ML-Based Resource Allocation for Digital Contact Centers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for resource allocation in digital contact centers are inadequate in handling multiple concurrent communications across various channels, as they rely on outdated methods like the Erlang C formula and simulations that fail to capture the complexity of digital mediums and diverse user behaviors.
Innovation Solution
A computerized system using machine learning models, specifically deep learning neural networks, to predict service metrics and optimize staffing requirements by transforming initial allocation matrices into updated matrices that account for multi-functional resources capable of handling multiple tasks simultaneously.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Erlang C formula and simulation methods are used for resource allocation, then staffing requirements can be approximated, but the complexity of digital mediums and diverse user behaviors cannot be captured
Solution Approach 1:
The patent replaces traditional mechanical/mathematical systems (Erlang C formula, discrete event simulations) with a machine learning-based predictive system. The ML model learns complex patterns from historical data across multiple digital channels without requiring explicit mathematical formulations, thereby capturing the complexity of digital mediums while improving prediction accuracy.
Solution Approach 2:
The patent transforms the resource allocation approach by changing from fixed mathematical parameters (AHT, concurrency) to dynamic ML-predicted service metrics. The system uses ML models to predict service metrics under different allocation scenarios, allowing the system to adapt to varying digital channel complexities and user behaviors.
2Productivity
If agents are allocated to handle multiple concurrent contacts across digital channels, then resource utilization improves, but service quality may deteriorate due to divided attention
Solution Approach 1:
The patent implements dynamic resource allocation that adjusts staffing levels and allocations based on predicted service metrics. The system evaluates multiple allocation candidates with different concurrency levels and selects optimal allocations that balance resource utilization with service quality requirements, adapting to changing digital channel demands.
Solution Approach 2:
The system uses ML predictions of service metrics as feedback to iteratively refine resource allocation decisions. By comparing predicted service metrics against required thresholds, the system adjusts allocations to maintain service quality while optimizing resource utilization across digital channels.
3Ease of operation
If average handling time is used to approximate service level, then staffing requirements can be calculated, but the diversity of communication channels and user behaviors is not accounted for
Solution Approach 1:
The patent creates a universal ML-based prediction system that handles multiple digital channels (chat, email, SMS, social media) and diverse user behaviors through a single framework. The ML model learns patterns across different channels without requiring channel-specific calculations, providing both simplicity and adaptability.
Solution Approach 2:
The system replaces simple AHT-based calculations with ML predictions that automatically adapt to channel diversity. The ML model captures complex relationships between channel types, user behaviors, and service metrics without requiring manual adjustments or separate calculation methods for each channel.
Data Source
AI summary
A computerized system and method for allocating multi-functional or multi-feature resources (which may handle multiple functions or tasks, e.g., simultaneously) for a plurality of time intervals, including: transforming an initial allocation matrix (which may associate each resource with a single function, task, or feature - and may not address simultaneous handling of tasks or task types by the resources) into an updated allocation matrix, where the updated allocation matrix includes a plurality of feature matrices describing different multi-feature resources to be allocated; predicting, using a machine learning (ML) model, expected service metrics for the updated allocation matrix; and providing a final allocation matrix based on the expected service metrics. Embodiments may perform iterative calculations and/or transformations of data to improve allocation matrices and provide a final allocation matrix for which predicted service metrics correspond to required or optimal service metrics.