HOW STUDENTS BENEFIT
The Krenicki center will organize student events such as data dives, case competitions and workshops. It will also be an ideal platform for students to conduct research in data analytics. The center will award seed grants to faculty and scholarships or assistantships to students to work on research projects in this domain. Qualified students will be employed to assist in various activities, including company-sponsored projects which will provide them with valuable experiential learning. Students will receive certificates of expertise upon gaining certain competencies.
We developed mathematical models to support and augment the efficiency of legacy heuristics that were followed retail locations selling spares. We obtained binary results for about more than 14000 SKUs, and determined which SKUs to stock based on various constraints like cost, space, minimum quantity.
Optimization Solution to Minimize Costs
If retailers can better handle inventory that will be more successful at fulfilling customer needs, while simultaneously managing the firm’s cash flows for more efficient operations and opportunities. We found that the client’s predicted demand had the highest total cost associated with it due to poor forecasted demand accuracy. As a model with highly accurate predicted demand showed least total cost, we suggest the client focus their efforts in improving the accuracy of their demand forecasts.
Machine Learning for Demand Forecasting
In this study, we observed the applicability of a deep-learning based workflow to complex scenarios which retailers can face like adding new locations, new products, new tastes, and unsystematic external factors. We had a chance to build our own workflow from scratch on a problem by using open-source technologies only, and compare how our solution performed versus a data science team we collaborated with's own proprietary solution.
Promotion Effects on Profits
The purpose of this study is to understand the impact of historical promotions on the demand of a product and create predictive models that best predict demand based on parameters like price, discount, and holidays. Using a traditional log-log model provides some insight about elasticity, but the relationship among demand and price is often non-linear and we posit our demand model can provide a dynamic decreasing elasticity curve that would generate more revenue and yield better profit than traditional approaches.
MIS & Analytics Big 10 Research Conference
The 2019 MIS and Analytics Big 10 Research Conference will provide a forum for all of our Big 10 partners working in the Information System and Analytics research field to present their findings and exchange views on the latest advances in the field. The conference will showcase Information System and Analytics research from world-renowned researchers within the Big 10 and also research from top-notch students from each university.
Esport Data Hackathon: 24 Hour Challenge
Centered on the Purdue University Campus, West Lafayette, IN teams competed in a 24-hour Challenge, using cutting-edge technology (Machine Learning, AI, AR/VR, NLP, Immersive Media, Machine Vision, Blockchain) to develop innovative data solutions that solve problems for Data Visualizations and Predictive Analytics to support Fan Experience, eAthlete Training & Performance.
Purdue/IU Case Challenge Sponsored by Eli Lilly
Similar to real-world business projects, the STAMINA4 case competition is an intensive, experiential learning opportunity that allows students to showcase their critical thinking and analytical abilities, communicate their ideas, and demonstrate mental tenacity. STAMINA4 participants only have four hours to analyze a case and create a presentation to share their recommendations.
Data Science for Business and Economics Conference
The objective of this conference was to feature speakers from business, economics, statistics, computer science, engineering, and other areas as they explore the use of data science to solve real-world problems.