Wednesday, July 2nd, 2025
Media Contact: Tanner Holubar | Communication Specialist | 405-744-2065 | tanner.holubar@okstate.edu
Research using machine learning and data analytics spans a wide range of industries, from infrastructure to healthcare. Researchers from the University of Engineering, Architecture and Technology in the Faculty of Industrial Engineering and Management at Oklahoma State University aim to improve Jewelers Mutual's business processes by employing machine learning and data science.
The project is part of a collaboration between the University of Wisconsin Madison University and OSU, and is working with Jewelers Mutual. The main contribution of OSU is to create custom business credit score models that will be added to the company's tool belt.
The collaboration spans multiple research projects and focuses on preventing losses for the company's clients. By studying historical data on crime-related losses, safety measures, and store locations, researchers identify patterns and risk factors to improve the company's underwriting process and customer guidance.
IEM assistant professor Dr. Akash Deep has been working together in the collaboration for five years, and this is the latest venture. He co-adapted for his Ph.D. In Wisconsin, Dr. Raj Veramani and Dr. Seiming. Deep began working on the collaboration before becoming an OSU teacher in 2022.
“The overall goal of this project is to see how machine learning models and data science insights can be used to improve business processes,” Deep said.
Custom Business Credit Score Models improve risk assessment decisions during the company's underwriting process. Create models using third-party data, including credit-based consumer information from major credit agencies.
This study has three main goals: 1) Define opportunities for improvement and discover ways data science can mitigate it. 2) develop and apply data science and machine learning to address opportunities including data analysis, model construction and prototype creation. 3) They share their findings with jewellers to potentially apply to the company's business practices.
“We can usually claim to be data science experts, but of course the company is an expert in their business,” Deep said. “We start with descriptive analysis and try and understand what we can do through the data. We want to handle the data properly so that we can effectively optimize their business.”
Deep's team wants to reach a research point where it can predict certain outcomes, such as losses. This is a collaboration that Deep enjoys taking part in multiple research efforts.
“We're very fortunate to be able to work with Jewelers Mutual,” Deep said. “It was an interesting experience for me personally, I can do a lot of mathematics and create models, but when used for certain beneficial reasons, it creates real value.
“It's a great project to work to really see how we can help. It's a great, synergistic collaboration. Working with business experts and very clever students, I think it was a very, very wonderful experience.”
