Perishable Product Temperature Prediction via Environmental Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for monitoring the temperature of perishable goods are limited by the need for physical sensors inserted into products, which can be invasive and may not provide comprehensive data on environmental conditions affecting product quality.
Innovation Solution
A method and system using temperature probes and environmental sensors to collect data, which generates a machine-learning model that predicts product temperatures based on product type, package type, and environmental conditions, allowing for accurate temperature forecasting without affixing sensors to the products.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical temperature sensors are inserted into perishable products for real-time monitoring, then temperature data can be obtained during harvesting, manufacturing, processing, cooling, storage, and transit, but the sensing method becomes invasive and may not provide comprehensive environmental condition data
Solution Approach 1:
The system separates the sensing function into two parts: invasive temperature probes inserted into products for accurate product temperature measurement, and non-invasive environmental sensors placed in the storage/transport environment for comprehensive environmental condition monitoring. This segmentation allows each sensor type to perform its optimal function without the drawbacks of the other approach.
Solution Approach 2:
The patent introduces environmental sensors as intermediary devices that indirectly measure product conditions by monitoring the environment in which products are stored or transported. These sensors detect temperature, humidity, and other environmental parameters without direct contact with the products, providing comprehensive data while avoiding invasive insertion.
2Reliability
If physical temperature sensors are inserted into products, then real-time product temperature can be monitored, but the system complexity increases and requires sensor insertion/removal operations
Solution Approach 1:
The patent extracts the environmental sensing function from the product itself and places sensors in the surrounding environment. This eliminates the need for repeated insertion and removal of sensors during product handling, while still providing reliable temperature and environmental data through the machine learning model that correlates environmental conditions with product temperature.
Solution Approach 2:
Instead of directly measuring product temperature through physical contact, the system creates a predictive model that copies product temperature information from environmental sensor readings. The machine learning model learns the relationship between environmental conditions and product temperature, allowing indirect but accurate temperature monitoring without physical sensor insertion.
3Adaptability or versatility
If environmental sensors are used without product attachment, then comprehensive environmental data can be collected, but direct product temperature measurement capability is reduced
Solution Approach 1:
The system uses feedback from both invasive temperature probes (for training data) and environmental sensors (for operational data) to continuously improve the machine learning model. The model learns from historical data where both types of measurements were available, then uses only environmental sensor data for predictions, with the feedback loop ensuring ongoing accuracy improvement.
Solution Approach 2:
The patent transforms the measurement approach by changing from direct physical measurement to indirect predictive measurement. Environmental parameters (temperature, humidity, gas composition) are measured and transformed into product temperature predictions through the machine learning model, which learns the complex relationships between these parameters and product temperature.
Data Source
AI summary
Provided is a system, method, and computer program product for predicting product temperatures. The system includes at least one processor programmed or configured to collect, from temperature probes, temperature data of the plurality of different perishable products during a time period, collect, from at least one environmental sensor, environmental data of the at least one environment during the time period, generate a machine-learning model based on the temperature data and the environmental data, collect second environmental data from the at least one environmental sensor and/or from a separate data source while at least one package of at least one perishable product is in the at least one environment, and generate a predicted temperature of the at least one perishable product based on inputting the second environmental data into the machine-learning model.