Container Internal Weather Modeling Without Transit Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating internal conditions of shipping containers lack accuracy and timeliness, especially under varying weather conditions, and often require expensive sensor installations that can fail during transit.
Innovation Solution
A method utilizing external weather data and machine learning models, combined with physics-based features, to predict container temperature and relative humidity without sensors, by resampling weather information in response to shipping updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed in containers to monitor internal conditions, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a virtual copy of the sensor measurement system by using machine learning models that replicate sensor functionality. Instead of physically installing sensors in containers, the system uses external weather data and ML algorithms to generate synthetic sensor readings, thereby achieving measurement precision without the complexity of physical sensor deployment
Solution Approach 2:
The patent replaces the mechanical sensor installation system with an information processing system. Rather than using physical sensors to detect internal conditions, the system substitutes a computational approach using machine learning models that process weather data to predict internal container conditions, eliminating the need for mechanical sensor deployment
2Measurement precision
If sensors are installed in containers to monitor internal conditions, then measurement precision is improved, but reliability deteriorates due to sensor failure during transit
Solution Approach 1:
The patent creates a virtual copy of the sensor measurement system by using machine learning models that replicate sensor functionality. Instead of physically installing sensors in containers, the system uses external weather data and ML algorithms to generate synthetic sensor readings, thereby achieving measurement precision without the complexity of physical sensor deployment
Solution Approach 2:
The patent introduces external weather data as an intermediary between the environment and the prediction model. Rather than relying on direct sensor contact with container internals, the system uses weather information from external sources as a mediator to infer internal conditions, improving reliability by removing the single point of failure that physical sensors represent
3Device complexity
If linear models are used for estimating internal climate conditions, then device complexity is reduced, but measurement precision deteriorates under varying weather conditions
Solution Approach 1:
The patent transitions from static linear models to dynamic machine learning models that can adapt to varying weather conditions. The ML models are trained on diverse weather data and can dynamically adjust their predictions based on current weather patterns, maintaining low device complexity while significantly improving measurement precision across different environmental conditions
Solution Approach 2:
The patent changes the parameters of the estimation model from fixed linear relationships to flexible machine learning parameters that can be optimized through training. By using ML models with learnable parameters instead of fixed linear coefficients, the system maintains simplicity while achieving high accuracy across varying weather conditions through data-driven parameter optimization
4Ease of operation
If route-specific humidity and temperature estimations are performed without physics theory, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent replaces the mechanical approach of direct sensor measurement with an information processing system that uses machine learning. The ML models substitute physical measurement mechanisms with computational inference, maintaining ease of operation while improving precision by leveraging learned patterns from training data rather than relying on simple extrapolation
Data Source
AI summary
A method for estimating status of a container. The method comprising obtaining, by a processor, shipping information of the container; extracting, by the processor, weather information received from one or more databases from one or more locations corresponding to a location and time interval of the shipping information of the container; executing, by the processor, pre-processing on the weather information for an input to a feature generator to output intermediate features; and using, by the processor, the intermediate features to predict container temperature and container relative humidity, wherein the weather information is periodically resampled in response to updates to the shipping information of the container.