Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Interpretable Taxi-Out Time Prediction of Departure Flights Using Stacking Ensemble Learning and SHAP Analysis.

Tao Wu1, Yanfeng Mao1, Junchuan Huang2

  • 1China Academy of Civil Aviation Science and Technology, Beijing, China.

Scientific Reports
|February 20, 2026
PubMed
Summary

Related Concept Videos

Prediction Intervals01:03

Prediction Intervals

3.5K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y. 
3.5K
End Point Prediction: Gran Plot01:07

End Point Prediction: Gran Plot

1.3K
A Gran plot is used to predict the equivalence volume or endpoint of a potentiometric or acid-base titration without reaching the endpoint. Typically, titration data is collected as a function of the titrant's volume up to a point less than the equivalence volume and then transformed into a linear format. The straight line is extended to the x-axis, indicating the necessary titrant volume to achieve the equivalence point.
For potentiometric titration, the Gran plot is created by plotting...
1.3K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

DEENet: an edge-enhanced CNN-Transformer dual-encoder model for steel surface defect detection.

Scientific reports·2026
Same author

The mechanistic insights into sex determination of melon: Integrating environmental factors, hormones and genetic regulation.

Plant science : an international journal of experimental plant biology·2026
Same author

Social user geolocation based on K-medoids and Gaussian Kernel graph attention network.

Scientific reports·2026
Same author

Genomic and antigenic analysis of G1P[8] rotavirus strains in Ningxia, China.

International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases·2026
Same author

A three-decade lake dataset on the Mongolian plateau tracking water area and quality dynamics (1990-2020).

Scientific data·2025
Same author

Study on forecasting method of power engineering cost based on BIM and DynGCN.

PloS one·2025

This study introduces a new taxi-out time prediction model using Stacking ensemble learning and SHAP analysis for improved accuracy and interpretability in air traffic management.

Area of Science:

  • Aviation Operations Research
  • Artificial Intelligence in Transportation
  • Air Traffic Management Systems

Background:

  • Existing taxi-out time prediction models suffer from poor interpretability and generalization.
  • Accurate prediction of taxi-out time is crucial for efficient airport operations and air traffic control.

Purpose of the Study:

  • To develop a novel taxi-out time prediction model with enhanced interpretability and generalization.
  • To decompose taxi-out time and analyze influencing factors for improved prediction accuracy.
  • To leverage Stacking ensemble learning and Shapley Additive Explanations (SHAP) for model development and validation.

Main Methods:

  • Decomposition of taxi-out time into unimpeded and dynamic components.
  • Correlation analysis of influencing factors for each component.

Related Experiment Videos

  • Construction of a Stacking-based prediction model comparing holistic and phased approaches.
  • Implementation of SHAP analysis for feature importance quantification and model interpretability.

    • Main Results:

    • Unimpeded taxi-out time is primarily influenced by airport configuration; dynamic taxi-out time by surface traffic flow.
    • Phased prediction offers better interpretability with slightly lower performance (MAPE: 12.0%) compared to holistic prediction.
    • The Stacking model demonstrates superior accuracy (41.0% within ±60s) and generalization.
    • A dual feature selection mechanism (SHAP and correlation analysis) enhances prediction accuracy and reduces feature dimensions.
    • SHAP analysis effectively explains feature impacts and interactions, demystifying the model.

    Conclusions:

    • The proposed Stacking model with SHAP analysis significantly improves taxi-out time prediction accuracy and interpretability.
    • Decomposing taxi-out time and employing phased prediction provides valuable insights for air traffic controllers.
    • The model offers actionable intelligence for optimizing airport surface operations and decision-making.