Network of Time Series Platform for Multi-Source Data Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for time series analysis lack a standardized platform to effectively integrate and predict values across multiple time series, limiting their ability to leverage complex global relationships and provide actionable insights for businesses.
Innovation Solution
A network of time series (NOTS) platform that standardizes time series data using a layering scheme, allowing for the combination of sources, operations, and models as nodes, enabling users to customize data evaluation and predict future, present, or past values by integrating multiple time series.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a standardized NOTS platform is implemented to integrate multiple time series, then the ability to predict values and leverage complex relationships is improved, but the device complexity and system architecture become more complex
Solution Approach 1:
The system segments time series analysis into independent standardized nodes that can be individually developed, deployed, and maintained. Each node represents a specific time series or processing unit with defined interfaces, allowing the overall system to scale without proportionally increasing complexity. This modular segmentation enables organizations to start with simple implementations and progressively add complexity only where needed.
Solution Approach 2:
The patent implements universal standardized interfaces and protocols that allow different time series sources (sensor data, business metrics, market data) to be integrated through common connection points. This universality enables a single platform architecture to handle diverse data types and prediction scenarios without requiring custom integration logic for each case, thereby improving versatility while controlling complexity through reuse.
2Quantity of substance
If hundreds or thousands of time series are integrated into the network, then the scale and value of predictions increase, but the difficulty of detecting and measuring relationships becomes more difficult
Solution Approach 1:
The system introduces intermediary components including standardized data schemas, protocol buffers for data exchange, and intermediate processing nodes that translate between different time series formats. These intermediaries act as mediators that simplify the detection of relationships by providing uniform interfaces and preprocessing capabilities, reducing the complexity of analyzing relationships across thousands of time series.
Solution Approach 2:
The patent transforms the complexity of detecting relationships across many time series by adding structural dimensions through hierarchical organization and temporal abstraction. Instead of directly analyzing relationships between all pairs of time series, the system introduces intermediate temporal aggregations and hierarchical groupings that reduce the dimensional complexity of relationship detection while preserving predictive insights.
3Ease of operation
If a layering scheme is used to customize data evaluation, then the ease of operation and customization improve, but the device complexity increases due to multiple layers
Solution Approach 1:
The layering scheme is implemented dynamically rather than as a fixed static structure. Users can configure, enable, or disable specific layers based on their needs without restructuring the entire system. The layers are evaluated in configurable sequences and can be adapted at runtime, making the system easy to operate for different use cases while avoiding the complexity of permanently maintaining a rigid multi-layer architecture for all scenarios.
Solution Approach 2:
The system implements self-service capabilities where the layering structure automatically adapts to user requirements through configuration files and declarative definitions. Rather than requiring complex manual setup of each layer's interactions, the system automatically manages layer coordination, data flow routing, and evaluation sequencing based on user-specified configurations, thereby improving ease of operation while minimizing the operational complexity of managing multiple layers.
Data Source
AI summary
This patent specification relates to systems and methods that use networks of standardized time series and models in a generic and extensible platform. More particularly, this patent specification relates to standardizing time series and models, using standardized time series and models in a network of time series (NOTS), and creating networks of time series in a NOTS platform. In addition, this patent specification relates to use of nodes that can be arranged in any user defined order, and each node is evaluated to return a time array that may be used to populate a time series. A layering framework may be used to define how each node is evaluated.