Automated Workflow Evaluation System Using Network Graph Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional process workflow improvements require time-intensive, multi-person communication and expertise, with conventional comparison methods being painstaking and reliant on subject-matter experts to identify similarities between intricate workflows.

Innovation Solution

An automated process evaluation computing system that uses isomorphism-based, computer vision-based, and knowledge-based network graph comparisons to evaluate and cluster workflow structures, reducing evaluation time and providing recommended improvements such as consolidation, standardization, and simplification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional manual methods are used to compare and evaluate process workflows, then evaluation accuracy and depth can be achieved through expert analysis, but the process requires time-intensive multi-person communication and subject-matter expertise

Engineering Contradiction:

Improveworkflow evaluation accuracyVSAvoidevaluation time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual expert analysis with automated computing systems that use machine learning models, natural language processing, and computer vision to evaluate process workflows. The system automatically compares workflow structures, extracts insights, and generates recommendations without requiring human experts to manually analyze each workflow, thereby eliminating time-intensive communication while maintaining evaluation accuracy through algorithmic precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates digital representations and models of process workflows that can be replicated and analyzed computationally. By converting workflows into structured data formats and visual models, the system enables automated comparison and evaluation across multiple workflows simultaneously, eliminating the need for repeated manual analysis while preserving the detailed insights that experts would otherwise provide.

Inventive Principle:

Principle #26Copying

2Measurement precision

If conventional comparison methods are used to identify similarities between workflows, then thorough analysis can be achieved, but the process is painstaking and requires subject-matter experts

Engineering Contradiction:

Improvesimilarity detection accuracyVSAvoidevaluation system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex manual comparison methodologies with automated computing systems that use machine learning algorithms, natural language processing, and computer vision techniques. These systems automatically detect similarities between workflows by analyzing structured data and visual representations, eliminating the need for painstaking expert analysis while maintaining high detection accuracy through sophisticated pattern recognition capabilities.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a universal evaluation system that can analyze multiple types of workflows across different domains using the same automated techniques. The system handles diverse workflow structures and formats through standardized processing pipelines, reducing the need for domain-specific expert knowledge while maintaining accurate similarity detection across various business processes and organizational contexts.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Productivity

If automated evaluation systems are implemented to reduce evaluation time, then turnaround time improves, but the system requires sophisticated technological approaches and automation capabilities

Engineering Contradiction:

Improveevaluation throughputVSAvoidsystem technological complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements automated computing systems that use machine learning models, natural language processing, and computer vision to evaluate process workflows at scale. These systems process multiple workflows simultaneously through automated comparison algorithms and generate evaluation reports without human intervention, dramatically increasing evaluation throughput while managing technological complexity through standardized processing pipelines and modular system architecture.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11847599B1Computing system for automated evaluation of process workflows

Publication Date: 2023.12.19 WELLS FARGO BANK NA

AI summary

An example method includes obtaining data associated with a plurality of process workflows, performing an automated workflow analysis of the data at least by (i) evaluating workflow structures to identify at least one similarity between one or more portions of the process workflows, wherein the workflow structures are modeled from the process workflow data and, (ii) responsive to evaluating the workflow structures, clustering, based on the at least one similarity, the portions of the process workflows into at least one common process group that is shared between the process workflows, identifying, based on the at least one common process group shared between the process workflows, at least one process improvement that is associated with at least one of the process workflows, and outputting, by the computing system, a recommendation associated with the at least one process improvement.