Forecasting the Short-Term Passenger Flow on High-Speed Railway with Neural Networks

Abstract

Short-term passenger flow forecasting is an important component of transportation systems. The forecasting result can be applied to support transportation system operation and management such as operation planning and revenue management. In this paper, a divide-and-conquer method based on neural network and origin-destination (OD) matrix estimation is developed to forecast the short-term passenger flow in high-speed railway system. There are three steps in the forecasting method. Firstly, the numbers of passengers who arrive at each station or depart from each station are obtained from historical passenger flow data, which are OD matrices in this paper. Secondly, short-term passenger flow forecasting of the numbers of passengers who arrive at each station or depart from each station based on neural network is realized. At last, the OD matrices in short-term time are obtained with an OD matrix estimation method. The experimental results indicate that the proposed divide-and-conquer method performs well in forecasting the short-term passenger flow on high-speed railway.

Improvement of technology for management of freight rolling stock on railway transport

“…The experiment performed in [13] demonstrated high accuracy of prediction at short-term forecasting of passenger traffic. The result of prediction can be used for planning cargo operations and for managing revenues of the industry.…”

Türkiye Demiryolu Yolcu Taşıma Talebinin Tahmini

“…Bağlantılar yoluyla ara katmana gelen bilgiyi karıştırır ve değişkenlerin yeni özelliklerini belirler. Çıktı katmanı tahminleri oluşturur ve hatayı yayar [14][15]1].…”

“…Yolcu talep tahminini, ulaştırma sistemlerinin düzenli çalışması, işletme planlaması, yolcu yoğunluğunun yönetilmesi ve gelir yönetiminin kontrol altında tutulmasını sağlamaktadır [1]. Demiryolu yolcu taşımacılığında daha iyi hizmet sunmak için hat planlama ve zamanlamanın tahmin edilmesi, yolcuların seyahat seçme ölçütlerinin belirlenmesi ve uygun yolcu akışını verecek modeller oluşturulmalıdır [2].…”

“…Xie vd. [1] çalışmalarında, bir güzergâh üzerinde numaralandırdığı istasyonlara gelen ve ayrılan yolcu sayısına ait geçmiş verileri kullanarak kısa vadeli planlamada yapay sinir ağlarından faydalanmışlardır. Önerdikleri DC (Divide and Conquer) yöntemi ile kısa vadeli yüksek hızlı tren yolcu akışının tahmininde başarılı sonuçlar oluşturduğunu belirlemişlerdir.…”

Exploring Intercity Trip Patterns of Railway Systems on National Holidays Using Deep Auto-Encoder

“…In some studies, direct demand models, which can be estimated by using aggregate data and require relatively low development costs, are adopted for specific Origin-Destination (O-D) pairs or simpler networks (6)(7)(8). For short-term ridership forecasting, time-series approaches are commonly applied (9)(10)(11)(12)(13), and in more recent years, some studies also employ neural networks or hybrid models that combine neural networks and time-series approaches, seeking to account for the nonlinearity among ridership fluctuation and thereby further enhance prediction accuracy (14)(15)(16). By contrast, probably because of the lack of intact data and representative samples, trip patterns for railway systems on national holidays are rarely discussed in the literature.…”