Challenge
The shipping company encountered a significant challenge concerning the inefficient utilization of dock space, primarily due to the uncertainty surrounding the availability of empty containers on any given day.
The Solution
A machine learning solution was created to predict the number of empty containers arriving at the port at any given date in the next month. It consisted of the following features:
- ETL Implementation: Developed an automated ETL pipeline for migration of tables to the cloud, thereby reducing the cost and time needed to upload data.
- AI/ML Use Case Implementation: Designed an advanced AI/ML model for precise prediction of the destination port for container returns.
Services
- AI/ML modeling
Technologies Used
GCP Data Flow
GCP Big Query
GCP Cloud Composer
GCP Cloud Storage
Impact by the NUMBERS
86%
accuracy in predicting the number of empty containers at the port
11%
increase in previous average precision after adding just 20 days of more data
More Success Stories
