نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مهندسی عمران، دانشگاه ملی مهارت، تهران، ایران.
2 گروه سیستم های اطلاعات مکانی، دانشکده مهندسی ژئودزی و ژئوماتیک، دانشگاه صنعتی خواجه نصیرالدین طوسی، تهران، ایران.
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Introduction: Flight delays are one of the major problems in the aviation industry. Previous studies have looked less at the reserve aircraft and the effect of their deployment location in reducing delays. These few studies show that most airlines deploy these aircraft at their hubs. As such, in this research was used agent-based simulation in combination with the Bees Meta-Heuristic Algorithm for site selection of these aircraft. The agent-based simulation allows the placement of reserve aircraft at different airports and the calculation of the average delay in flights. The bees algorithm also allows the finding of the optimal position of the reserve aircraft to reduce flight delays in a short time. On the other hand, this algorithm was combined with the ant colony optimization meta-heuristic algorithm. The review of results of this combination in improving the performance of the bees algorithm is one of the main goals of the research. Another purpose is to investigate the effect of proper site selection of reserve aircraft in reducing delays compared to their deployment in hubs.
Materials and methods: In this study, Anylogic software and its GIS map were used. Also, the flight data of Qeshm Air airline was used as real data. The simulation was done as agent-based, including the main agent, the airport agent, the main aircraft agent, and the reserve aircraft agent. At the beginning of the simulation, the main aircraft will fly to their destination based on time of their flights per week and then return to the origin. In the following, airports will call their nearest reserve aircraft in the event of an aircraft failure. The reserve plane will immediately move to the destination of the cancelled flight after arriving at the requesting airport. It will then return to its origin airport. The average flight time of all reserve aircraft to the requesting airports in the total simulation time was considered as the average delay in flights. This average delay is used as the objective function required for each bee in the execution of the bees algorithm. On the other hand, the bees meta-heuristic algorithm was combined with the ant colony optimization algorithm in 3 proposed scenarios. The first, second, and third scenarios were considered as combination in the local search section of the algorithm, combination in the global search section, and simultaneous combination in both sections, respectively. Then, with 50 independent runs of each of the combined algorithms and the initial algorithm with 15 replications each, the effect of these scenarios on improving the performance of the bees algorithm was investigated.
Results and Discussion: The results showed the significant effect of all three scenarios in improving the performance of the bees algorithm in three parts of the algorithm convergence, the repeatability of the results, as well as the average number of repetitions to achieve the optimal solution (the best result found). The third scenario, as the most effective scenario, in the repeatability of the results section, led to 50 optimal solutions out of 50 implementations compared to 3 optimal solutions in the initial bees algorithm, which is a very impressive result. Also, this scenario, the average number of replications in the initial bees algorithm to achieve the best result, which was 14.56, with a 57% reduction, changed to 6.26 replications. The first and second scenarios were ranked next in order of effectiveness, respectively. Also, the proposed hybrid algorithm showed appropriate and acceptable performance compared to similar hybrid algorithms in other research. On the other hand, the results of the research showed the significant effect of about 60% of the appropriate site selection of reserve planes in reducing flight delays compared to deploy them in hubs. Also, the results showed an effect about 87% in reducing delays compared to the worst deployment of reserve aircraft, as the greatest effect of their suitable site selection.
Conclusion: The results show that proper site selection of reserve aircraft can significantly reduce flight delays. On the other hand, meta-heuristic algorithms showed their appropriate efficiency in the site selection of reserve aircraft. Also, the results of the research showed that creative combinations of meta-heuristic algorithms, such as the proposed combination in this study, can have a significant effect on improving their performance.
کلیدواژهها [English]
)