BEV Rental Booking Using Charge Prediction and Dynamic Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rental vehicle systems face challenges in optimizing vehicle suggestions, minimizing redundancy, dynamic scheduling, and predicting late returns, which affect fleet utilization and user experience.
Innovation Solution
A computer-implemented system that uses virtual voters to rank rental vehicles based on user preferences, groups vehicles by relevant properties, and employs dynamic scheduling to adjust bookings and predict charge levels for battery electric vehicles, while also notifying users and adjusting subsequent bookings for late returns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If virtual voters are used to rank vehicles based on multiple preference profiles, then user preference accuracy is improved, but system complexity increases
Solution Approach 1:
Virtual voters serve as intermediary entities that represent different user preference profiles. Instead of directly implementing complex preference matching logic, the system introduces virtual voter objects that encapsulate preference weightings and ranking behaviors, simplifying the overall system architecture while maintaining high preference accuracy
Solution Approach 2:
The system creates multiple copies of user preference profiles in the form of virtual voters. Each virtual voter is a copy representing a different weighting combination of vehicle criteria, allowing the system to evaluate and rank vehicles from multiple preference perspectives simultaneously without requiring complex real-time calculations
2Loss of information
If vehicles are grouped by relevant properties to minimize redundancy, then information clarity is improved, but processing time increases
Solution Approach 1:
The system performs preliminary grouping of vehicles by relevant properties before presenting results to users. By pre-organizing vehicles into groups based on properties like vehicle type, price range, or features, the system reduces information redundancy and improves clarity without requiring complex real-time processing during user interaction
Solution Approach 2:
The vehicle fleet is segmented into distinct groups based on relevant properties. This segmentation allows the system to present organized, non-redundant information to users while maintaining efficient processing by working with pre-defined property categories rather than analyzing individual vehicle attributes in real-time
3Productivity
If dynamic scheduling is used to adjust vehicle assignments, then fleet utilization is improved, but operational complexity increases
Solution Approach 1:
The scheduling system implements dynamic vehicle assignments that can be adjusted in real-time based on changing conditions. Vehicle assignments are not fixed but can be rearranged automatically to optimize fleet utilization, allowing the system to adapt to new booking requests, cancellations, or vehicle availability changes without manual intervention
Solution Approach 2:
The dynamic scheduling system incorporates feedback mechanisms that monitor fleet utilization and automatically adjust vehicle assignments. When conditions change (new bookings, cancellations, vehicle maintenance), the system receives feedback and re-optimizes assignments accordingly, maintaining high utilization while reducing operational complexity through automated decision-making
4Reliability
If charge levels are predicted for battery electric vehicles, then booking reliability is improved, but computational requirements increase
Solution Approach 1:
The system predicts charge levels for battery electric vehicles in advance before booking decisions are made. By performing charge level predictions preliminarily, the system ensures that only vehicles with sufficient charge are considered for booking, improving booking reliability without requiring complex real-time calculations during the booking process
Data Source
AI summary
Techniques are described for optimizing various aspects of rental vehicle systems. According to an embodiment, a system is described for optimizing fleet utilization of a battery electric vehicle (BEV) rental system based on predicted charge levels. The system comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. These computer executable components comprise a charge prediction component that determines forecasted charge levels of BEVs in association with usage of the BEVs in a vehicle sharing operation wherein the BEVs are available for renting for varying durations of time. The computer executable components further comprise a booking component that controls the renting of the BEVs based on the forecasted charge levels.