Intelligent Network Planning Tool Using Feedback-Driven Link Costs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network planning systems lack awareness of actual network demands, leading to inaccuracies in demand forecasting and resource allocation, as they rely solely on available resources and marketing inputs without considering real-time usage patterns.
Innovation Solution
An intelligent network planning and provisioning tool that utilizes recent demand patterns to generate link costs and constraints, incorporating feedback from actual provisioning experiences to refine routing decisions and resource allocation, ensuring a more accurate match between demand forecasts and network resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If external planning systems rely on demand forecasts from marketing personnel, then network planning can be performed before actual network deployment, but the demand forecasts miss fine details regarding actual network usage
Solution Approach 1:
The system implements a feedback mechanism where actual network provisioning data and usage patterns are continuously collected and fed back to the planning system. This feedback loop allows the planning system to refine and update demand forecasts based on real-world network performance and actual resource utilization, thereby improving forecast accuracy over time while maintaining the ability to perform planning before deployment.
2Device complexity
If planning systems use only available network resources information, then the system complexity remains low, but there is no true match between demand forecasts and actual resources
Solution Approach 1:
The system merges multiple data sources including network resource inventory, actual provisioning data, usage patterns, and demand forecasts into a unified planning model. By combining these previously separate information streams, the system achieves accurate resource-demand matching without requiring complex separate systems, as the integration itself provides the matching capability.
3Measurement precision
If the system incorporates actual provisioning experience and usage patterns, then demand forecast accuracy improves, but the system complexity increases
Solution Approach 1:
The system implements self-service capabilities where the planning system automatically collects, processes, and analyzes actual provisioning data and usage patterns without requiring manual intervention. The system self-updates its demand forecasts based on learned patterns from historical data, reducing the operational complexity despite the increased analytical sophistication.
Data Source
AI summary
An intelligent network planning and provisioning tool is provided. The intelligent network planning and provisioning tool includes a forecaster coupled to a network control plane. The forecaster receives an estimate of initial network resources from a user and generates a set of link costs and constraints for use by the network control plane in making routing decisions for a network. The network control plane transmits provisioning experience information, representing network traffic demands received by, and provisioning decisions made by, the network control plane, back to the forecaster. When the forecaster receives the provisioning experience from the network control plane, the forecaster generates a new set of link costs that are transmitted to the network control plane for use in making further provisioning decisions. The forecaster may optionally generate link costs, termed “smart costs”, that may be used to route traffic through the network using preferred links.