Fatigue Risk Indicators of the Brazilian Regular Aviation
Fatigue Risk Indicators of the Brazilian Regular Aviation: First Results from Fadigômetro Research Project Capt. Tulio Rodrigues, Ph. D
After more than 6 years of hard work, the Brazilian Congress finally approved a new labour law for aircrew on July, 12 th 2017!
… But now, what about our next challenges and goals?
Upcoming structure of the Brazilian Regulations Airlines should implement a fatigue Risk Management System (FRMS) FRMS Airlines should implement a Fatigue Risk Management (FRM) Prescriptive limits (RBAC 117) Airlines do not require to implement Fatigue Risk Management (System) Prescriptive limits (Law 13. 475/17)
Outline: 1. The Fadigômetro project 2. Modelling the fatigue risk 3. Results: ² Fatigue indicators from Fadigômetro ² PVT data, modelling and FDM pilot errors 4. Conclusions and lessons learned 5. Future work 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 5
1. The Fadigômetro Project In July of 2017, the Representative Institutions signed a Technical Cooperation Agreement! 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 6
1. The Fadigômetro Project Idealizers: Participants: 3/11/2021 Supporters: FRMS Forum 2019 Meeting San Francisco, California, USA 7
1. The Fadigômetro Project Objectives: (1) to perform a statistical mapping of the human fatigue in a convenient sample of the Brazilian regular aviation; (2) to identify cognitive performance degradation hazards; (3) to determine fatigue risk exposure in the Brazilian regular aviation, and (4) to propose fatigue risk analyses and mitigation strategies. Confidentiality and Ethical aspects: It is assured that all the personal data and any specific data from a given airline will not be disclosure. The participants joined the experiment voluntarily and anonymously. The project was approved by the Bioscience Institute Research Ethics Committee of the University of São Paulo (CAAE: 89058318. 7. 0000. 5464/Report: 2. 716. 654). 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 8
1. The Fadigômetro Project 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 9
1. The Fadigômetro Project ü Advantages of the tool: 1. Scientific reliability (without political interference); 2. Full confidentiality of participants and airlines; 3. Random sample of rosters; 4. Fatigue evaluation considering the combined effects of realistic and multiple prescriptive limits; 5. The establishment of a data base to support the National Commission of Human Fatigue (CNFH) in the updating process of its Fatigue Investigation Methodology. 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 10
1. The Fadigômetro Project ü More advantages 6. Identification of potential hazards and the most fatiguing schedules (hot spots); 7. Relative risk analyses during regulatory changes (new prescriptive limits versus current limits); 8. Evaluation of BMM parameters/criteria taking into account realistic operational circumstances and also supplementary information from social, health and behaviour optional questionnaire. 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 11
1. The Fadigômetro Project Rosters of Airline A Rosters of Airline B Rosters of Airline C File management and conversion web platform (to generate unidentified csv files) SAFTE-FAST Console Results and Reports Statitical Analyses of SAFTE-FAST outputs
2. Modelling the fatigue risk Definitions and parameters (SAFTE-FAST Model): • Critical Phase: first and last 30 minutes of each sector. • SAFTE-FAST Effectiveness (ESF): varies from 0 100%. [An effectiveness score of 77% is equivalent to a BAC of 0. 05% when measuring the reaction time of the individuals (*)]. • Minimum Effectiveness critical (EMC): lowest value of ESF during the critical phases. • Hazard Area Critical (HAC): the area of ESF line shape that drops below 77% during the critical phases. (*) http: //www. sleepapnea. org/assets/files/2011%20 Conference/Tuesday/Steven%20 Hursh. pdf 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 13
2. Modelling the fatigue risk Accident costs versus ESF (*) Low Risk 90% Moderate Risk ESF 77% High Risk (*) S. Hursh, J. Fanzone and T. Raslear (2011) Analysis of the Relationship between Operator Effectiveness Measures and Economic Impacts of Rail Accidents. (Report No. DOT/FRA/ORD-11/13). Washington, DC: U. S. Department of Transportation. 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 14
2. Modelling the fatigue risk HF Accidents x Fatigue a = 0. 27 ± 0. 20 b = 0. 58 ± 0. 16 c 2 = 3. 02 DOF = 4 Prob (c 2 > 3. 02) = 55. 5% Data from: S. Hursh, T. Raslear, A. Kaye and J. Fanzone, Jr. (2006). (Report No. DOT/FRA/ORD 06/21). Washington, DC: U. S. Department of Transportation.
