Starting Semester: Fall 2022
Client ProfileThe Operations Research group at American Airlines provides internal consulting to various functions of the company, such as Revenue Management, Network Planning, Loyalty, Sales, Marketing, and more. This project in particular supports the Technical Operations organization, which oversees the maintenance of aircraft.
Project DescriptionCapstone project - Long Term Part Demand Forecasting To support our operation at American Airlines (AA), we need to ensure availability and reliability of our aircrafts to fly. This requires our maintenance activities to be performed when needed. The supply chain team at AA plans and executes the procurement, repair, and supply of all parts for maintaining the aircraft. In short, they need to fulfill various parts' demand at the right price, at the right time and at the right place. Given a historical part usage across different maintenance stations, we would like to predict the usage of every part at a specific maintenance location for twelve months in the future. There are around 300K distinct parts in AA inventory. However, we will focus on Rotable parts accounting for around 60K parts across ~20 maintenance replenishing stations. We expect the team to perform statistical analysis on parts' usage behavior, understand their demand distribution and analyze parts with different demand behaviors. The goal is to cluster/categorize parts by their demand behavior and propose effective models to predict their future demand levels. The choice of models can vary from history-based models such as exponential smoothing or time series models to Machine Leaning based models that use forward looking information. Proper evaluation metrics such as mean absolute errors or mean absolute percentage error can be used to compare the performance of models.
SkillsStatistics, data science