Clustering and Forecasting Urban Bus Passenger Demand with a Combination of Time Series Models

The current paper concentrates on the examination of extensive datasets derived from public transportation networks, specifically addressing the prediction of urban bus passenger demand. The approach involves a series of steps designed to enhance the comprehension of passenger demand. Initially, due...

Descripción completa

Detalles Bibliográficos
Autores: Mariñas Collado, Irene, Sipols, Ana E., Santos Martín, M. Teresa, Frutos Bernal, Elisa
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad Rey Juan Carlos
Repositorio:BURJC-Digital. Repositorio Institucional de la Universidad Rey Juan Carlos
OAI Identifier:oai:burjcdigital.urjc.es:10115/28340
Acceso en línea:https://hdl.handle.net/10115/28340
Access Level:acceso abierto
Palabra clave:forecasting
time series models
Big Data
Clustering
Cointegration
Combination
Descripción
Sumario:The current paper concentrates on the examination of extensive datasets derived from public transportation networks, specifically addressing the prediction of urban bus passenger demand. The approach involves a series of steps designed to enhance the comprehension of passenger demand. Initially, due to the substantial number of bus stops in the network, they are categorized into clusters, and distinct models are subsequently developed for a representative from each cluster. The objective is to compare and integrate predictions generated by conventional methods like exponential smoothing or ARIMA with those from machine learning techniques, such as support vector machines or artificial neural networks. Furthermore, the accuracy of support vector machine predictions is refined by incorporating explanatory variables with temporal structures and moving averages. Ultimately, through cointegration techniques, the outcomes obtained for the representative of each group are extrapolated to the remaining series within the same cluster. The paper illustrates the application of these methods through a case study conducted in the city of Salamanca, Spain.