Accessing airports can be considered as a crucial issue since passengers need not miss their flights. This issue makes the mode choice to access the airports important to study on and develop policies regarding it. Many studies show destination type as domestic or international affects the airport access mode choice, along with other factors. In this study, we investigate the effect of destination type of mode choice using mixed logit, using market segmentation approach. Market segmentation regarding destination type as domestic or international is a first in airport access mode choice modeling. Revealed-preference data was collected by face-to-face passenger surveys at Ataturk International Airport in Istanbul, Turkey, in 2015. We did market segmentation analysis for Multinomial Logit (MNL) and Mixed Logit (ML) models. When MNL and ML models were compared, it was observed that ML was superior to MNL. Further, results of market segmentation analysis revealed that using segmented models produced more accurate results than using the pooled model; both in MNL and ML. This finding was also supported by the value of time estimates; there were significant differences between domestic and international travel markets in terms of airport access mode choice. These results showed that different transportation policies may be introduced for domestic and international traveler segments, which also were explained.

由于旅客不必错过航班，因此进入机场可被视为一个关键问题。这个问题使得进入机场的模式选择对于研究和制定有关政策至关重要。许多研究表明，目的地类型是国内还是国际都会影响机场访问模式的选择以及其他因素。在这项研究中，我们使用市场细分方法来研究使用混合logit的模式选择的目的地类型的影响。关于目的地类型为国内还是国际的市场细分是机场进入模式选择建模中的第一项。显示偏好数据是通过2015年在土耳其伊斯坦布尔阿塔图尔克国际机场进行的面对面乘客调查收集的。我们对多项Logit（MNL）和Mixed Logit（ML）模型进行了市场细分分析。当比较MNL和ML模型时，可以发现ML优于MNL。此外，市场细分分析的结果表明，使用细分模型比使用汇总模型产生的结果更准确；无论是在MNL和ML中。时间估计值也支持这一发现；在机场准入​​方式选择方面，国内和国际旅行市场之间存在显着差异。这些结果表明，可能会针对国内和国际旅行者细分市场采用不同的运输政策，并对此进行了解释。时间估计值也支持这一发现；在机场准入​​方式选择方面，国内和国际旅行市场之间存在显着差异。这些结果表明，可能会针对国内和国际旅行者细分市场采用不同的运输政策，并对此进行了解释。时间估计值也支持这一发现；在机场准入​​方式选择方面，国内和国际旅行市场之间存在显着差异。这些结果表明，可能会针对国内和国际旅行者细分市场采用不同的运输政策，并对此进行了解释。