Nowadays this concept has been widely assessed due to its complexity and sensitivity for the beneficiaries, including passengers, airlines, regulatory agencies, and other organizations. To date, various methods (e.g., statistical and fuzzy techniques) and data mining algorithms (e.g., neural network) have been used to solve the issues of air traffic management (ATM) and delay the minimization problems. However, each of these techniques has some disadvantages, such as overlooking the data, computational complexities, and uncertainty. In this paper, to increase the air traffic management accuracy and legitimacy we used the bidirectional long short-term memory (Bi-LSTMs) and extreme learning machines (ELM) to design the structure of a deep learning network method. The Kaggle data set and different performance parameters and statistical criteria have been used in MATLAB to validate the proposed method. Using the proposed method has improved the criteria factors of this study. The proposed method has had a % increase in air traffic management in comparison to other papers. Therefore, it can be said that the proposed method has a much higher air traffic management capacity in comparison to the previous methods.

如今，由于这一概念对受益人（包括乘客，航空公司，监管机构和其他组织）的复杂性和敏感性，已经得到了广泛的评估。迄今为止，已经使用各种方法（例如统计和模糊技术）和数据挖掘算法（例如神经网络）来解决空中交通管理（ATM）的问题并延迟最小化问题。但是，这些技术中的每一种都有一些缺点，例如，忽略数据，计算复杂性和不确定性。在本文中，为了提高空中交通管理的准确性和合法性，我们使用了双向长短期记忆（Bi-LSTM）和极限学习机（ELM）来设计深度学习网络方法的结构。在MATLAB中已使用Kaggle数据集以及不同的性能参数和统计标准来验证所提出的方法。使用提出的方法改善了这项研究的标准因素。与其他论文相比，所提出的方法在空中交通管理方面增加了％。因此，可以说所提出的方法与以前的方法相比具有更高的空中交通管理能力。