Dynamic Contact Center Routing with Real-Time Agent Telemetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional contact center routing systems rely on static, pre-configured rules that do not consider various factors that could improve customer experience, resource utilization, and cost efficiency.
Innovation Solution
The system incorporates data such as agent device telemetry, local conditions, monetary cost data, and noise data into routing decisions to optimize agent selection and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If static, pre-configured routing rules are used, then system simplicity is maintained, but routing optimization and resource utilization deteriorate
Solution Approach 1:
The patent transforms static routing rules into dynamic routing decisions by continuously collecting real-time data from multiple sources (agent telemetry, environmental sensors, network conditions) and using machine learning models to adapt routing rules on-the-fly. This allows the system to optimize resource utilization without requiring complete reconfiguration, as the system dynamically adjusts to current conditions while maintaining operational simplicity.
Solution Approach 2:
The system implements closed-loop feedback by collecting performance data from agents and environmental conditions, analyzing this data through machine learning models, and using the insights to continuously refine routing decisions. This feedback mechanism enables the system to learn from past performance and automatically optimize routing without manual intervention, resolving the contradiction between system simplicity and optimization capability.
2Productivity
If multiple data sources are collected and analyzed, then routing optimization improves, but system complexity and processing requirements worsen
Solution Approach 1:
The patent introduces machine learning models as intermediary components that automatically process and synthesize data from multiple sources (agent telemetry, environmental sensors, network conditions). These models act as intelligent mediators that transform raw multi-source data into actionable routing insights, reducing the complexity burden on the routing system itself while maintaining high optimization capability through automated pattern recognition and decision support.
3Measurement precision
If real-time data processing is implemented, then routing accuracy improves, but processing time and computational cost worsen
Solution Approach 1:
The system performs preliminary actions by pre-processing and preparing routing data in advance, maintaining ready-to-use feature sets and model predictions for common scenarios. This allows the system to quickly make accurate routing decisions during peak periods without requiring extensive real-time computation, as much of the analytical work is completed beforehand or cached for rapid retrieval.
Data Source
AI summary
A contact center request is received at a customer datacenter. Respective factors for available agents including agent-specific factors, environmental factors, and agent-cost factors are determined. Respective weights for the respective factors are obtained. At least one agent of the available agents is selected based on the respective weights. The contact center request is then routed to the at least one agent. The weights can be obtained via a user interface or can be obtained from a machine-learning model that is trained to output the weights.