Conformance Diagnostics Using Constraint Mining From Process Nets

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing conformance checking methods in process mining, particularly those using procedural and declarative models, provide low-level, non-deterministic, and often non-scalable diagnostics that require significant user interpretation and are not easily extendable to detect new patterns.

Innovation Solution

A method that generates declarative constraints from procedural models, automatically instantiating constraints from a template library and using them to present conformance diagnostics in a more understandable form, with features like Γ-invariant constraints to reduce state space and an approximated redundancy resolution for scalability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If procedural models are used for conformance checking, then the model completeness and precision are improved, but the diagnostic interpretability and user understanding deteriorate

Engineering Contradiction:
Improvemodel precisionVSAvoiddiagnostic interpretability
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent introduces an intermediary layer that translates procedural model conformance results into natural language explanations. This mediator component bridges the gap between precise procedural models and user-understandable diagnostics, allowing the system to maintain model precision while improving diagnostic interpretability through automated explanation generation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If trace alignment techniques are used to detect deviations, then the detection precision is improved, but the computational complexity and scalability worsen

Engineering Contradiction:
Improvedeviation detection precisionVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the conformance checking process into multiple stages: first identifying candidate deviations using efficient filtering, then applying trace alignment only to relevant cases. This segmentation reduces the overall computational complexity while maintaining detection precision by avoiding exhaustive alignment computations on all traces.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial trace alignment by performing alignment computations only for traces that exhibit potential deviations identified in preliminary analysis steps. This partial action approach maintains detection precision for actual deviations while reducing unnecessary computational overhead on conformant traces.

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If multiple constraint templates are instantiated to detect diverse patterns, then the pattern detection versatility is improved, but the constraint redundancy and verification overhead worsen

Engineering Contradiction:
Improvepattern detection versatilityVSAvoidconstraint redundancy
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal constraint template system where a small set of parameterized templates can express multiple specific constraints. These multi-functional templates reduce redundancy by allowing the same template structure to generate different constraints through parameter substitution, thereby maintaining pattern detection versatility while reducing the total number of constraint definitions.

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

Solution Approach 2:

The patent uses parameter changes to generate specific constraints from general templates. By systematically varying template parameters based on the procedural model structure, the system generates diverse constraints without manually defining each one, thus improving versatility while managing redundancy through parameterized generation rather than explicit enumeration.

Inventive Principle:
Principle #35Parameter changes

4Ease of operation

If declarative models are used for conformance checking, then the diagnostic understandability is improved, but the model completeness and precision worsen

Engineering Contradiction:
Improvediagnostic understandabilityVSAvoidmodel precision
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The patent merges the strengths of both procedural and declarative models by using procedural models for precise conformance checking while presenting results in a declarative, natural language format. This combination allows the system to maintain the precision of procedural models while achieving the understandability of declarative models through automated explanation generation.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4703990A1Method of mining behavioral patterns for conformance diagnostics
Publication Date: 2026.03.04 CELONIS SE
  • EP4703990A1 patent drawingFigure 1
  • EP4703990A1 patent drawingFigure 2~3
  • EP4703990A1 patent drawingFigure 4

AI summary

Computer-implemented method of performing conformance diagnostics of traces of an event log with a net representing a model of a process, the process comprising predetermined steps, the process comprising predetermined allowed variants of sequences of activities, the event log comprising traces, each trace being representative of a sequence of observed activities, the traces being representative of actual sequences of activities, each activity being a step of the process, the net being representable by a la-beled accepting Petri net, comprising: - Instantiating a predetermined set of candidate constraints out of a library comprising predefined templates of constraints, a template being a function that returns a constraint for each k-tuple of activities, each constraint being defined as an object describing a language; - determining, a set of constraints which are satisfied by all traces in the Petri net's accepted language using a predetermined elation holds; - removing redundant constraints from the set of satisfied constraints by searching for a minimal set of redundancy-free constraints; - verifying the event log by checking whether the minimum set of constraints is satisfied by the traces σ of the event log.