Vehicle Vocation Classification Using Telematics Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods struggle to accurately determine a vehicle's vocation based on telematics data, as they often rely on inherent vehicle properties rather than operational behavior, leading to difficulties in categorizing vehicles with similar characteristics used for different purposes.
Innovation Solution
A method involving the retrieval and processing of telematics data to extract feature datasets related to distances traveled, stops completed, and round trips, which are then analyzed using trained cluster models, such as Gaussian mixture models, to determine vocation probabilities and classify vehicles into categories like local, regional-haul, door-to-door, and hub-and-spoke.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If vehicles are categorized by inherent properties (make, model, GVWR), then vehicles with similar characteristics can be grouped together, but vehicles used for different purposes cannot be distinguished
Solution Approach 1:
The patent transforms the classification basis from static inherent parameters (make, model, GVWR) to dynamic operational parameters (distance traveled, stops completed, round trips). This parameter change enables the system to distinguish vehicles by their actual usage patterns rather than just their physical characteristics, thereby improving classification accuracy while preserving operational behavior information through the telematics data collection system.
2Measurement precision
If telematics data is collected from vehicles, then operational behavior information is available, but it is challenging to determine how to use the data to determine vocation
Solution Approach 1:
The patent segments the complex telematics data into three distinct feature datasets: distance traveled, stops completed, and round trips. Each dataset is processed separately through specific cluster models, which simplifies the overall analysis complexity while maintaining high measurement precision. This segmentation approach allows the system to handle large volumes of telematics data effectively by breaking down the complex determination process into manageable, specialized analysis streams.
Solution Approach 2:
The patent introduces trained cluster models as intermediary components between the raw telematics data and the final vocation determination. These cluster models act as mediators that automatically process and interpret the feature datasets, converting complex operational data into meaningful vocation classifications. This intermediary layer reduces the complexity burden on the overall system while enhancing measurement precision through specialized analytical algorithms.
3Measurement precision
If multiple feature datasets are analyzed using multiple cluster models, then vocation classification precision is improved, but processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-training multiple specialized cluster models offline before deployment. These pre-trained models are ready to process feature datasets immediately without requiring complex real-time training computations. The distance cluster model, stops cluster model, and round trips cluster model are each pre-configured with their specific analytical parameters, enabling rapid processing of incoming telematics data while maintaining high classification accuracy.
Solution Approach 2:
The patent segments the analysis into three parallel processing streams, each handled by a dedicated cluster model. This segmentation allows simultaneous processing of different feature datasets without sequential bottlenecks, reducing overall processing time while maintaining comprehensive analysis. Each cluster model operates independently on its specific dataset, enabling parallel computation that preserves measurement precision without proportionally increasing total processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for determining vehicle vocation are provided. The method involves operating at least one processor to: retrieve telematics data associated with a vehicle, the telematics data originating from a telematics device installed in the vehicle; process the telematics data to extract a plurality of feature datasets, the plurality of feature datasets including: a first dataset associated with distances traveled by the vehicle; a second dataset associated with stops completed by the vehicle; and a third dataset associated with round trips performed by the vehicle; apply a plurality of trained cluster models to the plurality of feature datasets to determine a plurality of vocation probabilities; and determine the vocation of the vehicle based on the plurality of vocation probabilities.