AI Data Transformation for Process Automation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing automated systems for knowledge work struggle to efficiently convert voluminous process manuals into executable code, relying on manual programming and domain expert knowledge, which can be incomplete or outdated, leading to errors in process automation.

Innovation Solution

An AI-based data transformation system that processes documents of various formats, performs structural and semantic analysis, and generates platform-specific code to automate processes, allowing for iterative feedback and human validation to correct errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual programming and domain expert knowledge are used to convert process manuals into executable code, then the conversion can be performed with existing tools, but the process becomes time-consuming and error-prone due to incomplete or outdated domain knowledge

Engineering Contradiction:

Improveaccuracy of process automationVSAvoidtime for converting process manuals to code

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual programming mechanics with an AI-based natural language processing system that automatically converts process manual text into executable code. The AI model analyzes the semantic structure of process descriptions and generates corresponding automation code without requiring manual intervention, thereby eliminating the time-consuming nature of manual conversion while maintaining or improving accuracy through the AI's ability to understand and interpret process logic.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If domain experts manually create process automation code, then the code can be customized to specific needs, but the process requires significant expert knowledge and is difficult to maintain when processes change

Engineering Contradiction:

Improvecustomization of process automationVSAvoidcomplexity of maintaining process automation

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The AI-based system enables process automation to be self-updating by automatically re-analyzing process manuals when changes occur. The system can detect modifications in process descriptions and regenerate the corresponding automation code without requiring expert intervention, thereby maintaining high adaptability to process changes while significantly reducing the complexity of maintenance.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent creates a universal AI-based conversion system that can handle multiple types of process manuals and generate code for various automation platforms. This single system replaces the need for multiple specialized tools and expert knowledge areas, providing both customization capability and reduced maintenance complexity through its multi-functional nature.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Loss of information

If voluminous process manuals are processed manually, then detailed knowledge can be extracted, but the process becomes inefficient and cannot keep up with rapidly changing processes

Engineering Contradiction:

Improvecompleteness of process knowledge extractionVSAvoidspeed of process automation conversion

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The patent replaces manual information extraction with an AI-based natural language processing system that can rapidly analyze voluminous process manuals. The AI model uses semantic analysis and machine learning to extract complete process knowledge at high speed, maintaining information completeness while increasing productivity by orders of magnitude compared to manual extraction methods.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary analysis of process manuals by pre-processing and structuring the information before automation code generation. This preliminary action allows the AI to efficiently organize vast amounts of process knowledge, ensuring complete extraction while enabling rapid subsequent conversion to automation code, thereby resolving the contradiction between information completeness and conversion speed.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10936363B2Automatic data transformations for process automations

Publication Date: 2021.03.02 ACCENTURE GLOBAL SOLUTIONS LTD

AI summary

An Artificial Intelligence (AI) based data transformation system receives a process document and automatically generates processor-executable code which enables automatic execution of a process as detailed within the process document. Various structural elements of the process documents are identified and the data from the document is clustered based on common parameters which can include the structural elements or textual data from the process document. The contextual information including conditional and non-conditional statements along with the entities and entity attributes are also obtained. The domain knowledge is superimposed on the contextual information to generate flows that represent procedures which make up the process to be automated. Platform specific code for the automatic execution of the process is automatically generated from the flows.