Machine-Learning Container Yard Imaging for Real-Time Inventory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual yard checks in intermodal container yards are time-consuming and prone to errors, as workers visually observe and manually enter shipping container locations, leading to inefficiencies and resource wastage.
Innovation Solution
Implementing an inventory imaging system with machine-learning capabilities to automatically identify and determine the locations of shipping containers using imaging devices and maps, communicating data electronically only when a container is detected, thereby optimizing network bandwidth and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual yard checks are performed by workers visually observing and manually entering container locations, then inventory data can be collected, but the process becomes time-consuming and error-prone
Solution Approach 1:
The patent replaces the manual mechanical process of visual observation and data entry with an automated imaging system using cameras and machine learning algorithms. The system captures images of containers and automatically processes them to extract location and identification data, eliminating the need for manual intervention and significantly reducing both time and errors.
Solution Approach 2:
The system enables self-service by allowing the imaging and machine learning module to autonomously perform the entire yard check process without human assistance. The automated system captures images, identifies containers, determines their locations, and updates the inventory database independently, freeing workers from repetitive manual tasks.
2Loss of information
If continuous imaging and data transmission are performed throughout the yard, then real-time inventory data is obtained, but network bandwidth and computer resources are wasted
Solution Approach 1:
The patent applies local quality by making the data transmission selective rather than uniform. The system transmits information only from locations and at times when changes occur (when containers are detected or moved), rather than continuously from all areas. This targeted approach maintains real-time data availability for relevant containers while minimizing unnecessary network and computational resource consumption.
Data Source
AI summary
According to some embodiments, a method by a computing system includes accessing a plurality of images captured by an imaging device. The imaging device and the computing system are coupled to a vehicle that moves within an intermodal container yard. The method further includes determining, by analyzing the plurality of images using a machine-learning module, that a shipping container is depicted within at least one of the plurality of images. The method further includes determining, using a map of the intermodal container yard, a parking location of the shipping container. The method further includes electronically communicating, in response to determining that the shipping container is depicted within at least one of the plurality of images, a message comprising data about the shipping container. The data includes the determined parking location of the shipping container and one or more identification markings of the shipping container.