Event Log Resource Recommendation for Process Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Process optimization techniques are difficult to perform on event logs due to multiple factors impacting process performance, and manually observing workers to determine additional factors is time-consuming and infeasible for large organizations.
Innovation Solution
A recommender system that utilizes an event monitor to collect data from event logs, convert activities and resources into latent vectors, generate score values, and recommend optimal resources based on user preferences and historical data to optimize processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If process optimization techniques are applied to event logs, then process performance can be improved, but the complexity of analyzing multiple impacting factors increases
Solution Approach 1:
The patent segments the complex optimization problem into separate latent vectors for activities and resources. By dividing the analysis into independent activity representations and resource representations, the system can process multiple impacting factors without increasing overall analysis complexity. Each latent vector captures specific characteristics independently, making the multi-factor analysis manageable through mathematical operations rather than complex logical reasoning.
Solution Approach 2:
The patent introduces latent vectors as intermediary representations between raw event log data and optimization decisions. These latent vectors serve as mediators that transform complex multi-factor analysis into simpler vector operations. The latent vectors capture the essence of multiple impacting factors in a compressed format, enabling optimization without directly processing the full complexity of the original data.
2Measurement precision
If manual observation of workers is used to determine process factors, then comprehensive process analysis is achieved, but time consumption increases significantly
Solution Approach 1:
The patent replaces the mechanical process of manual worker observation with an automated computational system. Instead of human observers manually tracking and recording process factors, the system uses event log parsing and latent vector computation to automatically extract and analyze process characteristics. This substitution eliminates time consumption while maintaining measurement precision through systematic data processing.
Solution Approach 2:
The system performs self-service analysis by automatically extracting process factors from event logs without requiring external human observation. The event log data itself serves as the source of truth, and the system processes this data autonomously through standardized algorithms. This self-service approach eliminates the time-consuming manual observation process while maintaining comprehensive factor detection through systematic data mining.
3Manufacturing precision
If multiple factors are considered in process optimization, then optimization accuracy improves, but the difficulty of detecting and measuring all factors increases
Solution Approach 1:
The patent transforms the detection and measurement of multiple optimization factors into parameter changes within the latent vector space. Instead of directly measuring each factor separately, the system changes the representation of data into latent vectors that inherently capture multiple factors. This parameter transformation simplifies factor detection while maintaining optimization accuracy, as the latent vectors are computed through standardized mathematical operations rather than complex measurement processes.
Data Source
AI summary
Solutions for identifying and optimizing a process are provided herein. A recommender system creates a plurality of event pairs from an event log, and using historical data and features corresponding to the event pairs, the recommender system generates score values for each event pair. Using the score values and user preferences, the recommender system provides a recommended event pair to the user to include in a process.