Skip to main content



Chiranjoy Das, Chief Information Officer




 Data & Digital Marketing

Business Challenge

We set out to solve a business challenge to try to develop a platform that will provide insights for heavy equipment, including heatmaps of farm equipment across the country, using data science and machine learning techniques. Such information is of high interest for original equipment manufacturers (OEMs), dealers of farm equipment and financial institutions. In this multi-billion-dollar industry, this one-of-its-kind BI platform will offer intelligence to our clients who can target prospects based on financing details of equipment, geographic region, climatic conditions and numerous business specific criteria.

The goal was to answer the following questions:

  • Where do the farmers finance?
  • Are the farmers moving to commercial banks?
  • Who are the competitive lenders?
  • What is the frequency of purchase of equipment with collateral?
  • How can inventory be managed down to the county level?
  • How should we determine supply and demand based on economic and climatic conditions?

Analytics Solution

We decided to use data science, a statistical probabilistic algorithm called Linear Dirichlet Allocation (LDA) and supervised machine learning to achieve outputs from the equipment data. The second statistical model used to fine tune the LDA-generated output is Correlation Explanation (CorEX). This framework uses semi-supervised learning.

In short, we use probabilistic algorithms to process natural language and provide intelligence to financial institutions to strategize the loans to farmers and dealers. We visualize the data (heat maps and movement of equipment) to depict the trends and supply/demand in equipment in each county in the entire country.


In the absence of a BI system that provides intelligence about farm equipment, our solution has the opportunity to make a huge impact. In addition to OEMs such as Case and Deere, banks such as Wells Fargo and Bank of America have an immediate need for this information. This project is in a unique position to fill an informational void and provide an advantage to Randall-Reilly over all other companies and competitors.