Bills of Lading processing for shipping
98.46%
accuracy
99%+
reduction in manual errors
$2M+
in annual savings
CHALLENGE Overview
Commodity traders and logistics providers process thousands of Bills of Lading on a daily basis. Often it is done manually, which is prone to human errors. Further, previous attempts to automate the extraction of data from semi-structured documents had failed to meet the levels of accuracy required due to the variability of the BOLs.
For instance, after years of working with a legacy IDP vendor, a global logistics company was only able to achieve 30% accuracy, processing 10M+ BOLs annually with a majority needing to be processed manually.
Client
Tier 1 Global Logistics Company
Goal
Create a trained model to accurately extract data using a small number of sample documents.
WHAT WE DID
AYR automated the classification process, leveraging capabilities such as human-in-the-loop (HITL) capability to send exceptions for specialists for review to improve future models.
Our solution enabled greater accuracy and improved employee productivity, unlocking more time to focus on higher value activities.
IMPACT
- 98.46% accuracy of extraction achieved in <90 minutes training time
- 99%+ reduction in manual errors and extracting precise data
- Average processing of less than 30 seconds, reducing turnaround time from hours to seconds
- Results achieved in a total of 3 days, providing $2M in annual savings