IoT Failure Symptom Detection via User Feeling Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing failure detection systems are unable to detect service level failures independently of product failures and cannot proactively address user discomfort issues in facilities with distinct user and operator roles, as they require user intervention and focus on product errors rather than service-level issues.
Innovation Solution
A failure symptom detection system that collects and analyzes data from IoT devices to identify service level failures by matching user-reported feelings with historical failure data, providing proactive measures to facility managers before the failure occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a system compares log data of operation immediately before error with known error causes in a database, then the cause and measure can be provided to the user, but the system cannot detect service level failures independently of product failures and requires user intervention
Solution Approach 1:
The system segments failure detection into two independent pathways: one for product failures (using log data comparison) and one for service level failures (using IoT device data monitoring). The service level failure detection pathway independently monitors IoT device data without requiring user intervention or product failure reports, enabling separate detection of service issues from product issues.
Solution Approach 2:
The system introduces an intermediary data collection mechanism that continuously gathers IoT device data (temperature, humidity, device status) and compares it against historical failure patterns. This intermediary layer enables automatic detection of service level failures before they manifest as user complaints, bridging the gap between device operation and service quality.
2Reliability
If the system functions after a failure occurs by identifying and displaying cause and measure, then user feedback is obtained, but the system fails to detect failure symptoms proactively
Solution Approach 1:
The system performs preliminary monitoring of IoT device data continuously, comparing current device states against historical failure patterns before actual failures occur. By detecting failure symptoms in advance through pattern recognition in device data (temperature trends, operational anomalies), the system enables proactive intervention before service level failures manifest, reducing response time and preventing user dissatisfaction.
Data Source
AI summary
A failure symptom detection system includes a first storage to collect and store field data of each of a plurality of Internet of things devices, a feature extractor to acquire feature data of the field data based on a report on a failure in a service as a feeling of a user of a facility, a second storage to accumulate a failure at an occurrence of the failure associated with content of the failure as the feeling of the user, and a failure symptom detector to monitor the field data stored in the first storage, and produce, upon detecting feature data matching the feature data accumulated in the second storage, output indicating detection of a symptom of the failure associated with the feature data.