Delay Attribution

Delay attribution focuses on identifying the root causes of delays in complex systems, aiming to improve efficiency and predictability. Current research employs machine learning techniques, such as hierarchical classification models (e.g., Random Forests, Support Vector Machines) and reinforcement learning (e.g., Behavioral Cloning, Conservative Q-Learning), to automate delay analysis and prediction across diverse applications. These efforts are significant for optimizing resource allocation and improving decision-making in areas like air traffic management, train operations, and autonomous driving, where timely and accurate delay attribution is crucial for safety and efficiency. The challenge lies in accurately modeling complex, real-world uncertainties and developing robust algorithms that can handle fluctuating delays and diverse data sources.

Papers

May 14, 2024

Deep Reinforcement Learning for Real-Time Ground Delay Program Revision and Corresponding Flight Delay Assignments
Ke Liu, Fan Hu, Hui Lin, Xi Cheng, Jianan Chen, Jilin Song, Siyuan Feng, Gaofeng Su, Chen Zhu
Reinforcement Learning Deep Reinforcement Learning Air Traffic Flight Delay Delay Prediction Delay Attribution

February 6, 2024

Hierarchical Delay Attribution Classification using Unstructured Text in Train Management Systems
Anton Borg, Per Lingvall, Martin Svensson
Unstructured Text High Speed Train System Public Transit Delay Attribution

September 13, 2023

MTD: Multi-Timestep Detector for Delayed Streaming Perception
Yihui Huang, Ningjiang Chen
Autonomous Driving Multi Horizon Streaming Perception Multi Timestep Delay Attribution

Delay Attribution

Papers

Deep Reinforcement Learning for Real-Time Ground Delay Program Revision and Corresponding Flight Delay Assignments

Hierarchical Delay Attribution Classification using Unstructured Text in Train Management Systems

MTD: Multi-Timestep Detector for Delayed Streaming Perception