Regressão linear múltipla na identificação de fatores relacionados à demanda por serviços de compartilhamento de bicicletas

Abstract

Bike sharing systems provide several economic, climate and health benefits and therefore have become increasingly common. The objective of this study was to apply multiple linear regression analysis methods to identify the main factors influencing the demand for this type of service. In order to achieve the objective, the daily count of bicycles rented at the BikeIndia sharing provider, from the United States of America (USA), in the years 2018 and 2019 was considered as the dependent variable. The independent variables for the study were the season of the year, year, holiday period, working day, weekend, climate situation, temperature, wind chill, humidity and wind speed. The adjusted model showed a significant linear relationship between the rented bike count and the year, holiday, temperature, humidity, and wind speed variables. It was concluded that in 2019 there was a significant increase in bicycle rentals compared to 2018 and on holidays there was a reduction in rentals. In addition, there was a trend towards an increase in bicycle rentals with an increase in temperature, while the increase in humidity and wind speed tends to reduce the number of rentals.