Leveraging Artificial Intelligence in the Procurement Process: For Enhancing Automation and Predictive Risk Management

This thesis aims to address the inefficiencies of traditional procurement processes. Being inefficient in today's competitive business world costs the company in many ways. It leads to higher costs due to delays, wasted time, and difficulties in managing operations effectively. With this motivation, we design and build a potential hybrid system combining AI with Excel-based automation. This study explores how the hybrid prototype is built and how can it improve procurement efficiency by focusing on two main areas. The first one is automating repetitive tasks, in particular the preparation and generation of Purchase Order documents. The second focus is to use AI tools to predict risk related to today’s uncertain geopolitical events to give insights into the global risk scenarios. We compared and evaluated three different predictive models in order to figure out the prediction's actual potential. Based on the findings, I have presented the design framework prototype for implementing a hybrid approach to the procurement operational process, along with key change management recommendations to facilitate integration and facilitation of the procurement system. The goal of this thesis is to provide a tool for the company to enhance its efficiency by reducing processing time and minimizing unnecessary costs. In addition, the tool is expected to enhance decision-making by supporting more effective risk management within the procurement process.