ATM Failure Diagnosis Using First-Abnormal Sensor Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing banking ATMs face frequent recurrence of similar failures due to unresolved failure causes, leading to reduced lifespan and inefficient maintenance, as maintenance workers only repair the affected module without addressing the underlying cause.
Innovation Solution
A method that diagnoses failures by identifying the first abnormal sensor value during a transaction, predicts the failure cause-providing module, and provides repair information to maintenance workers, enabling them to address both the failure-occurred and cause-providing modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of repair
If maintenance workers only repair the module where failure occurred, then repair simplicity is maintained, but failure recurrence increases and lifespan decreases
Solution Approach 1:
The system performs preliminary detection of abnormal sensor values before the actual failure occurs by monitoring sensor data along the medium return route. This allows the maintenance worker to be alerted about potential failure causes in advance, enabling preventive maintenance that prevents failure recurrence while still maintaining straightforward repair procedures.
Solution Approach 2:
The system establishes a feedback loop where sensor values are continuously monitored and compared against normal ranges. When abnormal values are detected, the system provides feedback to the maintenance worker with specific information about the failure cause-providing module, enabling targeted repair actions that prevent recurrence while maintaining repair simplicity.
2Loss of information
If only failure-occurred module information is provided to maintenance worker, then information simplicity is maintained, but root cause analysis is insufficient
Solution Approach 1:
The system extracts and highlights only the critical information needed for root cause analysis - specifically the abnormal sensor values and their locations along the medium return route. This selective extraction provides complete diagnostic information without overwhelming the maintenance worker with unnecessary data, maintaining information simplicity while ensuring completeness.
Solution Approach 2:
The system segments the medium return route into multiple nodes with individual sensors, allowing the failure information to be presented in an organized, segmented manner. This segmentation makes the complex diagnostic information more manageable while ensuring all relevant data about the failure cause-providing module is included.
3Reliability
If maintenance worker visits repeatedly for similar failures, then all failure types are addressed, but maintenance efficiency decreases and time is lost
Solution Approach 1:
By detecting abnormal sensor values before failures occur and providing targeted information about the failure cause-providing module, the system enables maintenance workers to perform comprehensive repairs in fewer visits. This preliminary detection and information provision ensures all failure types are addressed efficiently, reducing repeated visits and time loss.
Data Source
AI summary
The present invention relates to diagnosing a failure of and providing maintenance information about an ATM. When a failure occurs in the ATM, a node at which an abnormal sensor value is initially measured during a failed transaction is identified on the basis of sensor values measured by various sensors provided in the ATM. A failure cause providing module that provided an abnormal operation cause for the failure is predicted and information about a failure occurrence module in which the failure actually occurred and information about the failure cause providing module are provided to a maintenance worker on site, supporting the maintenance worker so that the maintenance worker can, during a failure recovery process, repair the failure occurrence module along with the failure cause providing module that caused the failure.