Workload Routing With Dynamic Agent Capacity Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication systems fail to dynamically adjust agent capacity assignments based on the varying complexity and difficulty of user interactions, leading to inefficiencies and potential overload in workload distribution.

Innovation Solution

Implementing an intelligence-driven workload model that analyzes multiple factors to predict and distribute communications among agents, dynamically adjusting slot assignments to match the identified workload measurement, thereby optimizing resource use and preventing system imbalances.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If agents are assigned a predetermined number of slots to handle multiple user connections simultaneously, then agent productivity is improved, but the system cannot distinguish between different levels of conversation complexity leading to workload imbalance

Engineering Contradiction:
Improveagent productivityVSAvoidworkload measurement precision
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies dynamics by transitioning from static predetermined slot assignments to dynamic slot adjustments based on real-time workload measurements. The system continuously monitors conversation complexity and adjusts the number of active slots for each agent accordingly, allowing the system to adapt to varying workload conditions while maintaining high productivity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of slot allocation from fixed to variable. Instead of assigning a predetermined number of slots to each agent, the system dynamically modifies the slot count based on measured workload parameters such as conversation complexity, duration, and resource consumption, thereby achieving precise workload distribution.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If predetermined slot fractions are added or subtracted for simple or difficult conversations, then some workload adjustment is achieved, but the system still cannot identify predicted workload with precision

Engineering Contradiction:
Improveworkload adaptationVSAvoidworkload prediction precision
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements feedback mechanisms where the system continuously monitors actual conversation characteristics and compares them against predefined thresholds. Based on this feedback, the system adjusts slot assignments in real-time, creating a closed-loop system that progressively improves workload prediction accuracy and adaptation.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis of conversation characteristics before finalizing slot assignments. By evaluating factors such as conversation type, complexity indicators, and expected duration in advance, the system pre-adjusts slot allocations to match the predicted workload, improving both adaptability and precision.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If agents handle multiple user connections simultaneously to maximize efficiency, then system productivity increases, but agents may be overwhelmed by complex conversations requiring more time and analysis

Engineering Contradiction:
Improvesystem productivityVSAvoidagent capacity reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts the number of active slots for each agent based on real-time workload assessment. When an agent is assigned a complex conversation requiring additional time and analysis, the system automatically reduces the number of active slots or reallocates resources, preventing agent overload while maintaining high system productivity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system continuously monitors agent performance and conversation complexity, using this feedback to adjust slot assignments. When an agent demonstrates signs of being overwhelmed or when a conversation exceeds expected complexity thresholds, the system responds by modifying the agent's capacity, thereby ensuring reliability while preserving productivity.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3959675B1Smart capacity for workload routing
Publication Date: 2025.11.19 LIVEPERSON INC
  • EP3959675B1 patent drawingFigure 1
  • EP3959675B1 patent drawingFigure 2
  • EP3959675B1 patent drawingFigure 3

AI summary

Examples are described of smart capacity workload routing. One example involves storing a workload model in memory regarding a set of different factors associated with user communications, with each factor is associated with a measurement of workload. A received request including information regarding one or more of the factors is process and used in identifying a workload measurement for the requested user communication based on comparing the received request information to the stored workload model, and identifying an agent with capacity that is available to handle the requested user communication. A communication slot for the identified agent is activated and defined by the identified workload measurement. The request is then routed to the identified agent and updating available workload capacity in the system.