Multi-City Flight Route Optimization System Using Big-Data Analytics

This paper proposes a multi-city flight route optimization system using big-data analytics. When one plans a travel for visiting a set of cities, the sequence of visiting should be determined so the total travel cost is minimized. Such problem is known as the travelling salesman problem (TSP) in the literature. We present a variant of TSP considering the dynamic travel costs between two cities and a total travel duration. We present a mixed integer programming formulation for solving the problem, with developing a flight price trend prediction model. The goal of the flight price trend prediction model is to reduce time and cost that are required to construct the route optimization problem. We employed big-data analytics to build the appropriate price trend prediction model that involves massive web crawling, statistical analysis, and regression. A web-based system was, then, built on the developed models for demonstrating the validity of the proposed approach.