Short-Term Forecasting of Railway Passenger Flow Based on Clustering of Booking Curves

Abstract

For railway companies, the benefits from revenue management activities, like inventory control, dynamic pricing, and so forth, rely heavily on the accuracy of the short-term forecasting of the passenger flow. In this paper, based on the analysis of the relevance between final booking amounts and shapes of the booking curves, a novel short-term forecasting approach, which employs a specifically designed clustering algorithm and the data of both historical booking records and the bookings on hand, is proposed. The empirical study with real data sets from Chinese railway shows that the proposed approach outperforms the advanced pickup model (one of the most popular models in practice) during the early and middle stages of booking horizon when bookings are not concentrated in the final days before departure.

Cross-Industry Statistical Law: Average Booking Curves Draw Exponential Functions

“…The other booking curve is represented as a "gross" demand-based booking curve (on-hand-based booking curve generally). The latter is often used in studies for RM [3,16,18,26,28,12,10,24,19]. Each of these definitions is divided according to whether or not canceled bookings are included in the time series; the former does not include them (that is, aggregate only consumed bookings), The cumulative number of bookings: X(t)…”

“…The most traditional and representative forecasting model is the exponential smoothing (including its expansions) [36,37,38,22,23,16,12,24,25]. In recent studies, various approaches, such as stochastic process models (including Poisson processes, negative binomial processes, generalized linear mixed models,), neural nets, pick up algorithms, and advance booking models have been proposed [4,3,16,18,26,28,12,10,19]. Since some of them are versatile, they spread in various perishable assets industries with some vertical advances such as dealing with the peculiar parameter in each field.…”

“…Therefore, the use of booking curves for forecasting in RM has been widely studied by many researchers and professionals in the perishable assets industries. This has led to the development of several advanced and highly accurate forecasting models [16,17,18,4,9,11,10,19,20,14,21], including exponential smoothing [22,23,24,25], stochastic process models [3,26], generalized linear mixed models [27], and methods utilizing neural nets that incorporate seasonal and day-of-week trends into their parameters [28,29].…”

Passenger demand forecasting in scheduled transportation

“…There are also some non conventional approaches as well, that are worth considering. Ma et al (2014a), for example, has used a clustering algorithm to generate a booking curve forecast. Tsai (2014) has taken a very similar approach but used case based reasoning to forecast a booking curve from initial booking.…”

The Cross-Industry Universal Laws of Exponential Booking Curves

“…Accordingly, in service industries, purchasing products with reservations has become common. With the spread of online reservations, the booking curve, which is the concept of the time series in the cumulative number of reservations and has been used for sales optimization in the airline ticket and hotel industries, has been used in various industries [2][3][4]. Booking curves in specific industries have been studied, but a universally applicable model across various industries has not been developed [5][6][7][8].…”