AI Model of Models for RPA Workflow Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic process automation (RPA) systems face challenges in adapting to changes and optimizing performance, as they often require continuous rule updates and may not be suitable for all scenarios, leading to potential failures in recognizing new changes or inefficiencies.
Innovation Solution
The implementation of an AI-driven 'model of models' that analyzes and selects or chains machine learning (ML) models within an RPA workflow, replacing existing models with superior performing ones and modifying the workflow to enhance performance by experimenting with permutations of ML models in series, parallel, or combinations thereof.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single RPA workflow is built and deployed in a single robot, then the robot can perform certain tasks under certain conditions, but the robot may fail to recognize new changes or fail altogether when changes occur
Solution Approach 1:
The patent implements dynamic model selection where the RPA system automatically selects and switches between multiple ML models based on current performance metrics and changing conditions. This allows the system to adapt to new scenarios dynamically rather than relying on a static single model, resolving the contradiction between reliability and adaptability.
Solution Approach 2:
The system changes the parameter of model selection by evaluating multiple ML models and selecting the one with superior performance for specific scenarios. This parameter change enables the system to maintain reliability while adapting to different conditions by switching between models with different strengths.
2Device complexity
If a single robot is used for automation, then the system structure is simple, but the robot is not optimal for all scenarios
Solution Approach 1:
The patent implements a pool of multiple ML models where each model can handle different scenarios or task types. This multi-functionality approach allows a single RPA robot to efficiently handle diverse scenarios by selecting the appropriate model, improving productivity without requiring multiple specialized robots.
Solution Approach 2:
The system nests multiple ML models within a single RPA robot workflow, creating a hierarchical structure where the robot contains a collection of specialized models. This nesting allows the simple system structure to be maintained while incorporating multiple functional capabilities for different scenarios.
3Adaptability or versatility
If continuous rule updates are made to improve robot performance, then the robot can adapt to changes, but the system requires continuous maintenance and rule management
Solution Approach 1:
The patent implements self-service through automated model selection and performance evaluation. The system automatically monitors which ML models perform best in different scenarios and selects them without requiring manual rule updates. This reduces the complexity of rule management while maintaining high adaptability to changes.
Solution Approach 2:
The system incorporates feedback mechanisms where performance metrics from ML model executions are continuously monitored and used to automatically update model selection strategies. This feedback loop enables the system to adapt to changes automatically without manual intervention, reducing rule management complexity while improving adaptability.
Data Source
AI summary
Using artificial intelligence (AI) to select and/or chain robotic process automation (RPA) models a given problem is disclosed. A model of models (e.g., an RPA robot or an ML model) may serve as an additional layer on an existing system that makes the existing models more effective. This model of models may incorporate AI that learns an improved or best set of rules or an order from existing models, potentially taking certain activities from a model, feeding input from one model into another, and/or chaining models in some embodiments.