Smart Staffing Machine Learning for optimizing crew scheduling

Abstract

The problem of crew scheduling in airlines has received considerable attention in recent years. This problem is often broken down into two steps due to its large size and increasing complexity, first crew pairing, then crew assignment. This work focuses on the crew pairing problem, also known as "CPP ", which is an essential part in crew planning and consists of creating sequences (pairings) of flights, where the company Air Algeria encounters difficulties which require a lot of time to solve this problem in a feasible manner. The objective is to apply machine learning models in solving this problem in order to optimize the use of crews. At first, we tried some machine learning methods, but the results obtained weren't as good, so we used the Machine Learning methods to reduce the complexity of the problem instead by splitting the data, then suggested an algorithm that gave much better and feasible results