Task Support System for Unknown Event Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems face challenges in accurately evaluating the importance and confidence score of new events, leading to delayed handling of critical events and inability to prioritize based on multiple importance criteria, especially in systems with multiple application products.
Innovation Solution
A task support system that includes an event type evaluation unit, a calculation unit to extract high-confidence event type candidates, and a representation unit to display events in a coordinate space using confidence scores and importance, enabling operators to handle unknown events safely.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple event type candidates are proposed to the operator, then the accuracy of event type identification is improved, but the time required for event handling increases
Solution Approach 1:
The system performs preliminary evaluation of event type candidates by calculating confidence scores based on multiple criteria (event type, occurrence frequency, importance) before presenting them to the operator. This pre-processing reduces the operator's decision time by providing pre-ranked candidates rather than requiring review of all possible event types.
Solution Approach 2:
The system incorporates feedback mechanisms where the operator's corrections to event type identification are recorded and used to refine future evaluations. The confidence scores are adjusted based on actual operator feedback, improving the accuracy of subsequent event type assignments while reducing the need for operator intervention over time.
2Measurement precision
If the operator manually evaluates the importance of each event, then the accuracy of importance assessment is improved, but the productivity of event handling decreases
Solution Approach 1:
The system performs self-evaluation of event importance by automatically calculating importance scores based on predefined criteria and historical data. This self-service mechanism eliminates the need for manual importance assessment by operators, significantly improving productivity while maintaining reasonable accuracy through algorithmic evaluation.
Solution Approach 2:
The system changes the parameters for importance evaluation from subjective manual assessment to objective quantitative metrics. By transforming importance evaluation into measurable parameters (occurrence frequency, event type weightings, temporal patterns), the system enables automated processing while maintaining assessment accuracy.
3Quantity of substance
If the system displays all event type candidates without filtering, then the completeness of information is improved, but the ease of operation decreases
Solution Approach 1:
The system applies local quality by differentiating the presentation of event type candidates based on their confidence scores. High-confidence candidates are prominently displayed with additional details, while low-confidence candidates are summarized or require further interaction to view. This selective presentation maintains information completeness for critical events while improving operational ease for routine events.
Solution Approach 2:
The system segments the event type candidates into distinct groups based on confidence scores and presentation priorities. By dividing the list into high-confidence, medium-confidence, and low-confidence segments with different display characteristics, the system provides complete information where needed while simplifying the interface for less critical events.
Data Source
AI summary
A task support system and method improve an operator's handling of an unknown new event. The task support system and method handle an event that occurred in a management target, in relation to an unknown event, and one or more event types are evaluated as event type candidates. A first event type candidate group having a high confidence score among the evaluated event type candidates is extracted and a first candidate confidence score is calculated from each confidence score of each event type candidate of the first event type candidate group. A first candidate importance is calculated from first importance held by an event type corresponding to each event type candidate belonging to the first event type candidate group and information other than the first importance. A diagram corresponding to each event is represented in a coordinate space by using the first candidate confidence score and the first candidate importance.