AI-Driven RPA Workflow Auto-Completion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic process automation (RPA) technologies require repetitive steps in workflow creation, leading to reduced developer productivity due to the lack of intelligent prediction and adaptation in template-driven workflow designs.
Innovation Solution
The use of AI/ML models to analyze RPA workflow sequences, provide suggestions for next activities, and auto-complete code, reducing repetition by learning from developer patterns and preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If template-driven workflow designs are used, then workflow creation is standardized, but developer productivity decreases due to repetitive steps and lack of intelligent prediction
Solution Approach 1:
The system performs preliminary action by analyzing sequences of activities already present in the workflow and predicting the next required activities before the developer actually needs to create them. The AI/ML model pre-computes suggestions based on the current workflow state, allowing developers to simply select from predicted options rather than creating activities from scratch, thus maintaining standardization while dramatically improving productivity.
Solution Approach 2:
The workflow design system provides self-service by automatically generating activity sequence suggestions based on the workflow context and historical patterns. The AI/ML model serves itself by learning from captured workflow sequences and automatically providing completion suggestions without requiring manual template selection or intervention, enabling the system to assist developers intelligently while they focus on high-level design decisions.
2Adaptability or versatility
If developers manually create each workflow sequence, then customization and control are maximized, but time consumption and effort increase significantly
Solution Approach 1:
The system implements feedback by capturing the sequences of activities that developers create and use this information to train and improve the AI/ML model. The model continuously learns from the feedback provided by actual developer workflows, becoming increasingly accurate at predicting the specific sequences and patterns relevant to each organization's workflows. This allows the system to provide increasingly accurate suggestions that respect organizational standards while dramatically reducing creation time.
Solution Approach 2:
The system applies partial action by providing suggested completions for only the next logical sequence of activities rather than attempting to generate entire workflows. The AI/ML model focuses on predicting immediate next steps based on current context, allowing developers to maintain full control over overall workflow design while automatically handling routine sequence completions, thus balancing customization with time efficiency.
3Stability of the object's composition
If preset templates are used for workflow generation, then consistency is maintained, but the system cannot predict user intent or adapt to changes in requirements
Solution Approach 1:
The system transitions from static preset templates to dynamic AI/ML-based prediction that adapts to user intent and changing requirements. The model analyzes the current workflow context, captured activity sequences, and organizational patterns to dynamically generate suggestions that are consistent with established workflows yet adaptable to specific user needs and evolving requirements, maintaining consistency through learned patterns rather than rigid templates.
Solution Approach 2:
The system changes parameters by moving from fixed template structures to flexible, context-aware predictions. The AI/ML model adjusts its suggestions based on multiple parameters including the current workflow state, historical sequence patterns, organizational standards, and specific user preferences captured through feedback, allowing the system to maintain consistency while adapting to varying user intents and requirements through parameter-driven customization.
Data Source
AI summary
Training and using artificial intelligence (AI)/machine learning (ML) models to automatically supplement and/or complete code of RPA workflows is disclosed. A trained AI/ML model may intelligently and automatically predict and complete the next series of activities in RPA workflows (e.g., one, a few, many, the remainder of the workflow, etc.). Actions users take while creating workflows over a time period may be captured and stored. The AI/ML model may then be trained and used to match the stored actions with stored workflow sequences of actions in order to predict and complete the workflow. As more and more workflow sequences are captured and stored over time, the AI/ML model may be retrained to predict a larger number of sequences and/or to more accurately make predictions. Auto-completion may occur in real-time in some embodiments to save time and effort by the user.