Ridesharing Zone Reachability for Real-Time Vehicle Supply Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ridesharing management systems face challenges in accurately predicting the number of vehicles needed for a particular zone in real-time, requiring extensive user inputs and simulations that take days to produce results, and lack flexibility to switch between different service scenarios or adjust boundary geographies quickly.
Innovation Solution
A system that estimates the number of ridesharing vehicles needed based on historical data and user inputs, allowing for real-time or near real-time predictions of vehicle supply, and enables switching between fixed route, on-demand, and hybrid scenarios, with the ability to easily modify boundary geographies for immediate results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simulation methods are used to predict vehicle supply, then prediction accuracy can be improved, but the time required increases to days
Solution Approach 1:
The patent creates a simplified digital copy of the ridesharing system that replicates key behaviors and patterns from historical data. This virtual model can be queried instantly without running full simulations, providing accurate predictions in real-time by copying essential system dynamics rather than simulating every detail.
Solution Approach 2:
The system pre-processes historical ridesharing data to extract patterns, trends, and behavioral characteristics before they are needed for prediction. By performing this analysis in advance and storing the results in the digital model, the system can provide immediate predictions without needing to run time-consuming simulations when predictions are required.
2Reliability
If multiple simulation parameters are input to predict supply, then prediction comprehensiveness improves, but the number of user inputs increases to dozens or seventy parameters
Solution Approach 1:
The patent extracts only the most critical patterns and variables from the full set of simulation parameters that actually drive ridesharing supply dynamics. By identifying and isolating these key factors from the dozens of potential parameters, the system maintains comprehensive prediction capability while reducing user input requirements to a manageable few essential variables.
Solution Approach 2:
The digital model serves multiple functions simultaneously: it captures complex system behaviors, processes historical data, identifies patterns, and generates predictions all through a unified framework. This multi-functional approach eliminates the need for separate analysis steps and multiple parameter inputs, as the model inherently handles comprehensiveness through its integrated design.
3Ease of manufacture
If fixed boundary geographies are used for service zones, then service area definition is simplified, but the system lacks flexibility to adjust to changing demands or events
Solution Approach 1:
The patent implements dynamic boundaries that can automatically adjust based on real-time conditions, historical patterns, and user-defined parameters. The digital model allows service zones to expand, contract, or shift shape in response to changing demand patterns, events, or operational requirements, transforming static geographic definitions into adaptive, living boundaries that evolve with system needs.
4Measurement precision
If simulations are run to determine boundary geography impact, then geographic analysis accuracy improves, but the time required increases to days
Solution Approach 1:
The system creates a virtual replica of the geographic service model that can be instantly queried to assess boundary changes. Instead of running full geographic simulations, the digital copy allows users to immediately see how different boundary configurations would impact vehicle supply and service coverage, providing accurate geographic analysis in real-time without simulation delays.
Data Source
AI summary
Methods and systems for managing a fleet of ridesharing vehicles is provided. The methods and systems can allow for a communications interface to receive a plurality of user input boundaries specified on a geographical map, wherein the plurality of user input boundaries define a zone. The methods and systems also include at least one processor configured to estimate a number of riders based on a number of people that live within the zone and a number of people that work within the zone, to determine n number of sample trips within the zone, and to determine a number of ridesharing vehicles to supply for the zone over a time duration based on the estimate number of riders and the n number of sample trips within the zone.