Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports

Background: Modeling tools should be tested against real-world outcomes to confirm their predictive ability compared to random chance. Insights is an analytical tool within the biomathematical modeling software SAFTE-FAST that identifies work patterns that consistently result in elevated fatigue ris...

Full description

Bibliographic Details
Published in:Safety
Main Authors: Jaime K. Devine, Jake Choynowski, Steven R. Hursh
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Subjects:
Online Access:https://www.mdpi.com/2313-576X/11/1/4
_version_ 1849872380765143040
author Jaime K. Devine
Jake Choynowski
Steven R. Hursh
author_facet Jaime K. Devine
Jake Choynowski
Steven R. Hursh
author_sort Jaime K. Devine
collection DOAJ
container_title Safety
description Background: Modeling tools should be tested against real-world outcomes to confirm their predictive ability compared to random chance. Insights is an analytical tool within the biomathematical modeling software SAFTE-FAST that identifies work patterns that consistently result in elevated fatigue risk. This study investigated the ability of Insights to correctly identify duties with an associated fatigue report using previously collected flight schedule and report data. Methods: Planned and completed flight roster schedules were analyzed using SAFTE-FAST Insights after the rosters had been flown. Fatigue reports were independently linked to planned and completed schedules at the duty level. Odds ratio (OR) analysis investigated the ability of Insights to predict which duties would be linked to a fatigue report. Differences in duties were compared using a one-way analysis of variance (ANOVA) and a two-sample <i>t</i>-test. Results: There were 157 fatigue reports out of 78,061 planned duties and 235 fatigue reports out of 82,612 completed duties. Insights had 3.04 odds of correctly identifying fatigue reports in planned duties but 0.41 odds for completed duties. Discussion: Insights showed good odds of correctly identifying a fatigue report duty using planned schedules but poor odds of identifying a fatigue report duty from completed schedules. Completed duties started later in the day and were shorter in duration than planned duties. Day-of-operations schedule changes may have reduced the fatigue risk in response to the fatigue reports.
format Article
id doaj-art-17e9f12db7ec4d00a62580a77abe72f0
institution Directory of Open Access Journals
issn 2313-576X
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
spelling doaj-art-17e9f12db7ec4d00a62580a77abe72f02025-08-20T01:13:47ZengMDPI AGSafety2313-576X2025-01-01111410.3390/safety11010004Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue ReportsJaime K. Devine0Jake Choynowski1Steven R. Hursh2Institutes for Behavior Resources, 2104 Maryland Ave, Baltimore, MD 21218, USAInstitutes for Behavior Resources, 2104 Maryland Ave, Baltimore, MD 21218, USAInstitutes for Behavior Resources, 2104 Maryland Ave, Baltimore, MD 21218, USABackground: Modeling tools should be tested against real-world outcomes to confirm their predictive ability compared to random chance. Insights is an analytical tool within the biomathematical modeling software SAFTE-FAST that identifies work patterns that consistently result in elevated fatigue risk. This study investigated the ability of Insights to correctly identify duties with an associated fatigue report using previously collected flight schedule and report data. Methods: Planned and completed flight roster schedules were analyzed using SAFTE-FAST Insights after the rosters had been flown. Fatigue reports were independently linked to planned and completed schedules at the duty level. Odds ratio (OR) analysis investigated the ability of Insights to predict which duties would be linked to a fatigue report. Differences in duties were compared using a one-way analysis of variance (ANOVA) and a two-sample <i>t</i>-test. Results: There were 157 fatigue reports out of 78,061 planned duties and 235 fatigue reports out of 82,612 completed duties. Insights had 3.04 odds of correctly identifying fatigue reports in planned duties but 0.41 odds for completed duties. Discussion: Insights showed good odds of correctly identifying a fatigue report duty using planned schedules but poor odds of identifying a fatigue report duty from completed schedules. Completed duties started later in the day and were shorter in duration than planned duties. Day-of-operations schedule changes may have reduced the fatigue risk in response to the fatigue reports.https://www.mdpi.com/2313-576X/11/1/4fatigue risk managementbiomathematical modelingfatigue reportingaviationodds ratio aviation accidents and investigationsaviation safety
spellingShingle Jaime K. Devine
Jake Choynowski
Steven R. Hursh
Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
fatigue risk management
biomathematical modeling
fatigue reporting
aviation
odds ratio aviation accidents and investigations
aviation safety
title Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
title_full Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
title_fullStr Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
title_full_unstemmed Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
title_short Evaluation of a Biomathematical Modeling Software Tool for the Prediction of Risk in Flight Schedules Compared Against Incidence of Fatigue Reports
title_sort evaluation of a biomathematical modeling software tool for the prediction of risk in flight schedules compared against incidence of fatigue reports
topic fatigue risk management
biomathematical modeling
fatigue reporting
aviation
odds ratio aviation accidents and investigations
aviation safety
url https://www.mdpi.com/2313-576X/11/1/4
work_keys_str_mv AT jaimekdevine evaluationofabiomathematicalmodelingsoftwaretoolforthepredictionofriskinflightschedulescomparedagainstincidenceoffatiguereports
AT jakechoynowski evaluationofabiomathematicalmodelingsoftwaretoolforthepredictionofriskinflightschedulescomparedagainstincidenceoffatiguereports
AT stevenrhursh evaluationofabiomathematicalmodelingsoftwaretoolforthepredictionofriskinflightschedulescomparedagainstincidenceoffatiguereports