Ride-sharing Vehicle Allocation Graph Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ride-sharing systems face inefficiencies in vehicle allocation, particularly when customers are traveling to different destinations, leading to high route deviations and increased carbon emissions, without considering factors like traffic conditions and demand-supply patterns, resulting in unsatisfactory experiences for customers and losses for service providers.
Innovation Solution
A method and system for optimizing vehicle allocation in ride-sharing environments using a graph-based approach that considers multiple share-ride parameters such as route deviation, gross merchandise value, sharing efficiency, number of customers, pick-up time, and customer satisfaction, along with historical data and real-time booking requests, to allocate vehicles efficiently and reduce emissions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If vehicles are allocated to customers traveling to different destinations, then more customers can be served simultaneously, but route deviation increases and customer satisfaction decreases
Solution Approach 1:
The system dynamically adjusts the share-ride parameter thresholds based on real-time conditions such as traffic patterns, demand-supply ratios, and customer preferences. By changing the parameters of route deviation tolerance and sharing efficiency requirements, the system can accommodate more customers in shared rides without excessively compromising route efficiency, thus resolving the contradiction between serving more customers and maintaining acceptable route deviations.
2Device complexity
If ride-sharing is implemented without considering traffic conditions and demand-supply patterns, then vehicle allocation is simplified, but allocation optimality and customer experience deteriorate
Solution Approach 1:
The allocation system automatically incorporates real-time traffic conditions, demand-supply patterns, and historical data without requiring manual intervention. The system self-adjusts by processing multiple share-ride parameters and making optimal allocation decisions autonomously, thus maintaining high allocation optimality while keeping the operational complexity manageable through automation rather than manual complexity.
3Ease of operation
If vehicles travel with minimum occupancy on individual bookings, then customer privacy and comfort are maintained, but carbon emissions and environmental impact increase
Solution Approach 1:
The system merges multiple individual booking requests into shared rides by identifying customers with compatible routes and timing. By combining separate trips into consolidated shared journeys, the system reduces the total number of vehicles on the road and minimizes carbon emissions while still providing privacy-protected individual booking experiences within the shared vehicle environment.
4Speed
If ride-sharing allocation is based on single factor such as shortest-path, then allocation speed is maintained, but overall optimization and customer satisfaction decrease
Solution Approach 1:
The system performs preliminary calculations and pre-processes booking requests by organizing them into groups based on route compatibility and timing. This preliminary action allows the system to quickly evaluate pre-processed data using multiple share-ride parameters without sacrificing allocation speed, thus achieving both fast allocation and comprehensive optimization by having data ready for multi-parameter evaluation before final matching occurs.
Data Source
AI summary
Vehicle allocation method and system for ride-sharing are provided. The method includes receiving a first set of booking requests is received, at a first time instance, from a set of customer devices for sharing one or more rides. A second set of booking requests is determined, at a second time instance, based on the first set of booking requests and a third set of booking requests. The third set of booking requests is determined based on at least one of historical booking data and booking requests received after the first time instance. A set of vehicles available for the one or more rides is determined at the second time instance. A graph is generated based on the second set of booking requests and the set of available vehicles. An available vehicle is allocated to one or more customers based on optimal matching between nodes of the graph.