3. Results: Fatigue indicators from Fadigômetro General Statistics (until 17 -Sep-2019) participants 502 number of rosters 8584 3/11/2021 responders of the questionnaire 474 male female 341 (71. 9%) 133 (28. 1%) pilots 256 (54. 0%) cabin 218 (46. 0%) total duty hours (h) 994 k total number of sectors 236 k FRMS Forum 2019 Meeting San Francisco, California, USA 16
3. Results: Fatigue indicators from Fadigômetro High Season Narrow-Body operations for rosters processed until 28 -Sep-2018 Low Season High Season 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 17
3. Results: Fatigue indicators from Fadigômetro High Season Narrow-Body operations for rosters processed until 28 -Sep-2018 Low Season High Season 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 18
3. Results: Fatigue indicators from Fadigômetro Normality Test Analyses Non-parametric independent samples tests Kolmogorov-Smirnov Variables Minimum Effectiveness critical (EMC) Group K DOF Jan 18 0. 146 248 May 18 0. 095 259 Jul 18 0. 133 261 Jan 18 Hazard Area Critical (HAC) 0. 212 Kruskal-Wallis p 0. 76 259 Jul 18 0. 245 261 N <0. 001 H DOF p 46. 248 Jan 18, May 18 & Jul 18 248 May 18 Groups Mann-Whitney 768 2 49. 829 <0. 001 Groups N Z Jan 18 & May 18 507 -5. 866 May 18 & Jul 18 520 -5. 886 Jan 18 & Jul 18 509 -0. 289 Jan 18 & May 18 507 -6. 146 May 18 & Jul 18 520 -6. 131 Jan 18 & Jul 18 509 -0. 234 p <0. 001 0. 773 <0. 001 H 0 hypothesis is confirmed when comparing two high season periods 0. 815
3. Results: Fatigue indicators from Fadigômetro The study* received the Award “Dr. Bernardo Bedrikow - Inovação e novas práticas em Saúde e Segurança do Trabalho” from the 17 st National Congress of the National Association of Occupational Medicine! (*) Rodrigues TE, Fischer FM, Bastos EM, Baia L, Bocces R, Gonçalves FP, Licati PR, Menquini A, Spyer P, Stefenon E and Helene AF. Systematic Evaluation of Fatigue Indicators in the Brazilian Regular Aviation: Fadigômetro Project. Submitted to the Brazilian Journal of Occupational Medicine, 2019.
3. Results: PVT data, modelling and FDM pilot errors PVT data from: Roma PG, Hursh SR, Mead AM, Nesthus TE. Flight Attendant Work/Rest Patterns, Alertness, and Performance Assessment: Field Validation of Biomathematical Fatigue Modeling. (Report No. DOT/FAA/AM-12/12). Washington, DC: Office of Aerospace Medicine; 2012.
3. Results: PVT data, modelling and FDM pilot errors PVT data from: Roma PG, Hursh SR, Mead AM, Nesthus TE. Flight Attendant Work/Rest Patterns, Alertness, and Performance Assessment: Field Validation of Biomathematical Fatigue Modeling. (Report No. DOT/FAA/AM-12/12). Washington, DC: Office of Aerospace Medicine; 2012.
3. Results: PVT data, modelling and FDM pilot errors Fitting model PVT data A 1 A 2 Sigmoid Lapses 5. 24 ± 0. 24 3. 40 ± 0. 08 Linear RT (msec) 330 ± 4 301. 2 ± 1. 4 Speed 3. 617 ± 0. 022 3. 795 ± 0. 011 Lapses 7. 7 ± 0. 8 -0. 044 ± 0. 009 E 0 (%) DE (%) 81. 16 ± 0. 65 2. 83 ± 0. 52 -- c 2 & DOF p 17. 1 & 16 0. 38 54. 0 & 18 1. 9 x 10 -5 -- RT (msec) 382 ± 14 -0. 84 ± 0. 15 -- -- Speed 3. 22 ± 0. 09 (6. 0 ± 1. 0)x 10 -3 -- -- Preliminary
3. Results: PVT data, modelling and FDM pilot errors PVT Lapses versus ESF (Sigmoid model) Preliminary Data from: Roma PG, Hursh SR, Mead AM, Nesthus TE. Flight Attendant Work/Rest Patterns, Alertness, and Performance Assessment: Field Validation of Biomathematical Fatigue Modeling. (Report No. DOT/FAA/AM-12/12). Washington, DC: Office of Aerospace Medicine; 2012.
