Automatic Calling List Scheduler for Contact Centers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face challenges in determining when to switch calling lists to maximize agent productivity, as existing methods require significant experience and are not flexible or easy to administer, leading to inefficiencies due to poor performance of calling lists.
Innovation Solution
An automated system that learns from administrator actions to determine when to switch calling lists based on measured parameters such as time, call completion rate, and agent utilization, allowing for flexible and accurate decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual switching of calling lists is used, then flexibility and adaptability are maintained, but it requires significant experience and is not easy to administer
Solution Approach 1:
The system performs self-service by automatically monitoring performance parameters and switching calling lists without requiring continuous human intervention or extensive administrative experience. The automated mechanism learns from historical data and makes independent decisions about when to switch lists, reducing the operational burden on administrators.
Solution Approach 2:
The system implements feedback loops where performance parameters (call completion rate, agent utilization, etc.) are continuously monitored and fed back into the system. This feedback enables the automated mechanism to adjust calling list selection dynamically based on actual performance data, improving ease of operation while maintaining high automation levels.
2Ease of operation
If automated switching mechanism is implemented, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The automated mechanism is segmented into distinct functional modules: parameter monitoring module, performance analysis module, decision logic module, and calling list switching module. This segmentation reduces overall system complexity by making each component independent and easier to understand, maintain, and debug, while still providing automated operation.
Solution Approach 2:
The system performs preliminary actions by pre-defining performance thresholds, historical performance data, and switching criteria before actual operation begins. This preparation simplifies the automated decision-making process during execution, as the system only needs to compare current performance against pre-established criteria rather than making complex real-time decisions.
3Measurement precision
If calling list switching is based on historical data and learned patterns, then accuracy of switching decisions is improved, but loss of time is required for learning period
Solution Approach 1:
The system performs preliminary learning during off-peak times or using historical data from previous periods to establish performance patterns and thresholds before actual automated switching begins. This preliminary action allows the system to accumulate accurate decision-making models without impacting real-time operation, reducing the effective learning time loss.
Solution Approach 2:
The system uses partial learning by focusing on the most critical performance parameters and switching criteria rather than analyzing every possible variable. This partial action approach accelerates the learning process by concentrating resources on high-impact factors, reducing the time required to achieve accurate switching decisions while maintaining sufficient precision.
4Measurement precision
If multiple performance parameters are monitored, then accuracy of calling list selection is improved, but device complexity and measurement difficulty increase
Solution Approach 1:
The system merges multiple performance parameter monitoring functions into a unified monitoring framework that handles call completion rate, agent utilization, and other metrics through a single integrated process. This consolidation reduces the complexity of detecting and measuring parameters by using shared data collection, processing, and analysis mechanisms rather than separate systems for each parameter.
Data Source
AI summary
An automatic list scheduling (“ALS”) system monitors usage of calling lists by an administrator where the administrator terminates usage of a calling list based on various measured parameters. The ALS system is configure to learn the conditions associated with the measured parameters and, in one embodiment, recommends to the administrator when to terminate use of a current calling list in favor of another list. A learning mode gathers various samples of the measured parameters associated with calling lists, and analyzes the sample to determine a threshold value. Upon monitoring the measured values for subsequent usage of calling lists, the current measured parameters are compared to the threshold value to determine whether to generate a recommendation to the administrator. In one embodiment, the time zone of the called party and the applicable calling window is used to determine whether the current list can continued to be used.