Modeling Engine Structuring Unstructured BPMN into Process Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional modeling techniques struggle with structuring complex, unstructured, and cyclic business process models, often duplicating tasks and mishandling synchronization conditions, which complicates critical analysis and proper definition of process steps.
Innovation Solution
A modeling engine translates unstructured existing processes into well-structured process trees by detecting and structuring rigid components, using AI/ML capabilities to import and translate unstructured process definitions, and adding supplement, correction, and synchronization edges to create usable data structures like patched process trees.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional graphical modeling notation is used to define existing processes, then the processes can be visually depicted with over fifty model elements, but the model becomes unable to properly handle complex structures, cyclic models, and synchronization conditions
Solution Approach 1:
The patent segments the complex BPMN model into a hierarchical structure consisting of process trees and rigid components. By dividing the model into manageable segments (process trees for structured parts and rigids for unstructured parts), the system can reliably process each segment separately while maintaining overall model fidelity, thus resolving the contradiction between handling diverse structures and ensuring accurate process definition.
Solution Approach 2:
The patent introduces an intermediary representation layer between the original BPMN model and the final process analysis. This intermediary uses process trees and rigid components as a bridge, allowing the system to handle both structured and unstructured process elements while maintaining reliability in process step definition through the structured intermediary representation.
2Ease of manufacture
If conventional models attempt to structure unstructured processes, then they can provide a graphical notation framework, but they duplicate tasks and mishandle synchronization conditions
Solution Approach 1:
The patent implements a dynamic structuring approach where the system automatically identifies and separates structured portions (convertible to process trees) from unstructured portions (rigids). This dynamic classification allows the model to adapt its structuring strategy based on the specific characteristics of each process segment, maintaining both ease of model creation and precision in process structuring.
Solution Approach 2:
The patent changes the structural parameters of the process model by transforming it from a purely graphical notation into a hybrid representation with process trees and rigids. This parameter change allows the system to maintain ease of graphical model creation while achieving precision in process structuring through the tree-based hierarchical organization of structured elements.
3Adaptability or versatility
If conventional models use rigid components for parallelism, then they can represent unstructured processes, but they create duplications and fail to properly synchronize process conditions
Solution Approach 1:
The patent segments the model into process trees and rigids, where process trees represent structured, non-duplicative process flows and rigids represent unstructured portions. This segmentation eliminates task duplication by ensuring that each process element is represented exactly once in the appropriate structure, preventing information loss while maintaining the ability to represent both structured and unstructured processes.
Solution Approach 2:
The patent uses selective copying where structured process elements are transformed into process tree nodes rather than being duplicated as rigid components. This copying strategy preserves process information without duplication by creating a hierarchical representation in the process tree that maintains the original process logic while eliminating redundant representations.
Data Source
AI summary
A method is provided. The method is executed by a modeling engine implemented as a computer program within a computing environment. The method includes translating a pre-processed model based on an unstructured existing process into a patched process tree. The translating includes detecting, structuring, and translating rigids of the pre-processed model into corresponding process tree equivalent elements of the patched process tree.


