Virtual Interline Route Optimization with Feedback-Based Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to computationally optimize travel routes involving multiple carriers lacking interline agreements, making it impractical to identify efficient combinations that meet individual traveler preferences due to the complexity of combining segments from disparate carriers.
Innovation Solution
A system that assigns weights to virtual interline routes based on user feedback and comparison with real routes, dynamically updating these weights to rank and select optimal travel combinations, reducing computational burden and network congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If every conceivable route formed by combining segments from every potential carrier is calculated, then complete route coverage is achieved, but computational complexity becomes infeasible
Solution Approach 1:
The patent segments the route optimization problem into distinct components: (1) identifying individual flight segments from multiple carriers, (2) filtering segments based on user preferences and constraints, (3) combining filtered segments into candidate routes, and (4) ranking candidate routes. This segmentation reduces computational complexity by processing manageable portions rather than evaluating all possible route combinations simultaneously.
Solution Approach 2:
The system performs preliminary filtering of flight segments before route combination, pre-processing data by removing segments that don't meet user criteria. This preliminary action reduces the dataset size early in the process, making subsequent route combination and optimization computationally feasible while maintaining reliability of results.
2Adaptability or versatility
If route combinations are narrowed down to suit specific user needs, then user preference satisfaction is improved, but the difficulty of identifying optimal routes increases
Solution Approach 1:
The patent applies local quality by implementing different filtering and evaluation criteria for different segments of the route based on user preferences. Each flight segment can be evaluated against specific user requirements (e.g., preferred carriers, time constraints, price ranges), allowing customized optimization without requiring complete re-evaluation of all routes. This localized approach accommodates user preferences while managing optimization difficulty.
Solution Approach 2:
The system incorporates feedback mechanisms where user preferences and selections from previous queries inform subsequent route recommendations. The system learns from user interactions and adjusts filtering criteria and route rankings accordingly, improving adaptability to user needs over time while reducing the difficulty of identifying optimal routes through iterative refinement.
3Adaptability or versatility
If data from multiple disconnected carrier systems is integrated, then travel choice variety is improved, but system complexity and data management burden increase
Solution Approach 1:
The patent introduces an intermediary system that acts as a mediator between multiple disconnected carrier systems and the user interface. This intermediary layer collects, standardizes, and manages flight segment data from various carriers without requiring direct integration between carrier systems. It handles data formatting, constraint application, and route assembly, thereby improving travel option diversity while containing system integration complexity within the intermediary layer.
Data Source
AI summary
A system and method for generating optimized travel itineraries using virtual interlining. A system receives a travel request from an application installed on a client device, the request including a plurality of user-defined parameters. The system identifies a set of candidate virtual interline routes between an origin and a destination based on a graph constructed from historical user travel data, where each route includes segments from different carriers lacking interline agreements. Each route is associated with a weight determined in part by prior user input and comparisons with historical data for corresponding real routes. A subset of the parameters is applied to a constraint function, and the candidate routes are ranked based on the function's output and the route weights. The system selects itineraries that satisfy the user's parameters, including at least one itinerary with components from non-contiguous segments, and generates for display updated travel data using the client-side application.


