Agent Group Query Routing With Parallel Distribution Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional call distribution methods in customer interaction centers, such as simultaneous, sequential, and rotating ringing, often result in increased wait times due to understaffing or remote agents not actively logged in, leading to inefficient query handling and extended wait periods for callers.
Innovation Solution
Implementing parallel processing and skip criteria for query distribution within agent groups, where queries are distributed based on a defined threshold number of available agents and excluding agents that fail to meet skip criteria, such as those with locked devices or external phone numbers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional call distribution methods (simultaneous, sequential, rotating ringing) are used, then queries are distributed to agents, but wait times increase due to understaffing or remote agents not actively logged in
Solution Approach 1:
The system dynamically adjusts query distribution by monitoring real-time agent availability status. Instead of static distribution methods, the system continuously evaluates which agents are actively logged in and capable of handling queries, adapting the distribution pattern in real-time to match current staffing conditions and minimize caller wait times.
Solution Approach 2:
The system automatically identifies and excludes unavailable agents from the distribution pool without manual intervention. By self-managing the agent availability assessment and dynamically adjusting distribution parameters, the system eliminates the need for manual agent status tracking and ensures queries are only routed to actively available agents.
2Productivity
If all agents in an agent group are included in query distribution, then more queries can be handled, but agents with locked devices or external phone numbers cannot receive queries
Solution Approach 1:
The system applies different distribution rules to different subsets of agents based on their specific characteristics. Instead of treating all agents uniformly, it identifies agents with locked devices or external phone numbers and applies exclusion criteria specifically to them, while maintaining inclusion of agents with valid, accessible contact information.
Solution Approach 2:
The system extracts and excludes agents who fail to meet skip criteria (locked devices, external phone numbers) from the query distribution pool. By removing these unavailable agents from consideration, the system ensures that queries are only distributed to agents who can actually receive and handle them, improving delivery success rate.
3Loss of time
If parallel processing is implemented for query distribution, then wait times are reduced, but system complexity increases
Solution Approach 1:
The system implements parallel processing selectively rather than universally. It evaluates the queue depth and agent availability to determine when parallel distribution is appropriate, applying it only when conditions warrant the additional processing power, thereby balancing performance improvement against system complexity.
Solution Approach 2:
The system uses feedback from agent availability monitoring and query queue status to dynamically control the degree of parallel processing applied. By continuously monitoring system state and adjusting distribution strategies accordingly, the system optimizes wait time reduction while avoiding unnecessary complexity in normal operating conditions.
Data Source
AI summary
The distribution of incoming queries to a customer interaction center agent group is parallel processed amongst agents of that group to improve queue wait times. A threshold number of queries that may be processed by agent devices associated with the agent group at a given time are defined based on a number of agents of the agent group that are available at the given time. In response to determining that the number of queries is satisfies the threshold number of queries based on the number of agents that are available at a current time, a number of queries awaiting processing are distributed to one or more agent devices of the agent group. The threshold number of queries may be based on half of the number of agents that are available at the given time.