Object-Centric Data Model Scope Attribute for Process Mining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional process mining data models lack the ability to exploit hierarchical structures, preventing the use of OLAP operations like drill-down and roll-up due to unstructured information, which limits aggregation and analysis capabilities.
Innovation Solution
A computer-implemented method generates an object-centric data model with a scope attribute that serves as a pointer to an emerging hierarchical structure, enabling OLAP operations and allowing for dynamic aggregation levels by storing scope values that represent 'part-of' relations, facilitating anonymization, filtering, and clustering of process data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional process mining data models store unstructured information in attributes, then the data model is simple to implement, but OLAP operations like drill-down and roll-up cannot be performed
Solution Approach 1:
The patent segments the data model into structured components by introducing a scope attribute that divides unstructured attribute data into hierarchical segments. Each scope value represents a segment of the hierarchy (e.g., department, team, resource), enabling OLAP operations to traverse and aggregate data across these segmented levels while maintaining implementability.
Solution Approach 2:
The patent adds a new dimension to the data model by introducing the scope attribute as a separate hierarchical structure. Instead of storing all information in unstructured attributes, the scope dimension provides a structured layer that enables multi-dimensional analysis and OLAP operations without complicating the core data storage mechanism.
2Adaptability or versatility
If events are stored at fixed predefined aggregation levels, then data storage is straightforward, but dynamic drill-down and roll-up operations are not possible
Solution Approach 1:
The patent makes the aggregation level dynamic by introducing scope values that can represent different hierarchical levels (e.g., company, department, team, resource). Users can dynamically drill down or roll up by navigating through these scope levels, transforming the fixed aggregation model into a flexible, adaptive structure that responds to analytical needs.
Solution Approach 2:
The patent performs preliminary action by pre-structuring the scope attribute with potential hierarchical relationships before analysis. The scope values are organized in advance to represent the hierarchy (e.g., parent-child relationships between departments and teams), enabling rapid drill-down and roll-up operations without requiring complex real-time computations.
3Loss of information
If unstructured attribute information is used, then data storage is simple, but causal relationships between events cannot be identified
Solution Approach 1:
The patent introduces the scope attribute as an intermediary between unstructured event data and causal relationship analysis. The scope values act as mediators that link events to their contextual hierarchy (e.g., which department or resource an event belongs to), enabling the identification of causal relationships without requiring complex restructuring of the underlying data storage.
4Adaptability or versatility
If resource information is kept at detailed level, then analysis precision is high, but anonymization and abstraction are difficult
Solution Approach 1:
The patent makes the level of resource information detail dynamic through the scope attribute. Users can anonymously aggregate data at higher scope levels (e.g., department level) when anonymization is needed, while still accessing detailed resource information when precision is required. This dynamic switching capability enables both anonymization and precise analysis using the same data structure.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is a computer-implemented method to generate an object-centric data model of recorded process data, wherein objects and events are enriched by scopes, i.e., scope values stored in at least one scope attribute. Each scope value refers to a node of an emerging hierarchical structure, such that operations on the aggregation level of the objects and events, e.g., drill-downs and roll-ups, are enabled.