Specialized Microbots for Contact Center Virtual Agents
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monolithic virtual agents in contact centers are expensive to create and maintain, complex to troubleshoot, and inefficient due to their attempt to perform all functions, leading to high resource usage and difficulty in scaling.
Innovation Solution
Implementing specialized virtual agents with modular designs that can be easily created, tested, and maintained, each handling specific interactions and leveraging public cloud services, allowing for cost-effective deployment and efficient resource utilization by leveraging economies of scale.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If monolithic virtual agents are used to perform all functions, then versatility is improved, but device complexity and ease of manufacture deteriorate
Solution Approach 1:
The patent divides the monolithic virtual agent into multiple specialized microbots, each handling specific tasks or domains. This segmentation reduces code complexity by creating modular, independent components while maintaining overall system versatility through their collective capabilities.
Solution Approach 2:
The patent creates a universal agent framework that can invoke multiple specialized microbots. This universal layer provides the multi-functionality needed to handle diverse customer interactions while delegating specific functions to specialized agents, avoiding the complexity of a single monolithic agent trying to do everything.
2Adaptability or versatility
If monolithic virtual agents attempt to cover all functions, then adaptability is improved, but ease of repair and manufacturing precision deteriorate
Solution Approach 1:
By segmenting the agent system into independent microbots, each with specific functionality, the patent enables easier maintenance and repair. Individual microbots can be updated, fixed, or replaced without affecting the entire system, unlike a monolithic agent where changes risk introducing system-wide errors.
Solution Approach 2:
The patent implements self-service mechanisms where the universal agent automatically selects and invokes appropriate microbots based on customer needs. This automated routing reduces the need for manual intervention in managing functional coverage while maintaining adaptability.
3Adaptability or versatility
If monolithic virtual agents are deployed with all functionality available, then adaptability is improved, but use of energy and productivity deteriorate
Solution Approach 1:
The patent segments computing resources by deploying only the specific microbots needed for each customer interaction. This avoids the overhead of maintaining and executing all possible functions simultaneously, reducing CPU, memory, and data storage usage while keeping all functionalities available on-demand.
Solution Approach 2:
The patent applies partial action by activating only the necessary microbots for each specific customer query rather than deploying all functionalities. This selective activation reduces resource consumption while maintaining the ability to provide comprehensive service when needed.
4Ease of manufacture
If specialized virtual agents are used for specific tasks, then ease of manufacture is improved, but device complexity in terms of system architecture worsens
Solution Approach 1:
The patent introduces a universal agent as an intermediary layer that manages the complexity of coordinating multiple specialized microbots. This mediator simplifies the system architecture by providing a single point of entry for customer interactions while handling the complexity of selecting and managing specialized agents behind the scenes.
Data Source
AI summary
Large monolithic virtual agents attempt to provide all skills and abilities in a contact center in order to resolve a work item (call) from a customer. However, as the issues become more complex this paradigm results in errors, inefficiencies and knowledge shortcomings on the information and data they have been trained with. Accordingly, systems and methods are provided that enable a number of individual specialized virtual agents, working together, to be deployed. Each has a finite set of abilities and a communication may be transferred to the particular specialized virtual agent having the ability required to resolve a work item. A discovery service is provided to enable an agent to identify particular specialized virtual agent(s) having the required ability in order to resolve the work item, part of the work item or multiple work items.