Predição do erro no comportamento ecológico dos motoristas em função das condições térmicas das cabines de ônibus urbano

Abstract

The constant increase in fuel prices and natural disasters due to global warming causes interest in research aimed at the economics of fossil fuels and the reduction of polluting gases from these fuels. In view of this context, this thesis aimed to predict the error in the ecological behavior of drivers as a function of the thermal conditions of urban bus cabins. This error was translated by the proportion of time that drivers cannot keep the bus speed within the ideal cruising speed limits (35km/h≤Ve≤50km/h) on urban roads. The thermal conditions were transcribed by the environmental and personal parameters, by the Predicted Mean Vote (PMV) and equivalent temperature from the thermal comfort of ISO 7730 (2005) and ISO 14505-2 (2006), respectively, and by the subjective votes of thermal sensation, evaluation and preference from ISO 10551 (2019). The field research lasted 60 days, from September to December 2021, and observed 3 daily trips by 3 drivers on the same bus route. Data analysis was performed using the Generalized Linear Models methodology to generate and evaluate the consistency of ordinal logistic regression models for subjective votes, and Beta regression for error in ecological behavior. Therefore, the environmental parameters proved to be significant to the models of the subjective votes and, among these parameters, the air temperature was present alone or together with the air velocity in the best models of thermal sensation, evaluation and preference, whose Pseudos’ R2 were of 0.669, 0.467 and 0.437, respectively. Furthermore, the PMV and equivalent temperature indices can be considered excellent predictors (Pseudo R2 =0.63 and Pseudo R2 =0.631 respectively) of the thermal sensation perceived by bus drivers. The ecological behavior error models, when adjusted, resulted in 2 models with distinct variables and low explanatory power (Pseudo R2 =0.1785 and Pseudo R2 =0.1853). However, statistical tests have indicated that they can be used to assess some of the variability in the proportion of time drivers fail to keep the bus speed within optimal cruising speed limits. However, the variables that influenced the ecological behavior error of urban bus drivers were: environmental parameter, air temperature; the personal parameter, the metabolic rate; and the Fanger´s PMV index.