Freddy Hernández-Barajas

1. Introduction The transportation sector on a global scale is a major contributor to pollution, carbon emissions, and dependence on finite fossil fuels [1,2,3,4]. In response, governments are implementing policies to enhance energy security, reduce emissions, and promote sustainable mobility [1,4,5]. The rapid expansion of the electric vehicles (EVs) market underscores this transition, with projections that EVs will account for approximately 57% of global vehicle sales by 2040 [3].

1 Introduction Although linear regression is commonly used to model numeric dependent variables (NDVs), this classic technique relies on the assumption that the distribution of the residuals has a normal shape. Because the shape of the residuals reflects the shape of the NDV, and NDVs often exhibit non-normal distributions in practice Bono et al. (2017), it is likely that estimates from linear models would be biased.

5. Results In this section, we outline the implementation of the proposed methodology using the NASA dataset. provides a brief overview of the data, followed by with a statistical description of the features utilized for RUL prediction. Subsequently, in , we detail the methodology implementation, including the identification of degraded data, feature smoothing, and RUL prediction. provides a comparison of RUL estimation models.