I have completed, or sufficiently optimized, my work with the training data (An arbitrary 80% of roadways from the study), and with that comes joy accompanied by ambition for further improvement— a sentiment shared by the general public, who now have an excuse to stay at least six feet away from me at all times.
I’m also re-reading a similar paper that addresses climate as a factor in pavement damage, though the heavy lifting seems to have been done by traditional statistical analysis as opposed to machine learning, which makes for some difficulty in understanding the isolation of a single variable affecting the pavement. It’s also a voluminous 260 pages long, and while it maintains density akin to that of a shorter paper, I’m only looking for a couple needles in a veritable haystack.
Even if I do manage to come up with some percentage of damage caused by load bearing vehicles, I sometimes wonder how that information would be implemented to make the world a better, or at least a more efficient, place. Something I need to consider in the following weeks.