Bills of Lading processing for a Commercial Bank
1
Business Analyst
20
Documents
4
days
98.46%
Accuracy
CHALLENGE Overview
Commodity traders and logistics providers process hundreds of Bills of Lading on a daily basis. Often it is done manually, which is prone to human errors.
Previous attempts to automate the extraction of data from semi-structured documents had failed to meet the levels of accuracy required.
Working with a legacy vendor achieved only 70% accuracy after 4 months, 5000 documents, and the work of 4 data scientists.
Client
Tier 1 Commercial Bank
Goal
Create a trained model to accurately extract data using a small number of sample documents.
WHAT WE DID
The client provided 20 data samples across 5 data variations and asked AYR to train a robust model to extract 12 fields.
- AYR automates this classification process, leveraging capabilities such as human-in-the-loop (HITL) capability to send exceptions for specialists for review to improve future models.
- The solution enables greater accuracy and improves employee productivity, unlocking more time to focus on higher value activities.
IMPACT
- 98.46% accuracy of extraction achieved in less than ninety minutes training time
- 99% reduction in manual errors and extracting precise data
- Reducing turnaround time from hours to seconds
- Results achieved in three days far exceeded the results from a competitor's solution, even after four months of training
