Interactive Multi-Object Process Graphs for Multidimensional Data Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing process modeling notations fail to effectively visualize and analyze real-world processes with multiple interacting entities, as they reduce multidimensional data to a linear representation, making it difficult to analyze details such as involved order items, shipments, and invoices.

Innovation Solution

A computer-implemented method creates a process graph that represents multidimensional process data by linking entity instances and event instances, allowing for the visualization of entity flows and relationships, and enabling direct analysis of interrelated entity types through a graph layout that maintains multidimensionality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If existing process modeling notations (BPMN, direct-follow-graph, event-driven process chain) are used to visualize processes, then the process flow can be represented in a linear dimension, but the multidimensional reality of multiple interacting entities (orders, invoices, shipments) cannot be captured

Engineering Contradiction:
ImproveVisual analysis of process behaviorVSAvoidMultidimensional entity relationships
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent transforms the traditional linear process view into a multidimensional entity-centric graph model. Instead of representing processes as sequential steps in one dimension, the invention creates a graph structure where entities (orders, invoices, shipments) are nodes and their relationships are edges, adding dimensional depth to capture multiple interacting entities simultaneously. This allows visualization of complex entity relationships that cannot be represented in linear process notations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Loss of information

If process data is transformed into event logs or process protocols with case identifiers, then the process can be visualized from a case point of view, but analyzing details of involved entities (order items, shipments, invoices) becomes very hard or visually impossible

Engineering Contradiction:
ImproveProcess traceabilityVSAvoidEntity detail analysis
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the process view into two complementary layers: an entity-level graph for detailed entity relationship analysis and a case-level view for overall process traceability. Each entity (order, invoice, shipment) is represented as a separate node with its own event history, allowing detailed analysis of entity-specific details while maintaining the ability to trace overall case progression through parent-child relationships between entities and cases.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention implements a nested structure where cases contain multiple entities, and each entity contains multiple events. This hierarchical nesting allows users to drill down from the case level to entity level to event level, maintaining traceability at each level while enabling detailed analysis of entity-specific information. The nested structure preserves the case-centric view while adding entity-level detail through nested representations.

Inventive Principle:
Principle #7Nested doll (Nesting)

3Adaptability or versatility

If multiple entity types (orders, invoices, shipments) are involved in the same process steps, then the process exhibits multidimensionality, but existing notations can only represent one linear dimension at a time

Engineering Contradiction:
ImproveMulti-entity process representationVSAvoidGraph structure complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal entity-centric graph model that can represent multiple entity types (orders, invoices, shipments, and any other business entities) within a single unified framework. The graph structure uses generic node and edge elements that can accommodate different entity types through type attributes, allowing the same model to represent diverse entity relationships without requiring separate notation systems for each entity type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4365795A1Interactive multi-object process graphs
Publication Date: 2024.05.08 CELONIS SE
  • EP4365795A1 patent drawingFigure 1
  • EP4365795A1 patent drawingFigure 2
  • EP4365795A1 patent drawingFigure 3

AI summary

The present invention relates to a computer-implemented method to provide a process graph representing process data, wherein the process data is a multidimensional data set which is generated during executions of processes and transformed into an entity-centric data model. The process graph is discovered from the entity-centric data model, wherein the representation of the process graph enables an interactive visual process analysis.