Machine Learning Helps Optimize Manual and Data Intensive Activities to Increase Productivity by 65%

Challenge

Trade Finance is one of the most complex banking operations involving complex international trade practices. The trading operations lifecycle was highly time consuming due to extensive manual effort required for document and data intensive transactions, lengthy process workflow, a lot of manual inspection and interactions with multiple interfacing internal and external systems.

Approach

AutoRec application was introduced for automating the Trade Finance process for Document Check and Bill Lodgement for Import Export Collections. Powered by Machine Learning, it classifies the documents, extracts data from the documents, applies rules to those documents and automatically updates multiple systems as part of operations workflow and compliance checks.

Solution

The end to end validation performed by iLink was one of the most critical factors that determined the success of the AI-ML-OCR system. Validation of the Machine Learning system involved multiple iterations to test intelligently all possible combinations and assessed how well the system was trained. The system was fine-tuned based on the results of testing till the expected level of confidence for outcome was attained.

Outcome

Increased efficiency and accuracy for client

Document Check & Bill Lodgement was reduced from 105 minutes to 60 minutes, a > 40% increase in efficiency.

Document classification accuracy increased by 95% while data extraction accuracy increased by 80%.

Overall the customer saw a 65% increase in productivity.

Machine Learning Helps Optimize Manual and Data Intensive Activities to Increase Productivity by 65%

Challenge

Approach

Solution

Outcome

Increased efficiency and accuracy for client

ADDRESS

PHONE

EMAIL