ESR13: Using truck sensors for road pavement performance investigation

Hi crew! Today an amazing post introducing the Random Forests method is waiting for you to be read! Yep, I know, I promised this post a long time ago but seriously crew: I had no time for it! Random Forest is a machine learning method mainly used for classification and regression purposes. These kind of statistics basically learns from a training set of data and gives answers associated to new data based on what the algorithm learnt. Yep crew it sounds very similar to what Artificial Neural Networks do. In fact, also Artificial Neural Networks is a machine learning method used for regression analysis. However the two methods are based on slightly different concepts. From the historical point of view, the first concepts behind the theory of "Random Forests" are introduced for the first time by Ho in 1995 [2][3], but it is only in 2001 that they are defined as we currently know them by Breimann [1]. The algorithm itself is based on the theory of Decision Trees … ...

Hey crew! Are you ready for a new post?! Here we go then! Today I’m going to introduce you to a more technical topic. In fact, we are going to talk about machine learning and Artificial Neural Networks (ANNs) in particular. Machine learning is a subfield of computer science which aims to give computers the ability to do something without being explicitly programmed for doing that. Originally, it comes from the study of pattern recognition and computational learning theory in artificial intelligence. Exploring the study and construction of algorithms that can learn from experience (historical data), the algorithm operates by building a model from example inputs in order to make data-driven predictions or decisions. ANN is just a branch of Machine Learning. They are data processing paradigms inspired by the way the biological nervous system process information in human beings (Biological Neural Networks, BNNs). Usually they are used to estimate or approximate functions t...

Hello everybody! I hope you spent nice holidays and you came back to work cheerful and full of energy! Today, as promised, I’m going to explain a bit more about the data we collect and about the type of analysis we intend to perform. As I previously explained , the objective of the project is: to assess the relationship between truck fleet fuel consumption and road pavement conditions. Data about the fuel usage are collected by Microlise Ltd whilst data about road pavement conditions are collected by TRL Ltd (but owned by Highways England ). In both cases, data are stored in very big databases which can be remotely queried whenever needed. This study is going to consider only data about trucks travelling along the English SRN (Strategic Road Network). The SRN is completely managed by the Highways England authority. It is around 7’000 kilometers long and is mostly made of motorways and ‘A’ roads. Although the SRN network represents just 2% of the length of all roa...

Hello crew, how is it going? “R” you ready for a new post? Today I’ll introduce the software that I’m using for the data analysis in my project. The name of the software is “ R ”. Have you ever heard about it? Yes, when I arrived in NTEC I knew it just because my brother uses to work with it. My brother is a mathematician (strange people, but I love my brother anyway). By the way, I knew how to handle computer programming languages such as MATLAB® , C++, Python and Microsoft® Visual Basic for Applications (VBA)so, for this reason, it was not a problem to learn a new one. Futhermore, R is quite easy to learn and its syntax is very similar to Python. Yes, it can be confusing  at times because it is possible to write certain commands in another language instead of the one that you are using, but basically once you know how to handle one of them, you know (almost) all. Just change the syntax of the command or the keyword which recall a specific function that you need. ...