Travel Time Estimation Using Payment Card Data and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack an efficient method to determine travel time to an airport departure gate, particularly in real-time, accounting for passenger congestion and varying transportation modes, which can lead to delays and uncertainty for travelers.
Innovation Solution
A computer-based system that utilizes payment card transaction data and machine learning models to estimate user-specific airport processing time and travel time, integrating real-time flight information, TSA waiting times, and geographical location data to provide accurate departure times for each transportation mode, updated in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time data collection and machine learning modeling are implemented to calculate user-specific airport processing time, then measurement precision of travel time is improved, but device complexity increases
Solution Approach 1:
The patent introduces a server as an intermediary component that hosts the machine learning model and processes data. The server receives real-time data from multiple sources (payment card transactions, flight information, TSA waiting times), performs the complex machine learning calculations, and returns results to user devices. This intermediary architecture isolates the computational complexity from user devices while maintaining high measurement precision.
Solution Approach 2:
The server infrastructure is designed to perform multiple functions: collecting real-time data from diverse sources, hosting and executing machine learning models, processing payment card transaction data, integrating flight information, and providing results to multiple users simultaneously. This multi-functional design consolidates complexity into a single platform that serves many purposes.
2Reliability
If multiple real-time data sources are integrated to account for passenger congestion and transportation modes, then reliability of travel time estimation is improved, but device complexity increases
Solution Approach 1:
The patent merges multiple independent data sources (payment card transaction data indicating passenger presence, flight information systems, TSA waiting time databases, and navigation system transportation data) into a unified model on the server. This consolidation integrates diverse data types into a single coherent travel time estimation, improving reliability while centralizing integration complexity.
Solution Approach 2:
The system continuously collects real-time data from multiple sources and uses machine learning to process this feedback loop. Payment card transactions provide feedback on actual passenger presence and congestion patterns, which are fed back into the model to continuously refine travel time estimates. This feedback mechanism enhances reliability by adapting to current conditions.
3Measurement precision
If user-specific airport processing time is calculated based on passenger congestion from payment card data, then measurement precision is improved, but loss of information increases due to privacy concerns
Solution Approach 1:
The system extracts only the necessary information from payment card transaction data - specifically, the presence and timing of passengers at the airport - while leaving out sensitive personal information. The machine learning model uses aggregated transaction patterns to infer congestion levels without requiring or storing individual user identities, account numbers, or personal details, thus maintaining measurement precision while protecting privacy.
Solution Approach 2:
The patent applies different levels of data processing to different types of information. Payment card transaction data undergoes local transformation into anonymized congestion metrics at the airport location, while personal identifying information is excluded entirely. This local quality approach allows the system to use transaction data for its intended purpose (measuring congestion) without compromising user privacy elsewhere.
Data Source
AI summary
A method and system include identifying, by a processor, departing flight information that designates departure airports and departure times in payment card transaction data of a plurality of users. Airport-specific data for a departure airport before a departure time of a departing flight of a user from the plurality of users is received. The airport-specific data is inputted into a machine learning model that outputs a user-specific airport processing time for the user to reach a departure gate upon arriving to the departure airport. A travel time from a geographical location of the computing device of the user to the departure airport is received from a navigation system. The computing device displays a time for the user to start travel to the departure airport based on the user-specific airport processing time and the travel time to the departure airport for the user to reach the departure gate by the departure time.