3. Results: PVT data, modelling and FDM pilot errors Reaction Time and PVT Speed versus ESF (Sigmoid model) Preliminary Dada from: Roma PG, Hursh SR, Mead AM, Nesthus TE. Flight Attendant Work/Rest Patterns, Alertness, and Performance Assessment: Field Validation of Biomathematical Fatigue Modeling. (Report No. DOT/FAA/AM-12/12). Washington, DC: Office of Aerospace Medicine; 2012.
3. Results: PVT data, modelling and FDM pilot errors Average SAFTE-FAST Effectiveness as a function of time of the day Preliminary 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 26
3. Results: PVT data, modelling and FDM pilot errors Average PVT Lapses versus time of the day L (ESF) ESF(h) L (h) Preliminary
3. Results: PVT data, modelling and FDM pilot errors Average PVT Lapses versus time of the day Preliminary 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 28
3. Results: PVT data, modelling and FDM pilot errors Pilot errors relative likelihood increases ~ 50% between 0: 00 and 05: 59! 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 29
3. Results: PVT data, modelling and FDM pilot errors Pilot Error: objective data versus Fadigômetro predictions +44% +20% 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 30
4. Conclusions and lessons learned • The Fadigômetro research project provided reliable indicators for the fatigue risk quantification in a convenience sample of rosters from pilots and fight attendants of the Brazilian regular aviation via the SAFTE-FAST model; • The distributions of minimum effectiveness during the critical phases indicate high fatigue likelihood (between 1/2 and 3/4 of the monthly rosters with at least one event with ESF < 77%) and the need of a more suitable management of the flight schedules; 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 31
4. Conclusions and lessons learned • The distributions of EMC and HAC show significant seasonal oscillations. These results will be relevant in relative analyses after the implementation of the new rules (RBAC 117); • The combination of neurobehavioral measures with the time of the day average SAFTE-FAST effectiveness from Fadigômetro allowed a consistent interpretation (but still preliminary) of objective FDM pilot errors; 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 32
4. Conclusions and lessons learned • A preliminary analysis of the bio-psychosocial questionnaire from Fadigômetro suggests the need for customization of few parameters and criterion of SAFTE-FAST; • The results from Fadigômetro demonstrate that the current prescriptive limits in Brazil are not sufficient to mitigate the fatigue risks; • The State FRMS regulation should not be based on a “copy and paste” approach. Each State shall develop its own regulations taking into account its peculiarities and based on scientific knowledge, data and operational experience. 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 33
5. Future work • Development of specific filters for the investigation of the root causes of fatigue in the crew rosters; • Customization of few parameters and criterion of SAFTE-FAST with the help of the bio-psychosocial questionnaire; • Continue the data acquisition and interpretation in order to: (1) help the National Commission of Human Fatigue to update the “Guidance Manual for Fatigue Identification in Aeronautical Occurrences”, and (2) propitiate relative fatigue risk analyses after the new regulations are effective. 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 34
Fadigômetro Team Faculty Members/Researchers: • Prof. Andre Frazão Helene, Ph. D - Bioscience Institute - University of São Paulo • Prof. Frida Fischer, Ph. D - School of Public Health - University of São Paulo • Capt. Tulio Rodrigues, Ph. D - Physics Institute - University of São Paulo and ASAGOL Technical Committee: Operational Committee (staff members): • • • F. O. Eduardo Antunes - SNA F. O. Eduardo Mantovani - ABRAPAC Capt. Luciano Baia - ATL Capt. Paulo Licati - ABRAPAC Capt. Paulo Spyer - ATL Capt. Raul Bocces - ASAGOL Mr. Alfredo Menquini - ABRAPAC Mr. Eduardo Morteo - ASAGOL Mr. Fabiano Gonçalves - ATL Mrs. Karina Silva - SNA
Contact info: E-mail: tulio@if. usp. br Phone: +55 -11 -9 -82891105 Thank-you! 3/11/2021 FRMS Forum 2019 Meeting San Francisco, California, USA 36
- Slides: 36