Process Feedback Mapping for Linking Sentiment to Process Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current business transformation approaches neglect the interdependence between process-focused reality and experience-focused sentiments, leading to suboptimal business performance due to unaddressed bottlenecks or negative customer experiences.
Innovation Solution
An integrated framework that utilizes a process data store, process server, and machine learning to automatically map user feedback data to process models, enabling secure, automatic, and efficient analysis of both process-focused reality and corresponding sentiments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If process mining and BPM techniques are used to optimize process efficiency, then process execution speed is improved, but customer satisfaction may deteriorate due to negative sentiments not being addressed
Solution Approach 1:
The patent combines process mining techniques with sentiment analysis techniques into a unified framework. Process mining data (process graphs, nodes, edges) is integrated with sentiment analysis data (customer feedback, satisfaction metrics) to simultaneously optimize both process efficiency and customer satisfaction. This merging allows the system to identify bottlenecks while also understanding their impact on customer experience.
Solution Approach 2:
The system implements feedback loops where sentiment analysis results are fed back into process optimization decisions. Customer satisfaction metrics and sentiment data are continuously monitored and used to adjust process configurations, ensuring that efficiency improvements do not compromise customer experience. The framework enables bidirectional influence between process reality and customer experience.
2Loss of information
If sentiment analysis is performed without integrating process data, then customer experience understanding is improved, but process bottlenecks remain undetected
Solution Approach 1:
The patent merges sentiment analysis capabilities with process mining capabilities in a single integrated system. The framework processes both unstructured customer feedback data and structured process execution data simultaneously, enabling comprehensive analysis that links customer experiences to specific process steps, bottlenecks, and performance issues.
Solution Approach 2:
The system segments customer sentiment data and maps it to specific process nodes and edges in the process graph. This segmentation allows identification of which particular process steps are causing negative sentiments, enabling targeted process improvements rather than generic optimizations.
3Measurement precision
If manual mapping of survey questions to process data is performed, then mapping accuracy is improved, but analysis time and complexity increase
Solution Approach 1:
The system employs automated algorithms that self-map survey questions and sentiment data to relevant process elements without requiring manual intervention. The framework uses natural language processing and pattern recognition to automatically associate customer feedback with corresponding process nodes, edges, and attributes, significantly reducing analysis time while maintaining high mapping accuracy.
Solution Approach 2:
The patent replaces manual mapping mechanisms with automated computational methods. Instead of human analysts manually matching survey questions to process data, the system uses machine learning algorithms, text mining, and data association techniques to perform the mapping automatically, eliminating the time-consuming manual process while preserving or improving accuracy.
Data Source
AI summary
A process data store may contain a process model (e.g., a process graph, with process graph elements that include nodes and edges, as generated via process mining or a BPMN representation). A process server may retrieve information from the process data store and receive user feedback data. The server may determine if the information retrieved from the process data store is associated with a prior mapping of survey questions to the information retrieved from the process data store. If the information retrieved from the process data store is not associated with a prior mapping of survey questions, embodiments may utilize Machine Learning (“ML”) to automatically map the user feedback data to the information retrieved from the process data store. The server may then automatically assign, group, and analyze the user feedback data to generate a recommended alteration.