AI Accelerator Consumable Life Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Consumables in devices are often discarded prematurely due to unknown remaining usable life and lack of recycling options, leading to device malfunction and potential harm.
Innovation Solution
An AI acceleration component within a memory sub-system determines the life expectancy of consumables using manufacturer, third-party, and usage data, and recommends secondary uses or recycling instructions, ensuring timely replacement and reducing waste.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If consumables are discarded when remaining life is unknown, then device simplicity is maintained, but resource waste increases and device reliability decreases
Solution Approach 1:
The system performs preliminary actions by tracking consumable usage data and predicting remaining life before the consumable actually fails. The AI accelerator analyzes usage patterns and predicts when consumables will deplete, allowing users to replace them proactively rather than reactively, thus preventing device malfunction and reducing waste from premature disposal
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring consumable usage data and providing real-time information about remaining life to users. The AI accelerator processes usage data and feeds back predictions about consumable status, enabling users to make informed decisions about replacement timing, which improves device reliability and reduces unnecessary waste
2Loss of substance
If consumables are tracked and monitored, then resource efficiency improves, but system complexity increases
Solution Approach 1:
The system applies self-service by enabling the consumable itself or its associated device to automatically track and report its own usage data. The AI accelerator then autonomously analyzes this data and provides predictions without requiring user intervention, reducing the complexity burden on users while maintaining comprehensive tracking capabilities
Solution Approach 2:
The patent replaces complex mechanical tracking systems with AI-based predictive analytics. Instead of using elaborate physical sensors and monitoring mechanisms, the system uses machine learning models that analyze existing usage data to predict consumable life, significantly reducing system complexity while maintaining or improving accuracy
3Measurement precision
If AI operations are performed to determine life expectancy, then consumable management accuracy improves, but computing energy consumption increases
Solution Approach 1:
The system applies partial action by performing AI operations selectively rather than continuously. The AI accelerator processes usage data at strategic intervals or only when significant changes occur in consumable usage patterns, maintaining sufficient prediction accuracy while minimizing unnecessary computing operations and energy consumption
Solution Approach 2:
The system changes parameters by adjusting the frequency and depth of AI operations based on consumable criticality and usage patterns. For less critical consumables or stable usage patterns, the system reduces AI operation frequency, while increasing accuracy for critical consumables, thereby optimizing the balance between measurement precision and energy consumption
Data Source
AI summary
A method includes receiving, at an artificial intelligence (AI) accelerator of a computing system, at least one of: manufacturer data, third-party data, sensor data, or primary usage data of a consumable in a primary device and performing an AI operation on at least one of: the manufacturer data, the third-party data, the sensor data, or the primary usage data at the AI accelerator of the computing system using an AI model. The method further includes determining a primary life expectancy of the consumable in the primary device at the AI accelerator in response to performing the AI operation.