Multi-objective aircraft landing problem: a multi-population solution based on non-dominated sorting genetic algorithm-II

The aircraft landing problem (ALP) is a challenging scheduling and optimization problem in the industry and engineering, which has attracted attention in recent decades. Existing research has predominantly concentrated on optimizing aircraft delay and the financial implications of early or late landings. However, given the paramount significance of airport fuel costs at airports and the critical need for efficient fuel utilization, we aim to minimize airplane fuel consumption by streamlining operational time. In this paper, we present an innovative model with two main objectives: minimizing airplane fuel consumption by reducing dwell time and minimizing cost operation. To address these dual objectives concurrently, we propose a new method known as the multi populations of multiple objectives (MPMO) framework, which is modeled through a non-dominated sorting genetic algorithm-II (NSGA-II) called MPNSGA-II. First, MPNSGA-II employs two separate populations to optimize each objective. Second, to prevent populations from fixating solely on their respective single objectives, MPNSGA-II introduces an archive sharing strategy (ASS). This technique stores elite solutions gathered from two populations. Additionally, we introduce an archive update strategy (AUS) to enhance the quality of solutions stored in the archive. The proposed algorithm has been compared with other well-known algorithms, NSGA-II, multi-objective particle swarm optimization (MOPSO), and NSGA-III. The proposed algorithm shows a cost reduction in 18.01%, 16.75%, and 15.21%. Statistical precision, underscored through the application of the nonparametric Friedman test, corroborates the supremacy of the proposed method, clinching the highest ranking compared to state-of-the-art methods.