Asynchronous AI Data Fusion for Service Degradation Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing centralized machine learning models struggle to efficiently process and analyze both internal and external data in real-time to predict and address service degradation proactively without human intervention.
Innovation Solution
An asynchronous artificial intelligence (AAI) system that continuously collects internal and external data, trains an AI/ML model, and autonomously identifies service degradation issues, proposing proactive actions to improve service quality and minimize risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If centralized machine learning models process and analyze both internal and external data in real-time, then service degradation prediction accuracy is improved, but system complexity and computational resource requirements increase
Solution Approach 1:
The patent segments the centralized machine learning model into multiple distributed instances deployed across different service providers. Each instance processes local internal data independently while collectively consuming external data from multiple sources. This segmentation reduces the complexity burden on any single system while maintaining high prediction accuracy through ensemble capabilities.
Solution Approach 2:
The external data sources serve multiple functions simultaneously: they provide features for prediction models, enable anomaly detection, support service quality monitoring, and facilitate proactive intervention strategies. This multi-functionality reduces overall system complexity by consolidating multiple specialized systems into a unified platform.
2Reliability
If the system continuously collects and processes real-time data from multiple sources, then service quality monitoring capability is improved, but data processing time and computational energy consumption increase
Solution Approach 1:
Each distributed machine learning instance processes local internal data (specific to individual service providers) independently with optimized computational resources, while sharing only essential aggregated results. This local processing approach reduces overall energy consumption compared to a single centralized system processing all data, while maintaining comprehensive service quality monitoring.
Solution Approach 2:
The system implements continuous data collection and processing operations that run asynchronously without interruption. Background services continuously ingest data from multiple sources, update models in real-time, and prepare predictions before service degradation occurs. This continuous operation ensures reliable monitoring while optimizing energy usage through efficient batch processing and selective data transmission.
3Measurement precision
If the AI/ML model is trained on diverse internal and external data, then prediction accuracy for service degradation is improved, but data integration complexity and processing delays increase
Solution Approach 1:
The patent divides the data integration task across multiple distributed instances, each handling internal data from their respective service providers locally. External data is aggregated at the distributed level rather than requiring centralized processing. This segmentation enables parallel data processing, maintains prediction accuracy through diverse data training, and eliminates single-point processing bottlenecks that cause delays.
Solution Approach 2:
The system performs preliminary data processing and feature extraction at the source locations before data needs to be integrated into the prediction model. Pre-processing steps including data cleaning, normalization, and feature engineering are completed locally in advance, reducing the time required for subsequent data integration and model training operations.
Data Source
AI summary
An asynchronous artificial intelligence (AAI) system that is configured to receive internal data associated with a customer of the AAI system, wherein the customer is a service provider; search, in an ongoing manner, one or more external data sources that are external from the customer to obtain external data; and send machine learning output to the customer that presents a proposal of a customer action to address a predicted service degradation of the service provider based on the input of the internal data and the external data to an AI or machine learning (AI/ML) model that is trained based on historical data of the customer.