Forecasting the carsharing service demand using uni and multivariable models

Abstract

Carsharing is ana lternative to urban mobility that has been widely adopted recently. This service presents three main business models: two of these models base their services on stations while the remainder, the free-floating service, is free of fixed stations. Despite the notable advantages of carsharing, this service is prone to several problems, such as fleet imbalance due to the variance of the daily demand in large urban centers. Forecasting the demand for the service is a key task to deal with this issue. In this sense, in this work, we analyze the use of well-known techniques to forecast a carsharing service demand. More in deep, we evaluate the use of the Long Short-Term Memory (LSTM) and Prophet techniques to predict the demand of three real carsharing services. Moreover, we also evaluate seven state-of-the-art forecasting models on a given free-floating carsharing service, highlighting the potentials of each technique. In addition to historical carsharing service data, we have also used climatic series to enhance the forecasting. Indeed, the results of our analysis have shown that the addition of meteorological data improved the models’ performance. In this case, the mean absolute error of LSTM may fall by half, when using the climate data. When considering the free-floating carsharing service, and prediction for the short-term (i.e., 12 hours), the boosting algorithms (e.g. XGBoost, Catboost, and LightGBM) present superior performance, with less than 20% of mean absolute error when compared to the next best-ranked model (Prophet). On the other hand, Prophet performed better for predictions conducted on long-term periods.

Prediction of Bike‐sharing Trip Counts: Comparing Parametric Spatial Regression Models to a Geographically Weighted XGBoost Algorithm

“…In GAM, spatial coordinates can be directly included in the model estimation, whereas XGBoost does not offer this option. It has been applied to model temporal sharing data numerous times (Sathishkumar, Park, and Cho 2020; Yang et al 2020b; Alencar et al 2021), but in this study could be adapted to spatial data through the additional estimation of a GWR to account for spatial effects. The estimation of a GWR is a common technique to deal with spatial autocorrelation, but usually not in combination with XGBoost (Bao, Shi, and Zhang 2018; Ji et al 2018; Wang et al 2020b).…”

“…XGBoost ranks among the best models regarding computation time and symmetric mean absolute percentage error (sMAPE), while an LSTM leads to slightly better RMSE values. Alencar et al (2021) evaluate the qualities of LSTM for temporal modeling of car sharing demand to several other modeling approaches including XGBoost. Again, different methods perform best for different tasks and regarding different performance measures (MAE and RMSE).…”

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction

“…The authors of ref. [ 10 ] evaluated the use of Long Short-Term Memory (LSTM) and Prophet techniques for predicting the demand for car-sharing services. Furthermore, the authors of ref.…”

“…⊗ denotes the Hadamard product, W p , W D , W W , and W M are the weight matrices of the time scale-related layers, and b sp is the bias. The output of the temporal feature module is defined according to Equation (10).…”

A Space-Time Model for Demand in Free-Floating Carsharing Systems

“…Tey employed long short-term memory (LSTM) structure to forecast short-term future vehicle uses. Alencar et al [39] evaluate seven stateof-the-art forecasting models on a given free-foating carsharing service, highlighting the potential of each technique. Te assessed models include ARIMA and SARIMA, prophet, variants of boosting algorithms, and long short-term memory (LSTM).…”