Sequential Process Prediction Model Structure Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems fail to effectively predict outcomes in sequential processes involving multiple variables where the temporal sequence and dependencies are not clearly known, leading to challenges in identifying stages, dependencies, and predictors for multi-stage prediction settings.
Innovation Solution
A method is developed to identify sub-process sequences from temporal datasets, categorize predictors based on availability, group them into sub-processes, and generate a dependency graph to determine a final prediction model that captures the temporal order of predictors, allowing for adjustments in the sequential process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multi-stage prediction is applied step-by-step using predicted values from current time step to determine next time step, then prediction of sequential processes can be achieved, but error accumulation occurs and prediction accuracy deteriorates
Solution Approach 1:
The patent segments the sequential process into distinct stages with identified dependencies between them. By breaking down the complex prediction problem into manageable stages and modeling their interrelationships, the system avoids cumulative errors that occur in step-by-step prediction approaches.
2Ease of operation
If independent value prediction is used to build separate models for each prediction step, then learning difficulty is reduced, but the temporal dependencies and sequential nature of the process are not captured
Solution Approach 1:
The patent introduces an intermediary component that captures temporal dependencies and sequential relationships between process stages. This intermediary structure allows the model to learn from historical patterns while maintaining the benefits of staged prediction, bridging the gap between simple independent models and complex sequential models.
3Reliability
If the temporal sequence and dependencies in sequential processes are not clearly known, then prediction models cannot be effectively constructed, but discovering these dependencies increases system complexity
Solution Approach 1:
The patent employs self-service mechanisms where the system automatically discovers temporal sequences and dependencies from historical data without requiring manual specification. Through automated pattern recognition and dependency detection, the system identifies the sequential structure of processes, reducing the complexity burden on users while maintaining high prediction effectiveness.
Data Source
AI summary
A sub-process sequence is identified from a temporal dataset. Based on time information, predictors are categorized as being available or not available during time periods. The predictors are used to make predictions of quantities that will occur in a future time period. The predictors are grouped into groups of a sequence of sub-processes, each including a grouping of one or more of the predictors. Information is output that allows a human being to modify the groups. The groups are finalized, responsive to any modifications. Prediction models are extracted based on dependencies between groups and sub-processes. A final predication model is determined based on a prediction model from the prediction models that best meets criteria. A dependency graph is generated based on the final prediction model. Information is output to display the final dependency graph for use by a user to adjust or not adjust elements of the sequential process.