Ridesharing Pickup Wait Time Prediction Using Probe Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ridesharing companies face challenges in accurately predicting pickup wait times, especially in urban areas where the distance between service providers and requestors is short and data uncertainties are high, leading to inaccurate wait time estimates and a degraded user experience.
Innovation Solution
A system that collects and processes probe data from vehicles that have previously traversed the requested route to determine the most recent and route-specific individual vehicle speed data, providing a high-confidence wait time prediction and continuously updating it based on real-time traffic conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction methods are used to estimate pickup wait time, then the system is simple to operate, but the prediction accuracy deteriorates in short-distance urban scenarios with high data uncertainty
Solution Approach 1:
The patent segments the prediction task by separating probe data collection from final prediction computation. Probe vehicles independently collect and report their trajectory data to a server, which then aggregates this data to generate predictions. This segmentation allows the system to handle complex data processing centrally while keeping individual vehicle systems simple.
Solution Approach 2:
The patent introduces a server as an intermediary between probe vehicles and the prediction system. The server collects probe data from multiple vehicles, processes the information, and generates wait time predictions. This intermediary handles the computational complexity of data aggregation and uncertainty management, shielding end users from system complexity.
2Measurement precision
If probe data from multiple probe vehicles is collected and processed to identify travel times, then the prediction accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent merges data from multiple probe vehicles that have traversed similar routes. By combining trajectories from several vehicles and identifying common route segments, the system accumulates sufficient data to make accurate travel time predictions even for short distances where individual data points are uncertain.
Solution Approach 2:
The system performs preliminary actions by pre-collecting and storing probe data from multiple vehicles before a prediction is needed. This pre-processing creates a ready pool of trajectory information that can be quickly queried and processed when a wait time prediction is required, reducing real-time computational burden.
3Reliability
If real-time probe data is continuously collected and processed, then the prediction reliability improves under changing traffic conditions, but the computational load and energy consumption increase
Solution Approach 1:
The system implements periodic action by having probe vehicles report their trajectories at regular intervals or at key location points along their routes. Rather than continuously transmitting data, vehicles provide updates periodically, reducing communication overhead and energy consumption while still capturing sufficient information to detect traffic pattern changes.
Solution Approach 2:
Probe vehicles perform self-service by independently collecting and reporting their own trajectory data without requiring active intervention or heavy processing at the time of data generation. The vehicles autonomously track their positions and submit this passive probe data to the server, minimizing the energy and computational resources required at the vehicle端.
Data Source
AI summary
An approach is provided for accurate travel time prediction. The approach involves, for example, determining probe data collected within a threshold proximity of a vehicle location and a pickup location. The probe data may be collected, for example, from a location sensor of at least one probe vehicle that has previously traversed the vehicle location and the pickup location. The approach also involves processing the probe data to identify a travel time of a route taken by the at least one probe vehicle from the vehicle location to the pickup location. The approach also involves providing the travel time as an output indicating a wait time for a vehicle at the vehicle location to reach the pickup location.