Synthetic Ride Demand Modeling for Data-Sparse Locations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fleet management systems lack an efficient solution for predicting vehicle ride demand in areas with no previous record of vehicle activity, making it difficult to optimize routes and operations.
Innovation Solution
A method and system that generate synthetic demand data for a new location by analyzing historic demand data from a similar location using a clustering algorithm, such as KNN, and integrating map data to determine points of interest and generate pick-up and drop-off events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If historic demand data from a second location is used to generate synthetic demand data for a first location with no previous records, then demand prediction capability is improved for new locations, but data accuracy and reliability may deteriorate due to the synthetic nature of the generated data
Solution Approach 1:
The patent creates synthetic demand data for a first location by copying and adapting the statistical patterns and characteristics from historic demand data of a second location. The system generates synthetic pick-up and drop-off events that replicate the behavioral patterns observed in the source location, enabling demand prediction in areas without historical records while maintaining realistic demand characteristics through statistical replication
Solution Approach 2:
The system performs preliminary data preparation by pre-processing historic demand data from the second location before deployment to the first location. This includes pre-calculating statistical parameters, demand patterns, and spatial-temporal characteristics in advance, so that when the system needs to predict demand for new locations, it can quickly generate accurate synthetic data without real-time computation delays
2Adaptability or versatility
If a clustering algorithm like KNN is used to generate synthetic demand data, then the system can operate without historical data at the target location, but the complexity of the data processing system increases
Solution Approach 1:
The patent introduces a clustering algorithm as an intermediary processing layer between the source historic data and the target synthetic data generation. The KNN algorithm acts as a mediator that learns the underlying patterns from historic demand data at the second location and translates them into synthetic demand patterns for the first location, enabling the system to operate without direct historical data at the target while managing complexity through algorithmic mediation
Solution Approach 2:
The system changes the parameters of demand data representation by transforming detailed historic ride records into aggregated statistical parameters and patterns. By converting raw data into condensed statistical representations (such as demand density, temporal patterns, spatial distributions), the system reduces data complexity while preserving essential demand characteristics needed for synthetic data generation
Data Source
AI summary
The present invention provides a method, apparatus, and computer program product for generating synthetic demand data of vehicle rides corresponding to a first location. The method comprises obtaining historic demand data of vehicle rides corresponding to a second location. The historic demand data comprises at least one first point of interest (POI) associated with at least one of a pick-up event or a drop-off event. The method further comprises retrieving map data associated with the second location and further, determining at least one second POI associated with the first location. The method further comprises generating the synthetic demand data of vehicle rides corresponding to the first location, based on the at least one first POI, the map data, and the at least one second POI.