Intelligent Inventory and Personnel Management

Kent Building Supplies (referred to as Kent from hereon) has 49 retail locations across Atlantic Canada. We have started developing Robotic Process Automation (RPA). RPA significantly reduce the human intervention and generate exception-based flagging reports. We would like to continue deployment of RPA for new tasks and use unsupervised and supervised machine learning to improve the rules used and exceptions generated in RPA processes.
Kent has also initiated predictive and prescriptive analytics using machine learning for their inventory management. One of their first forays was into the use of statistical and machine learning based regression analysis to predict an acceptable range of inventory for every product at any point of time in every store. The results are encouraging. We would like to refine these predictions models to include additional variables such as promotions, seasons, weather. Kent stores vary in size from a small rural store to a mega-store in an urban setting. We would like to explore the possibility of using transfer learning so that stores can benefit from the data across the entire chain. Furthermore, we would like to use AI based optimization to extend the inventory management over a sub-region.

Faculty Supervisor:

Pawan Lingras

Student:

Partner:

Kent Building Supplies

Discipline:

Computer science

Sector:

Retail trade

University:

Saint Mary's University

Program: