ETA Prediction Using ETC Toll Data and Vehicle Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current GPS-based systems for predicting the expected time of arrival (ETA) of freight vehicles face challenges due to the use of general data instead of specific vehicle data, lack of real-time access to important data, and technology limitations, leading to reduced accuracy and increased costs.
Innovation Solution
The proposed system uses specific data for each vehicle in real-time to predict ETA, leveraging data from electronic toll collection (ETC) systems, vehicle type characteristics, and data from adjacent vehicles to enhance accuracy. This system also utilizes machine learning algorithms and statistical methods to identify similar vehicles and calculate ETA.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If GPS-based tracking systems use general aggregated data for all vehicle types, then the system complexity is reduced and ease of operation is improved, but the measurement precision of ETA prediction deteriorates
Solution Approach 1:
The patent segments the homogeneous vehicle population into heterogeneous groups based on vehicle type characteristics (size, weight, cargo type). Instead of using a single aggregated speed profile for all vehicles, the system creates separate speed profiles for different vehicle categories (e.g., empty trucks, loaded trucks, different cargo types), thereby improving prediction accuracy while maintaining system simplicity through standardized segmentation criteria.
Solution Approach 2:
The patent applies local quality by tailoring the speed prediction to the specific characteristics of each vehicle type rather than using a universal approach. Each vehicle type receives customized speed profiles based on its physical characteristics, cargo type, and operational constraints, allowing the system to maintain ease of operation through automated classification while achieving higher measurement precision through vehicle-specific predictions.
2Reliability
If GPS devices are installed in all freight vehicles for real-time tracking, then the reliability of data collection is improved, but the loss of substance (cost) increases due to initial investment and periodic maintenance
Solution Approach 1:
The patent introduces electronic toll collection (ETC) systems as an intermediary data source. Instead of directly installing GPS devices in vehicles, the system leverages the existing ETC infrastructure to obtain location and speed data. This intermediary approach maintains reliable data collection through the established toll collection network while eliminating the need for separate GPS device installation and maintenance in each vehicle.
Solution Approach 2:
The system utilizes the ETC system's own data collection capabilities to serve the tracking function. The ETC infrastructure, already in place for toll collection, automatically provides location and speed data as part of its normal operation, eliminating the need for additional dedicated tracking devices and reducing overall system costs while maintaining data reliability.
3Ease of manufacture
If historical data is used for ETA prediction, then the ease of manufacture and data availability are improved, but the productivity of prediction accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-processing and organizing historical speed data into vehicle-type-specific profiles before actual ETA prediction is needed. The system预先 segments historical data by vehicle characteristics and creates standardized speed profiles for different vehicle types, making the actual prediction process more efficient and accurate when real-time predictions are required.
Solution Approach 2:
The system transitions from static historical averages to dynamic, vehicle-type-specific speed profiles. By continuously updating and refining speed predictions based on accumulated data for each vehicle category, the system adapts to changing conditions while maintaining the ease of using pre-processed historical data, thereby improving prediction accuracy without sacrificing operational simplicity.
4Ease of operation
If data from vehicles with GPS devices is collected and fragmented across multiple vendors, then the ease of operation is improved, but the loss of information increases due to incomplete dataset
Solution Approach 1:
The patent merges data from multiple sources including ETC systems, GPS devices, and other vendors into a unified dataset. By consolidating fragmented information from different vendors and supplementing with ETC data, the system creates a complete vehicle-type-specific speed profile that captures the full range of operational conditions, thereby reducing information loss while maintaining ease of data collection through automated integration.
Data Source
AI summary
Provided is a system for predicting expected time of arrival (ETA) of vehicles. The scanning module is installed at a toll plaza for electronic toll collection (ETC) scan of a tracked vehicle and same and similar (SaSi) vehicles crossing the toll plaza. The SaSi vehicles are identified based on a plurality of predetermined factors. The computation module computes a segment time for plurality of segments from time difference between the ETC scans in two consecutive toll plazas for the SaSi vehicles moving ahead of the tracked vehicle. The segment time is computed based on a defined criteria. A segment time is computed for plurality of segments in an event of presence of non-SaSi vehicles or absence of toll road. Based on the computed segment time and periodic error correction module between successive segments, the ETA of the tracked vehicle is predicted.