基于蚁群算法的航班网络座位优化研究

座位优化是航空公司增加收益的有效方法，航班网络座位优化是目前主要的研究方向。针对起始地 目的地 舱位票价（ODF）和座位数组合的复杂性，传统的优化模型由于决策变量数多，难以用于实际计算；改进的线性规划方法在一定程度上改善了模型的实用性，但在求解大规模的网络问题时，计算时间长，复杂度高。采用蚁群算法求解网络座位优化问题能克服以上不足。实验结果表明，蚁群算法能快速得到令人满意的解；同时，蚁群算法简化了问题复杂度，思想简单，易于实现。

Ant colony algorithm in airline seat inventory optimization

Airline seat inventory optimization is a very profitable tool for airline. Current researches are focused on network seat inventory optimization, which has high complication of combination of the ODF (Origin, Destination, Fare) and seat number. Due to the large number of decision variables, traditional optimization models are hard to compute. Although some LP approximation methods of traditional models improve their practical applicability, they still take long time to compute and have high complexity when network is large. We used ant colony algorithm to solve network seat inventory optimization in this paper. It is shown that ant colony algorithm can solve problem quickly and gain good results, and it is easy to implement.