Dynamic Ride Assignment for High-Capacity Ride Pooling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ride-sharing systems are inefficient in handling large-scale ride pooling due to limitations in matching multiple riders with shared vehicles in real-time, primarily restricted to two passengers, and lack scalable mathematical models and algorithms to optimize vehicle allocation.
Innovation Solution
A system and technique that utilizes a reactive anytime optimal method to dynamically generate optimal routes and assign ridesharing requests to vehicles, capable of handling up to 10 simultaneous passengers per vehicle, and rebalances the fleet to service demand efficiently, using a pairwise request-vehicle graph and request-trip-vehicle graph to solve the unified problem of passenger and vehicle assignment through integer linear programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If ride-sharing systems restrict to two passengers per vehicle, then vehicle allocation is simpler, but ride-pooling efficiency and scalability are limited
Solution Approach 1:
The system segments the ride-sharing problem into discrete matching units using graph theory, where riders and vehicles are nodes and potential matches are edges. This segmentation allows complex multi-passenger allocation to be broken down into manageable matching operations that can be solved efficiently
Solution Approach 2:
The patent introduces mathematical models and algorithms as intermediaries between ride requests and vehicle assignments. These algorithms act as mediators that process multiple passenger requests simultaneously and generate optimized vehicle allocation plans, resolving the complexity of matching large groups without requiring direct complex interactions
2Adaptability or versatility
If real-time matching of large groups of riders to fleet vehicles is implemented, then ride-sharing capacity increases, but computational complexity and processing time increase
Solution Approach 1:
The system changes the parameter of vehicle capacity from the conventional limit of 2 passengers to accommodate large groups of riders. This parameter change enables high-capacity ride-pooling where vehicles can transport entire groups simultaneously, fundamentally increasing ride-sharing capacity while the mathematical models ensure computational tractability
3Productivity
If optimal routes are generated for multiple passengers with different destinations, then ride-pooling efficiency improves, but route optimization complexity increases
Solution Approach 1:
The system performs preliminary route optimization by pre-calculating optimal paths that accommodate multiple passengers with different destinations. The mathematical models determine efficient routing strategies in advance, considering all passenger destinations and vehicle capacities, before actual ride assignments are made
Solution Approach 2:
The route optimization system serves multiple functions simultaneously: it optimizes for ride-pooling efficiency, minimizes total travel distance, accommodates various destination combinations, and adapts to different vehicle capacities. This universal optimization approach handles diverse routing scenarios through a unified mathematical framework
Data Source
AI summary
Described are concepts, systems, devices and techniques for real-time high-capacity ride-sharing that (i) scales to large numbers of passengers and trips and (ii) dynamically generates routes (e.g. optimal or near optimal routes) in response to online demand and vehicle locations.