Show Filters

Top Results

Automation in Supply Chain

At a Glance

Machine Learning

Reduced manual processes by developing cloud-based machine learning service

Time Savings

Elimination of more than 50% of targeted manual work

Process Automation

Provided client the ability to process thousands of pages of tariff packages in minutes

Challenge

A large electronics product distributor, more than 300 locations worldwide, was experiencing issues and delays with their tariff drawback process. The objective was to pilot an automation program for reading import forms and product invoices to extract pertinent information used in the drawback process. ScottMadden developed a solution that leveraged a cloud-based machine learning service and custom code to generate a structured output which paired with federal customs reports to facilitate applications for drawback and reduce manual processing.

Process

  • Set up cloud architecture and permissions
  • Scripted API calls to machine learning service to transform unstructured data into semi-structured data set
  • Developed custom script for data cleaning to transform semi-structured data to structured data set
  • Output clean, structured data to CSV for use in existing processes

Result

  • Demonstrated ability to process thousands of pages of tariff packages in minutes
  • Elimination of more than 50% of targeted manual work
  • Matching with customs report eliminated risk of missed packages in manual processing
  • Output clearly denoted areas where continued manual processing was required

Related Insights

Let’s Work Together

We don’t solve problems with canned methodologies. We help you solve the right problem in the right way. Our experience ensures that the solution works for you.