plus process-safety authority approve the envelope and rollback logic.","world_novelty_boundary":"This bounded public-source review finds the broad intervention package already represented in risk-tiered MOC guidance and commercial adaptive/self-tuning control systems. It does not establish world novelty, patentability, freedom to operate, market size, routine adoption at every plant, or realized impact. Any remaining claim is limited to causal effectiveness and safe implementation in a specified local plant context."}