Consumable Usage Sensing With Edge Summaries and Raw-Data Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for determining consumable product usage lack accuracy and confidence in reported consumption data, as they often provide information at a single point in time rather than over a long period and rely on potentially inaccurate consumer feedback.
Innovation Solution
A system with sensors and edge algorithms that collect raw data during sample time periods, execute edge algorithms to summarize data into single outputs, and utilize a cloud-based server for independent analysis to enhance confidence in consumption data, incorporating feedback loops for algorithm refinement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If consumer feedback is used to obtain usage information, then data collection is simple, but accuracy and reliability of consumption data deteriorates
Solution Approach 1:
The patent replaces manual consumer feedback (mechanical/human system) with automated sensor-based measurement systems. Sensors continuously monitor product usage parameters such as weight changes, container opening/closing events, and product dispensing, substituting human reporting with objective mechanical measurements that provide both ease of collection and high reliability.
Solution Approach 2:
The patent introduces sensors and edge algorithms as intermediaries between the consumable product and the data collection system. These intermediaries automatically capture usage data without requiring direct consumer input, mediating the measurement process to eliminate inaccuracies in self-reported feedback while maintaining simple data collection infrastructure.
2Device complexity
If usage data is collected at a single point in time, then data processing is simplified, but ability to indicate actual usage over long period deteriorates
Solution Approach 1:
The patent implements continuous usage monitoring through sensors that continuously measure product consumption parameters over extended periods. Weight sensors continuously track product mass, container sensors continuously detect opening/closing events, and dispensing sensors continuously monitor product usage rates, providing an unbroken data stream that accurately reflects actual usage patterns over time rather than isolated snapshots.
Solution Approach 2:
The patent performs preliminary data processing and filtering at the edge device before transmission to central systems. Edge algorithms pre-process continuous sensor data to identify meaningful usage events, filter noise, and aggregate measurements, reducing the complexity of subsequent centralized analysis while preserving the benefits of continuous long-term monitoring.
3Measurement precision
If all raw sensor data is transmitted to server, then analysis accuracy is improved, but data transmission and processing load increases
Solution Approach 1:
The patent segments the data processing function between edge devices and central servers. Edge devices perform local preprocessing, filtering, and aggregation of raw sensor data, transmitting only processed results or selectively sampled data to the server. This segmentation reduces transmission volume and energy consumption while preserving measurement precision through distributed intelligence.
Solution Approach 2:
The patent transmits only the necessary portion of raw sensor data rather than all collected data. Edge algorithms identify and transmit only meaningful usage events, filtered measurements, or aggregated statistics that are sufficient for accurate usage determination, avoiding unnecessary transmission of redundant or noisy data points that would consume energy without improving precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for determining usage of a consumable product includes a device assembly to receive the consumable product, and a server. The device assembly includes sensors to collect raw sensor data during sample time periods, and a controller to execute edge algorithms based on the collected raw sensor data. The edge algorithms summarize the raw sensor data collected during the sample time periods to single edge algorithm outputs. The server analyzes the single edge algorithm outputs to make a determination on the usage and consumption of the consumable product. To obtain a higher degree of confidence in performance of the edge algorithms, the server independently analyzes the raw sensor data. As refinements are made to the edge algorithms, the confidence of the edge algorithms is increased. Updated versions of the device assembly are simplified so that a minimal amount of raw sensor data is reported.