Station and line-based passenger demand forecasting for Yenikapi M1-Kirazli M1 line and prioritization of critical success factors in headway determination in railway systems with multi-criteria decision-making methods

The rapid increase levels of the population and the number of motor vehicles brought about the transportation problem today. This problem has increased the importance of passenger demand forecasting studies in order to minimize passenger waiting times at stops and increase passenger satisfaction. In this study, unlike other studies in the literature, demand forecasting for Yenikapı M1 – Kirazlı M1 metro line is made in two stages, on a line basis and on a station basis. Station-based passenger demand forecasting is made using artificial neural network and machine learning (ML) algorithms and error values (MAE, BIAS, MSE, MAPE and RMSE) are compared. In the second stage of the study, passenger demand forecasting is made on a line basis with statistical techniques such as regression analysis and simple average; mean absolute percent error values are calculated and these values are compared. As a result of the study, it is seen that the most successful and reliable results for station-based demand forecasting are obtained with the decision tree, which is one of the ML algorithms; the best demand forecasting results on a line basis are obtained with the simple average method. As a second study, the prioritization of critical success factors in determining the headway for railway systems is discussed. While doing this study, the main criteria and decision hierarchy of decision alternatives SWARA, TOPSIS and AHP are revealed; for the other steps of AHP, firstly, pairwise comparison matrices are created. AHP pairwise comparison matrices are presented to expert engineers in Excel tables; based on the opinion of experts, the SWARA, TOPSIS, AHP approach is applied to the critical success factors model in headway optimization. As a result of AHP and TOPSIS model, the working rule alternative is found as the primary critical success factor with the weight of 0.197.Similarly, the main criteria are prioritized by the AHP and SWARA method, and it is seen that the reliability is the first priority main criterion. According to the findings obtained in this research, it was concluded that the lyophilization as powdered minimized the antioxidant activity loses rather than drying as whole or drying in oven.