Business Process Conformance Checking via Probabilistic Model Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for business process conformance checking struggle to effectively measure and address deviations between actual business process behaviors and theoretical models, particularly in noisy and complex log data, which can lead to inefficiencies in process optimization and decision-making.
Innovation Solution
A computer-implemented method and system that matches empirical process models with theoretical models, computes diagnostics for unlabeled process instances, and outputs human-readable diagnostics, including behavior expectedness and frequency, using fuzzy NOT XOR values and decision matrices to identify and correct deviations in process models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current methods are used for business process conformance checking, then process monitoring is performed, but measurement precision of process deviations is insufficient
Solution Approach 1:
The patent transforms qualitative process conformance assessment into quantitative measurement by introducing fitness scores, frequency metrics, and expectedness calculations. These parameter changes enable precise measurement of process deviations through mathematical models that compute deviation magnitude and significance, directly resolving the contradiction between measurement precision and reliability.
Solution Approach 2:
The patent replaces traditional mechanical/conventional conformance checking methods with probabilistic models and fuzzy logic systems. This substitution introduces mathematical rigor through probability distributions and fuzzy set theory, enabling precise yet reliable deviation measurement in complex, noisy environments where traditional methods fail.
2Difficulty of detecting and measuring
If traditional conformance checking methods are applied, then process adherence is monitored, but difficulty of detecting and measuring deviations increases in noisy log data
Solution Approach 1:
The patent introduces probabilistic models as intermediaries between raw log data and conformance assessment. These models act as mediators that filter noise while preserving significant deviation information, reducing the difficulty of detecting deviations without losing critical process information. The probabilistic framework transforms noisy observations into meaningful statistical signals.
Solution Approach 2:
The patent implements feedback mechanisms where deviation measurements inform model refinement and process improvement. By continuously measuring deviations and feeding this information back into the system, the method improves its ability to detect and measure deviations over time, reducing information loss through iterative learning and adaptation.
3Manufacturing precision
If empirical process models are matched with theoretical models, then process alignment is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex model matching task into manageable components: empirical model construction from logs, theoretical model specification, and systematic comparison algorithms. This segmentation reduces device complexity by breaking down the overall complexity into modular, independently implementable steps while maintaining high process alignment precision.
Solution Approach 2:
The patent creates simplified copies or representations of complex process models through probabilistic abstractions and fuzzy logic rules. These copied representations capture essential process characteristics without full complexity, enabling efficient matching while maintaining alignment precision. The copying approach reduces computational complexity by working with condensed model representations.
Data Source
AI summary
The present invention generally relates to systems and methods for checking the conformance of a process behavior against a theoretical process model, and for further providing a detailed diagnostic regarding the process behavior's expectedness and frequency. The provision of this detailed diagnostic includes discovering the empirical models generated by the system's business processes, and matching the defined process model to its corresponding empirical model.