Cognitive Utility Path Prediction Using Spatial Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for locating buried utilities are often inaccurate and prone to damage, especially when tracer wires are destroyed, making it difficult and expensive to find underground conduits without prior markings, and current systems are limited in versatility and effectiveness.
Innovation Solution
A cognitive system is trained with historical data from known utility installations to predict the path of underground utilities by applying spatial correlation and cognitive analysis, using geographic and non-geographic features to generate candidate paths with confidence scores, allowing for more accurate and efficient location prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If electronic markers or warning tapes are used to indicate buried utilities, then the presence of utilities can be detected, but the accuracy and reliability of location information deteriorates when markers are damaged, lost, or destroyed
Solution Approach 1:
The system performs preliminary actions by collecting and analyzing historical utility installation data, geographic features, and environmental information before excavation occurs. This predictive approach establishes utility paths in advance without relying on physical markers that could be damaged or lost, thereby maintaining reliable location information throughout the utility's service life.
Solution Approach 2:
The patent replaces mechanical marker systems (warning tapes, painted symbols, physical markers) with a cognitive computing system that uses spatial correlation and historical data analysis. This substitution eliminates the vulnerability of physical markers to damage while maintaining the ability to indicate utility locations through digital prediction and visualization.
2Reliability
If ground-penetrating radar or electromagnetic detection methods are used to locate buried utilities, then utility paths can be detected without markers, but the cost and complexity of the detection process increases significantly
Solution Approach 1:
The system creates a digital copy or model of utility paths by analyzing historical installation data, geographic features, and spatial relationships. This cognitive model replicates the information that would otherwise require expensive physical detection equipment, providing accurate utility location predictions through data analysis rather than complex electromagnetic scanning.
Solution Approach 2:
The patent develops a universal cognitive system that can predict multiple types of utilities (water, gas, electric, communication) using the same historical data and spatial correlation methodology. This multi-functional approach replaces the need for specialized detection equipment for each utility type, reducing overall system complexity while maintaining comprehensive detection capability.
3Ease of manufacture
If manual excavation and visual inspection methods are used to locate utilities, then no specialized equipment is needed, but the time consumption and labor costs increase significantly
Solution Approach 1:
The system performs self-service by automatically analyzing historical utility data, geographic features, and spatial relationships to generate predicted utility paths without requiring manual excavation or visual inspection. The cognitive system independently processes multiple data sources and produces location predictions, eliminating the need for time-consuming manual search methods while maintaining implementation simplicity through software-based solutions.
4Measurement precision
If extensive historical data and multiple geographic features are analyzed to improve prediction accuracy, then utility path prediction reliability improves, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent segments the complex prediction task into distinct analytical components: spatial correlation of geographic features, temporal analysis of historical installation data, identification of relevant environmental factors, and confidence score calculation. This segmentation allows the cognitive system to process multiple data sources systematically, improving prediction accuracy while managing computational complexity through modular processing steps.
Data Source
AI summary
The most likely path for an underground utility having an unknown location is predicted using artificial intelligence. A cognitive system is trained with details associated with historical utility installations whose underground paths are known. The system is applied to whatever installation details are available for the unknown underground utility such as geographic features which are subjected to spatial correlation to derive geographic locations relative to a region of interest for the underground utility. Cognitive analysis is performed on the locations in association with the features to generate candidate paths for the utility. The candidate paths are presented to the user along with computed confidence scores. The installation details may further include nongeographic features used in the cognitive analysis such as a date of installation of the utility or an entity associated with the utility.