Airline restructuring: A dynamic programming approach combined with genetic algorithm for network and fleet downsizing

In the decade following the 2007–2008 financial crisis, more than 1,400 airlines, commercial air services, and charter companies went bankrupt. During the COVID-19 pandemic, hundreds of airlines applied to declare bankruptcy. Despite these challenges, many airlines have demonstrated the ability to recover through successful restructuring following bankruptcy protection. A famous example is Japan Airlines (JAL), which faced bankruptcy in 2010 but successfully returned to the market through the Tokyo Stock Exchange in just two years and eight months. At the core of JAL's restructuring was network and fleet downsizing. This study formulates a dynamic programming (DP) model to assist the restructuring airline's action on each route and aircraft, considering the airline's ability to repay debt. A genetic algorithm (GA) is devised to solve the proposed problem. The numerical experiment results suggest that airlines should withdraw routes with huge losses, deficient load factors, low profits, and few connections. Furthermore, the restructuring airline must stop providing flights on routes that do not have direct links to its base country, despite the high connectivity of these routes. For routes with high connectivity and potential profitability, the results suggest that the airline terminates operations in the early stage of restructuring but resumes them later. Regarding fleet downsizing, the airline is required to remove old aircraft types, retain those with moderate debt and assets, and introduce new aircraft types into the fleet. The results closely mirror the patterns observed in JAL's restructuring, suggesting the feasibility and validity of the proposed approach.