Anomaly Detection in Supply Chain Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing supply chain planning and execution systems face challenges in accurately detecting and removing faulty data anomalies, leading to reduced reliability and precision in supply chain plans and forecasts.
Innovation Solution
A system and method utilizing K-means clustering and local outlier factor analysis, combined with machine learning techniques, to automatically detect and categorize valid data outliers and faulty data anomalies in supply chain data, generating a prioritized list of anomalies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If static rules are used to identify faulty data, then the detection process is simple, but it results in false negatives and false positives requiring significant correction effort
Solution Approach 1:
The patent replaces static mechanical rules with a machine learning-based anomaly detection system that uses clustering algorithms (K-means) and local outlier factor analysis. This substitution enables the system to automatically learn patterns from data and identify anomalies adaptively, eliminating the need for manual rule creation while improving detection accuracy and reducing false positives.
Solution Approach 2:
The system performs self-learning by automatically analyzing data patterns and adjusting its anomaly detection criteria without human intervention. The machine learning models continuously improve their ability to distinguish between normal variations and actual anomalies, making the system self-optimizing and reducing the need for manual correction of false positives.
2Measurement precision
If machine learning techniques are applied to detect anomalies, then detection accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The patent segments the anomaly detection process into distinct stages: data preprocessing, clustering (K-means), local outlier factor calculation, and anomaly classification. This segmentation allows each component to process data independently and efficiently, reducing overall computational burden while maintaining high precision through specialized algorithms designed for specific tasks.
Solution Approach 2:
The system applies machine learning techniques selectively rather than uniformly across all data. By using clustering to group similar data points and then applying local outlier factor analysis only to identified clusters, the system reduces computational effort compared to analyzing all data points with full machine learning algorithms, while still achieving high detection precision.
3Reliability
If manual correction of false positives is performed, then detection accuracy improves, but time and effort requirements increase significantly
Solution Approach 1:
The system incorporates feedback mechanisms where detected anomalies are reviewed and confirmed, with results fed back into the machine learning models to refine future detections. This feedback loop continuously improves the system's accuracy, reducing false positives over time and eliminating the need for manual correction of most anomalies while maintaining high data quality.
Data Source
AI summary
A system and method are disclosed to locate one or more data anomalies in a supply chain network comprising two or more supply chain entities. Embodiments receive supply chain data comprising a plurality of data points. Embodiments select categories and measures by which to cluster the supply chain data points. Embodiments cluster the data points into intersection clusters as measured by the selected categories and measures. Embodiments generate time interval clusters that divide the intersection clusters into one or more time intervals. Embodiments generate K-values for the plurality of data points in relation to the number of time interval clusters. Embodiments generate local outlier factors for the plurality of data points using the generated K-values.