{"id":"W3088075317","doi":"10.1016/j.ssci.2020.104981","title":"Application of Bayesian network and artificial intelligence to reduce accident/incident rates in oil & gas companies","year":2020,"lang":"en","type":"article","venue":"Safety Science","topic":"Risk and Safety Analysis","field":"Decision Sciences","cited_by":64,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Process safety; Incident report; Bayesian network; Asset (computer security); Process (computing); Reliability (semiconductor); Accident analysis; Poison control; Engineering; Computer science; Computer security; Risk analysis (engineering); Operations management; Transport engineering; Machine learning; Work in process; Business","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004450315,0.0001412828,0.0003852427,0.00028931,0.0003107259,0.000202961,0.001416234,0.00004271759,0.00007663511],"category_scores_gemma":[0.002886006,0.0001161697,0.0000698051,0.006772197,0.0006173696,0.0005094358,0.0004351028,0.0001330846,0.0001838316],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005173768,"about_ca_system_score_gemma":0.000158592,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003134353,"about_ca_topic_score_gemma":0.001856503,"domain_scores_codex":[0.9961815,0.0001469029,0.001031601,0.0008323732,0.001434269,0.0003733062],"domain_scores_gemma":[0.9979078,0.0006099024,0.0002958831,0.0005363537,0.000333758,0.0003162803],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0002212854,0.0000403929,0.06228956,0.000005517598,0.000005049807,0.000001800199,0.004728646,0.1373667,0.00701071,0.01738103,0.0001479508,0.7708014],"study_design_scores_gemma":[0.0001089769,0.0001436416,0.363008,0.00006425215,0.00002379007,0.000006653978,0.006022254,0.3144334,0.009404926,0.3018506,0.004409192,0.0005242805],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.523155,0.0002396939,0.4484219,0.02706147,0.000159815,0.0001994871,0.000006478908,0.00002701667,0.0007291037],"genre_scores_gemma":[0.9914695,0.0002235856,0.007783183,0.0003896354,0.00009000314,0.000008130786,0.000001642299,0.000004590775,0.00002969997],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7702771,"threshold_uncertainty_score":0.4737264,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08699401901798096,"score_gpt":0.3838259639997584,"score_spread":0.2968319449817775,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}