Contact Center Resource Scheduling via Attribute Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face challenges in accurately predicting staffing needs due to fluctuations in contact types and attributes, leading to inefficiencies in resource scheduling, resulting in either excessive or insufficient resources, which impacts service levels and customer experience.
Innovation Solution
The method involves assigning resources based on proficiency levels in specific attributes and identifying correlations between these attributes to forecast the minimum number of resources required to meet service levels, allowing for more efficient scheduling and reduced resource surplus or shortage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If attribute-based contact centers increase granularity for accurate forecasting, then forecasting accuracy is improved, but resource scheduling flexibility deteriorates due to fixed shift constraints
Solution Approach 1:
The patent applies dynamics by making the resource roster dynamically adjustable within shifts. Resources can be reassigned between different contact type groups during a shift based on real-time demand fluctuations, transforming the traditionally static shift-based scheduling into a flexible system that adapts to changing contact patterns while maintaining shift structure.
Solution Approach 2:
The patent segments the resource roster into multiple contact type groups, where each group is specialized for handling specific contact types. This segmentation allows independent adjustment of resources within each group during a shift, enabling precise matching of resources to forecasted contact patterns without requiring complete roster restructuring.
2Reliability
If resource roster is dynamically adjusted to match predicted contact patterns, then service level is improved, but operational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-segmenting the resource roster into contact type groups before shifts begin and pre-forecasting contact patterns. This advance preparation reduces the complexity of real-time adjustments during shifts, as the framework for dynamic reassignment is already established and resources are pre-positioned according to predicted demand.
Solution Approach 2:
The patent implements feedback mechanisms that monitor actual contact patterns during shifts and use this information to guide resource reassignment decisions. This closed-loop feedback system automates the adjustment process, reducing operational complexity by using data-driven rules rather than manual judgment for each reassignment decision.
3Reliability
If more resources are scheduled to handle contact fluctuations, then service level is improved, but labor cost increases
Solution Approach 1:
The patent applies universality by training resources to handle multiple contact types within their contact type group, making them multi-functional. Resources can be reassigned between different contact types based on demand fluctuations without requiring additional specialized hiring, thereby maintaining service levels across varying contact patterns while controlling labor costs through existing versatile workforce.
Data Source
AI summary
Managing resources in a contact center including assigning each resource to one of a first set of resources each comprising a proficiency level above a first threshold for a first resource attribute, or a second set of resources each comprising a proficiency level below the first threshold for the first resource attribute and a proficiency level above a second threshold for a second resource attribute. An expected number of contacts requiring resources possessing one of the first or second resource attribute is predicted for a time period, and a correlation between the first and second resource attribute is identified. Based on the correlation, a minimum number of resources from each set required to handle the expected number of contacts at a predetermined service level for the time period is forecasted. The minimum number of resources from the first set is less than a number of resources required without the correlation.