Search

Effective winter road maintenance relies on precise road friction estimation. Machine learning (ML) models have shown significant promise in this; however, their inherent complexity makes understanding their inner workings challenging. This paper addresses this issue by conducting a comparative analysis of road friction estimation models using four ML methods, including regression tree, random forest, eXtreme Gradient Boosting (XGBoost), and support vector regression (SVR). We then employ the SHapley Additive exPlanations (SHAP) explainable artificial intelligence (AI) to enhance model interpretability. Our analysis on an Alberta dataset reveals that the XGBoost model performs best with an accuracy of 91.39%. The SHAP analysis illustrates the logical relationships between predictor features and friction within all three tree-based models, but it also uncovers inconsistencies within the SVR model, potentially attributed to insufficient feature interactions. Thus, this paper not only showcase the role of explainable AI in improving the ML interpretability of models for road friction estimation, but also provides practical insights that could improve winter road maintenance decisions. [ABSTRACT FROM AUTHOR]

Basal gravel and sand mantling the bedrock floors of buried valleys throughout the Canadian Interior Plains, and conformably overlying proglacial lacustrine sediment, comprise the Empress Group. While previously conceptualized as stratigraphically equivalent deposits of preglacial rivers prior to the first and most extensive continental and montane glaciations, subsequent stratigraphic studies indicated that buried valley basal gravel must have been deposited between, or during, progressively more extensive continental glaciations and could not be stratigraphically equivalent throughout the buried valley network. However, in the general absence of formation-rank stratigraphic description of basal gravel units that might better inform the geologic history of the deposits, most workers simply consider Empress Group sediments time-transgressive. In this paper, we examine basal gravel at provincial and regional scales to understand its genesis and geologic history. At the provincial scale, we map basal gravel in three dimensions using a novel machine learning approach. At the regional scale, we formally define basal gravel formations at either end of the largest buried valley system in Alberta, which informs its glacial history and physiographic development and shows the importance of formation-rank stratigraphic description. Our results indicate that the buried valley network across Alberta is palimpsest in genesis and basal gravel units within it are chronostratigraphically intercalated between tills. We advocate that the Empress Group definition be extended across Alberta with modifications to improve its clarity and utility, and formally define the Old Fort, Unchaga, Ipiatik, and Winefred formations as part of the Empress Group. [ABSTRACT FROM AUTHOR]