A well-known phenomenon that happens and characterizes any busy civilized cities is traffic jam. All around the world more and more people are using their car not only to travel long distances but on a daily routine, as it can be going to work, doing grocery, carrying kids around for school, sports, meet up with friends. Researches showed that on average there is one car for every 7 inhabitants on Earth, which means that the world park is nine-zeroed. Considering these numbers and the fact that usually the roads people choose to take are the most frequently used to head from a place A to a destination B, and that they do it in similar time slots (due to the fact that our lives are all scheduled with similar timing that society has imposed), it seems inevitable to encounter traffic at some point of the day. The main complain about high traffic is the waste of time for the users. According to the AAA Foundation for traffic Safety [4], on average an American citizen spends 50 minutes/day on driving. In this report we will focus our attention on the city of Houston, Texas. In particular we will analyze and try to improve its road network (working on its weaknesses), in order to provide the user with a better service. What we’re asked to do is either reducing/modifying the traffic or improving the physical network of routes (PNR), or both. By reducing/modifying the traffic we mean asking people to adapt their behaviour on the roads by using detours, decreasing their cruising speed, use traffic lights or put parking limitation and so on. As pointed out on the project description, in general in Europe, most cities try to act upon the traffic directly, for example by asking people to use bicycles instead of cars. The Netherlands is a perfect example of this type of approach: with almost a quarter of the population cycling every day, it is the country with the biggest number of bicycles per inhabitants [2]. But this is not always the best choice as it may seem at first: even if the Dutch economy is doing well, what we’ve been suggested to think about is that taking such decisions can be a slowdown for the economic activity besides restricting the freedom of the user (which is not recommended). The only option left would then be the PNR modification. To improve the physical network of routes means enlarging the ways, remove obstacles, add lanes, build new routes and so on. Obviously, such an operation would be more costly due to the fact that we’re dealing with the ’hardware’ of the system. As previously mentioned, the goal of this project is to reduce the traffic in the city of Houston, Texas. In order to do this, we have been provided with a set of data, characterizing the vehicles’ fluxes around all Houston’s highway system, on an ordinary day: January 29th 2018, between 8 a.m and 9 a.m. In particular, for each segment of highway, we’ve been given: its number of lanes, its speed limit, its in-going and out-going flux, it’s location in the system and it’s length. We will firstly focus our attention on the identification of those segments of highway where traffic jams are mot likely to occur. In order to perform such a research we will develop a mathematical model for the maximal flux (the MVF) and we will compare the obtained results with the actual flux for each section(provided by the data set). After identifying an locating traffic jams, we will be able to get a clear global picture that will allow us to properly analyze the data provided and how our model fits the data. Finally, after getting a full view of the nature of all the fluxes characterizing all the branches of Houston’s highways system, we will propose some modifications of the PNR which could help decreasing traffic. For this last step we can proceed in different ways; the most common and user-friendly would be to choose a small and relevant number of characteristics trips, which are among the most used ones and determine benefits for these trips only. Even if this seems to be the better approach (it limits the number of simulations to perform allowing us to focus on qualitative and representative sections and it’s easier to be understood by the public opinion), our analysis and the results it would lead, will stress some conclusions that will make us prefer a different method