The seminars in Mathematical Modeling and Analysis are aimed at researchers, employees, and students.
This week's seminar is given by Jonas Westin, Umeå university.
Title: Modelling a Regional Electric Aviation PSO-Network in the Kvarken Region
Abstract: In many regions worldwide, steps are taken towards using electric-powered aircraft for regional air transport. In sparsely populated regions with long distances, such as the Kvarken Region in northern Scandinavia, electric regional aviation may both have a potential to improve intraregional accessibility and reduce emissions compared to traditional air transport. Around the world, flight services often operate on commercial bases where airlines choose flight routes based on profitability. As a consequence, regions and routes with low or insufficient passenger demand might be excluded from the commercial flight service market. In the European Union, Public Service Obligation (PSO) is a tool for subsidizing air routes in regions where commercial air services are considered unprofitable by airlines.
This purpose of the paper is to analyze the economic potential to create electric regional air networks in the Kvarken region with support of subsidies based on the PSO-system. The optimization model is an integrated flight scheduling and fleet assignment model that creates a schedule that meets a set of PSO-requirement windows while minimizing the operating and social costs of the network.
In the model, a fixed planning period is divided into a number of discrete time steps, each corresponding to a possible departure time (e.g. every 15 or 30 minutes). The movement of aircrafts and passengers is represented by a directed graph where each vertex in the graph corresponds to a possible departure time from an airport. The model has two layers; a flight layer and a service layer. The arcs in the flight layer represent potential aircraft movements between airports, including ground arcs representing aircraft staying on the ground. The edges in the service layer correspond to possible movements of passengers in the network. Both layers share the same vertices, and every arc in the flight layer is linked to a corresponding edge in the service layer for passengers using a flight on that arc. The capacity of each edge in the service layer depends on the number of aircraft on the corresponding arc in the flight layer multiplied by the passenger capacity of aircraft type.
Charging time is an important factor for the turnaround time for electric aviation. We model different charging strategies using combined flight arcs that allows aircrafts to choose arcs with multiple stops. Since a combined flight arc in the flight layer traverses multiple airports, it is connected to multiple edges in the service layer. The model is built in Matlab and solved using Gurobi.
We use the model to analyze a case study of electric aviation in the Kvarken Region. We compare the impact of different network configurations and investigate how different charging times impact fleet size and evaluates the compatibility of existing Swedish PSO regulations with electric aviation solutions.
