Carpool Matching System Using Two-Phase Travel Time Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing carpooling systems face computational inefficiencies when processing large numbers of drivers and passengers, leading to slowed response times and frustration from users due to unnecessary cancellation and rescheduling notifications.
Innovation Solution
The system estimates travel times using straight line distances and maximum vehicle speeds to quickly exclude ineligible passengers, then calculates actual road distances and times for eligible passengers, and automatically reassesses carpool assignments upon cancellations or rescheduling requests to minimize disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system calculates actual road distances and travel times for all driver-passenger combinations, then matching accuracy is improved, but computational intensity increases exponentially
Solution Approach 1:
The patent segments the calculation process into two distinct phases: (1) a fast filtering phase that uses straight-line distance and maximum speed to quickly eliminate ineligible passenger combinations, and (2) a detailed calculation phase that computes actual road distances and travel times only for the remaining eligible combinations. This segmentation resolves the contradiction by performing rough filtering first to reduce the problem size, then applying precise calculations only where necessary.
Solution Approach 2:
The patent performs preliminary filtering using simplified assumptions (straight-line distance and maximum vehicle speed) before committing to more computationally intensive actual route calculations. By establishing eligibility criteria in advance using the fast filtering method, the system prepares a reduced set of candidate combinations that are then subjected to detailed travel time calculations, thereby avoiding unnecessary computations for clearly ineligible pairs.
2Reliability
If the system processes carpooling requests in real-time with comprehensive calculations, then assignment quality is improved, but response time increases
Solution Approach 1:
The patent divides the real-time processing into two sequential stages: rapid pre-filtering using geometric distance and speed limits, followed by detailed route optimization only for qualified candidates. This ensures that the system responds quickly to new requests while still performing comprehensive calculations for the subset of viable matches, thereby maintaining assignment quality without excessive delay.
Solution Approach 2:
The system performs preliminary eligibility assessment using computationally inexpensive metrics (straight-line distance divided by maximum speed) before initiating time-consuming route calculations. This preliminary action establishes a threshold that must be met before detailed processing occurs, ensuring that real-time response requirements are met while preserving the ability to perform thorough optimization for promising candidates.
3Loss of information
If the system sends notifications for every cancellation and rescheduling, then user information awareness is improved, but user frustration increases due to unnecessary notifications
Solution Approach 1:
The patent extracts and removes clearly ineligible driver-passenger combinations from further processing using the fast filtering method. By taking out these obviously unsuitable pairs before detailed evaluation, the system avoids generating misleading or unnecessary notifications for combinations that would fail basic feasibility criteria, thereby reducing user frustration while maintaining notification completeness for viable options.
Solution Approach 2:
The patent creates a simplified model or copy of the travel time calculation using straight-line distance and maximum speed assumptions. This approximate copy is used for initial filtering and notification decisions, allowing the system to quickly determine whether a combination is worth pursuing further without committing to full route calculation, thus avoiding unnecessary notifications for obviously infeasible matches.
Data Source
AI summary
Distances between locations traveled by a carpool driver in a carpooling system may be initially estimated by calculating direct, straight line distances between each of the location points. Travel speeds may also be initially estimated using an expected maximum vehicle speed, which may be a maximum speed limit. An estimated travel time may then be calculated from this data to initially designate passengers as eligible or ineligible for carpooling with a carpool driver.