System and method for flight delay prediction

Patent metadata

Field	Value
Publication number	US-2019108758-A1
Application number	US-201816153102-A
Country	US
Kind code	A1
Filing date	Oct 5, 2018
Priority date	Oct 6, 2017
Publication date	Apr 11, 2019
Grant date	—

Abstract

This disclosure relates generally to a system and method to predict flight delay. Moreover, the embodiments herein further provide the system and method to predict timings of an airline in real time by considering historical operations (arrival and departure) data, historical airport data (captured at the time of arrival and departure) including congestion information, and weather data of the airport. The flight delays involves prediction of arrival and departure times of flight. Herein, the method categorizes input data related to an airline history, airline network, airport data and various airline reference data. Further, the method analyses the cause of delay which may be due to maintenance issues with the aircraft, fueling, weather, congestion in air traffic, and security issues etc. The system and method computes the flight delay due to multiple airline operations and different input datasets using stochastic approximation approach.

First claim

The claim is claimed:

1 . A system to predict flight delay, the system comprising:
at least one memory storing a plurality of instructions; one or more hardware processors communicatively coupled with the at least one memory, wherein the one or more hardware processors are configured to execute one or more modules; a receiving module configured to receive a plurality of historical operation data of the flight, a historical data of an airline operating the flight, a real time airlines data, a planned airlines data, an aircraft type data, a historical weather data and a real time weather data as an input for the system; an operation characterization module configured to analyze one or more operational scenarios considering one or more dimensional aspects and the received input to the system, wherein the one or more operational scenarios includes fleet deployment pattern, network flow characterization, and operational preferences; a learning module configured to learn the analyzed one or more operational scenarios, wherein the learning of operational scenarios comprises one or more operational levers, wherein the one or more operational levers include needs of a crew, operations of one or more gates, and a rescheduling priorities; a determination module configured to predict a taxi-in time, a taxi-out time, and an air time of the flight, wherein the taxi-in time is defined as time between a wheels-on time and a gate-in time, taxi-out time is defined as time between the actual pushback and takeoff of the flight, and the air time of the flight includes total time from a time that the aircraft first moves under its own power for the purpose of taking off until the time the aircraft comes to rest at the end of the flight; and a prediction module configured to predict the flight delay using a real time flight information, one or more operational scenarios, one or more operational levers, and the predicted taxi-in time, taxi-out time, and air time of the flight.

2 . The system claimed in claim 1 , wherein the one or more dimensional aspects include origin destination pairs and connections, a flight frequency, an operational delay classification, clock times, fleet types, an operational crew data and a network model.

3 . The system claimed in claim 1 , wherein the air time of the flight depends on one or more factors including a congested airspace, weather, traffic control actions, and type of the aircraft.

4 . The system claimed in claim 1 , wherein the taxi-out time and the taxi-in time of the flight depends on one or more factors including runway configuration, downstream restrictions, and arrival queue.

5 . A processor-implemented method to predict flight delay, the processors-implemented method comprising one or more steps of:
receiving, via the one or more hardware processors, a historical operation data of the flight, a historical airport data, and a weather data, as an input at a receiving module of the system; analyzing, via the one or more hardware processors, one or more operational scenarios at an operation characterization module of the system considering one or more dimensional aspects and the received input, wherein the one or more operational scenarios includes fleet deployment pattern, network flow characterization, and operational preferences; learning, via the one or more hardware processors, the analyzed one or more operational scenarios at a learning module of the system, wherein the learning of operational scenarios defines one or more operational levers, wherein the one or more operational levers include needs of a crew, operations of one or more gates, and a rescheduling priorities; predicting, via the one or more hardware processors, a taxi-in time, a taxi-out time, and air time of the flight at a determination module of the system, wherein the taxi-in time is defined as time between a wheels-on time and a gate-in time, taxi-out time is defined as time between the actual pushback and takeoff of the flight, and the air time of the flight includes total time from the moment that an aircraft first moves under its own power for the purpose of taking off until the moment the aircraft comes to rest at the end of the flight; predicting, via the one or more hardware processors, flight delay at a decision module of the system considering a real time flight information, one or more operational scenarios, one or more operational levers, and the predicted time of taxi-in, taxi-out, and air time of the aircraft.

6 . The method claimed in claim 5 , wherein the one or more dimensional aspects include origin destination pairs and connections, a flight frequency, an operational delay classification, a clock times, fleet types, an operational crew data and a network model.

7 . The method claimed in claim 5 , wherein the air time of the flight depends on one or more factors such as a congested airspace, weather, traffic control actions, and a type of the aircraft.

8 . The method claimed in claim 5 , wherein the taxi-out time and taxi-in time of the flight depends on one or more factors such as runway configuration, downstream restrictions, and arrival queue.

9 . A non-transitory computer readable medium storing one or more instructions which when executed by a processor on a system, cause the processor to perform method for predicting flight delay comprising:
receiving, via the one or more hardware processors, a historical operation data of the flight, a historical airport data, and a weather data, as an input at a receiving module of the system; analyzing, via the one or more hardware processors, one or more operational scenarios at an operation characterization module of the system considering one or more dimensional aspects and the received input, wherein the one or more operational scenarios includes fleet deployment pattern, network flow characterization, and operational preferences; learning, via the one or more hardware processors, the analyzed one or more operational scenarios at a learning module of the system, wherein the learning of operational scenarios defines one or more operational levers, wherein the one or more operational levers include needs of a crew, operations of one or more gates, and a rescheduling priorities; predicting, via the one or more hardware processors, a taxi-in time, a taxi-out time, and air time of the flight at a determination module of the system, wherein the taxi-in time is defined as time between a wheels-on time and a gate-in time, taxi-out time is defined as time between the actual pushback and takeoff of the flight, and the air time of the flight includes total time from the moment that an aircraft first moves under its own power for the purpose of taking off until the moment the aircraft comes to rest at the end of the flight; predicting, via the one or more hardware processors, flight delay at a decision module of the system considering a real time flight information, one or more operational scenarios, one or more operational levers, and the predicted time of taxi-in, taxi-out, and air time of the aircraft.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Byravan Satish
• Padmanabhan Kishore
• Natarajan Vijayarangan
• Narasimhadevara Kartick
• Ganesan Viswanath Kumar
• Gudla Sreedhar
• Perumal Ramesh Babu Sangaiaha
• Balakrishnan Subramaniam
• Jagannathan Balaji

Classifications

• G06Q10/1097

  using calendar-based scheduling for task assignment · CPC title

• G08G5/00 Primary

  Traffic control systems for aircraft · CPC title

• G06Q10/06 Primary

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title