Automated Process Transformation via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current process mining techniques require human intervention and subjectivity to improve operational processes, which can be time-consuming and inefficient, as they depend on expert judgment and do not fully automate the transformation of processes based on computer-generated event data.
Innovation Solution
The integration of machine learning models, specifically a first model for mapping key performance indicators (KPIs) to a discovered process model and a second model for identifying process transformations, calculates process value debt and transformation propensity scores to recommend automated improvements, leveraging AI to quantify and unlock improvement potential in operational processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If process mining techniques use human intervention and expert judgment to improve operational processes, then the quality of process improvement can be maintained through subjectivity, but the time consumption and efficiency deteriorate
Solution Approach 1:
The system enables automated process improvement by having the process mining technology perform analysis and transformation recommendations without human intervention. The automated system serves itself by using machine learning models to identify discrepancies, calculate enhancement potentials, and generate transformation recommendations, eliminating the need for expert judgment while maintaining improvement quality
Solution Approach 2:
The patent replaces the mechanical system of human expert judgment with an automated computational system using machine learning models. The first machine learning model maps KPIs to process models, and the second model identifies transformations, substituting human cognitive processes with automated algorithms that achieve the same improvement objectives without time loss
2Extent of automation
If process mining techniques do not fully automate the transformation of processes, then human expert judgment can be applied to maintain quality, but the automation level and efficiency deteriorate
Solution Approach 1:
The automated process improvement system is segmented into distinct functional components: event data retrieval, process model discovery, KPI mapping by the first machine learning model, discrepancy identification, enhancement potential calculation, and transformation recommendation by the second machine learning model. This segmentation enables full automation while maintaining efficiency by allowing each component to operate independently and contribute to the overall productivity improvement
3Adaptability or versatility
If manual process improvement methods are used, then flexibility and adaptability can be maintained, but the productivity and output per unit time deteriorate
Solution Approach 1:
The system maintains flexibility and adaptability through dynamic machine learning models that can adjust to different processes and KPIs. The automated system adapts to various operational contexts by learning from event data and adjusting transformation recommendations, while simultaneously increasing productivity through automated execution that eliminates manual processing bottlenecks
Data Source
AI summary
Process transformation can include mapping, by a first machine learning model, predetermined key performance indicators (KPIs) to a discovered process model. For each of the KPIs, a KPI gap and a KPI impact score can be determined. For each of the KPIs, a KPI-level enhancement potential value based on the KPI gap and KPI impact score of each KPI can be determined. Based on the KPI-level enhancement potential value of each of the KPIs, a process value debt (PVD) can be generated. Responsive to the PVD exceeding a predetermined threshold, a process transformation recommendation generated by a second machine learning model can be outputted to identify at least one modification to the process that is likely to reduce the PVD.