Buried Environment Classification Maps for Pipe Deterioration Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for predicting the deterioration of buried pipes, such as water pipes, lack accuracy in assessing the burial environment and its impact on pipe corrosion, leading to inefficiencies in maintenance and repair planning.

Innovation Solution

A system that utilizes machine learning to create an optimized burial environment classification map by selecting grounds based on past water leakage data, combining it with geological maps and pipe information to predict pipe deterioration accurately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If a general burial environment classification map is created based on pipeline map and geological map, then the coverage area is large, but the prediction accuracy for pipe deterioration is insufficient

Engineering Contradiction:
Improvecoverage areaVSAvoidprediction accuracy
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent segments the burial environment classification into two levels: a general classification map covering the entire region and an optimized classification map for specific areas with leakage accidents. This segmentation allows the system to maintain broad coverage while achieving high prediction accuracy in critical areas by applying machine learning optimization selectively to regions with actual leakage data.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If machine learning is applied to optimize burial environment classification for all grounds, then the prediction accuracy improves, but the computational complexity and time increase

Engineering Contradiction:
Improveprediction accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies machine learning optimization locally only to grounds where water leakage accidents have occurred, rather than uniformly across the entire region. This local quality approach concentrates computational resources on areas where accurate prediction is most critical and where sufficient training data exists, thereby improving prediction accuracy while minimizing computational complexity and processing time.

Inventive Principle:
Principle #3Local quality

3Ease of manufacture

If only general geological map data is used for classification, then the data collection is simple, but the specific burial environment conditions are not accurately captured

Engineering Contradiction:
Improvedata collection simplicityVSAvoidburial environment classification accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent merges multiple data sources including pipeline maps, geological maps, and water leakage accident records to create a comprehensive burial environment classification system. This combination allows the system to maintain the simplicity of using publicly available geological data while significantly improving classification accuracy by integrating it with actual leakage accident information to identify and optimize critical areas.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4641475A1Buried environment classification map creating device, buried pipe deterioration degree predicting device, buried environment classification map creating method, buried pipe deterioration degree predicting method, and program
Publication Date: 2025.10.29 KUBOTA CORP
  • EP4641475A1 patent drawingFigure 1
  • EP4641475A1 patent drawingFigure 2~3
  • EP4641475A1 patent drawingFigure 4~5

AI summary

A buried pipe deterioration degree prediction apparatus (1) includes a buried pipe deterioration degree calculation unit (132). The buried pipe deterioration degree calculation unit (132) calculates a deterioration degree for each of the buried pipes based on a buried pipe deterioration degree prediction model and a burial environment of each of the buried pipes identified by an optimized burial environment classification map (58a). The optimized burial environment classification map (58a) is created by optimizing the burial environment classification of a part of grounds in a general burial environment classification map (56a) by machine learning. The part of grounds is selected based on water leakage accident data (53) which is a past record of water leakage accidents for each of the buried pipes.