Utilizing Analytics to Predict Deliveries at CHS River Terminals

My client-based capstone project was done for CHS, Inc (Cenex Harvest States).  CHS is a diversified global agribusiness cooperative owned by farmers and local cooperatives across the United States.  CHS employs more than 10,000 people around the globe who provide their owners and customers access to global markets with a strong, efficient supply chain.  The main objective of this study dealt with helping that supply chain by identifying factors that influence drop-off traffic patterns of crops at their terminals and processing plants.  These terminals serve as the entry points for farmers and cooperatives into the global marketplace for crops such as corn, soybeans, and wheat.  Finding what these indicators are can be used to develop better staffing methods to serve customers in a more effective, cost-efficient manner.  Secondary objectives for this project include both empowering CHS business analysts with previously unavailable datasets and educating people on the power of a “data science” approach to problem-solving.   

Results of this project showed a correlation with different market price features on delivery, especially for the farmers who have contracted to deliver with CHS.  From a strictly analytical standpoint, this study also demonstrated the importance of testing different types of forecasting models and ensuring optimal hyperparameters are utilized within those models.