Machine learning?based stocks and flows modeling of road infrastructure

This paper introduces a new method to account for the stocks and flows of road infrastructure at national level based on material flow accounting (MFA). The proposed closes some current shortcomings in infrastructures that were identified through MFA: (1) insufficient implementation prospective analysis, (2) heavy use archetypes as way represent infrastructure, (3) inadequate attention inclusion dissipative flows, (4) limited coverage uncertainties. dynamic bottom-up MFA was tested Norwegian network estimate predict between 1980 2050. Here, supervised machine learning model introduced instead archetypical mapping different roads. dissipation materials from tire–pavement interaction incorporated. Moreover, this study utilizes iterative classified regression trees, lifetime distributions, randomized intensities, sensitivity analyses quantify