Automated fiber network design with Comsof Fiber in QGIS.  

Hello and welcome to this short demonstration of Comsof Fiber running on QJS. So what we're looking at is QJS, the open source software where Comsof Fiber is running as a plugin. So all of these icons in the third bar, that's the functionality of Comsof Fiber, which I'll be showing a bit of today. Of course, there's lots more advanced stuff, but that's not the scope of this short demo. And I'm looking at an area in the United States, more specifically Washington. And I have a set of JS data here on the left-hand side. So all the data is showing is geographical information data, but I will be making an abstraction for calculation purposes. So just hiding all of the streets and parks and so on that we don't really need, just focusing on the data that we do need for Comsof Fiber calculation. That is the street center lines. So the back lines running down the middle of the street and then locations of clients, which are the green points typically located at the front of the buildings. So Comsof Fiber has a number of data enrichment operations of which I've now just started the first one, which is the creation of possible trenches. Possible trenches are the locations where a trench could be dug and a dug cable could be placed into. So as you can see, the data that we initially had, the street center line data has now been doubled. Cable could be placed on both sides of the street. This is all done automatically. So no manual intervention necessary. This is based on the analysis of the street data as such. So this is the first operation, which I launched in the data processing stage. There are quite a number. I won't run through all of them. I'll just show the results. What is important for the mall points or client locations is that they are connected to the streets. So we have the ability to create these drop trenches as I've just added them on the map. You also need to have places to cross the streets. So you see that now a number of additional lines have been added that cross the streets over the whole area. And then there's a specific set of data that we also use for aerial. So on the left hand side, there are a number of poles and a number of poles and aerial cables and also existing pipes. But that's part of a larger demo. And on the right hand side, I have a special type of cable that could be placed. That's the facade cable, which will be placed on the front of the houses. So all of this data enrichment is done automatically with steps that take very little time. Making sure that the JS data that's given as input is enriched, analyzed, prepared, and made ready for a comms-servoir calculation. And as a last step before any calculation, I need to run a process input that really goes and checks, is all the data connected? Can all the clients be reached? And is the data ready for a comms-servoir calculation? There are no warnings right now. So everything is fine. That's the first bit, the geographical information. Then the second bit of preparation is the rules. Because this is a number of settings that define the architecture for any comms-servoir calculation. So it's quite an extensive UI, meaning that there are lots of options to configure and make sure that any network architecture can be modeled. For now, I'll just focus on the, or show a couple of the costs that can be set. Because costs is the important decision criteria that comms-servoir will be using. For example, a standard trench has a cost of $40 per meter. A standard crossing has a cost of $50 per meter. And there are lots and lots of other parameters that can be set. And the costs for these specific categories will be used for making decisions of, do we place a cable down one street or down the next street, depending on the cost for each of these streets. Then for each of the three layers. So we're creating a three layer architecture consisting of drop, distribution, and feeder. For each of the layers, there are a number of duct definitions and cable definitions that specify how the connectivity is reached. And then also a number of other configurations. So all of this combines into a rules configuration that decides how Comsoft Fiber will create the routes. So with that preparation done, I will start the calculation. Right now, all the GIS information is taken into account. The rules configuration that we've just looked at is also taken into account. And this is all run through the extensive optimization engine that Comsoft Fiber is. Decisions are being made of where to place the drop points, where to place cabinets, what is the best location to place ducts and cables, and so on. Everything is done in a very short time. As you can see, the calculation is now finished. And we see the same area, but now with additional information added. I will zoom in a bit. For example, here we will see that these two houses get a cable that goes to the street right there. If I click on this line, I can see it is a drop cable is being created that has two fibers. It is an underground cable with a certain length that goes from the house to the street. And then from the drop point, there will be another cable, a distribution cable, going up to the north, crossing the street here, and then further along to the distribution cable right there. It is again a two fiber cable with a bigger length of 234 meters. Of course, when you look at the cluster, so that's this attribute, sorry, this object here on the left hand side, this is all, these are all the homes served on this specific area. Distribution point. And you can highlight all the cables that start from this distribution point or cabinets towards all of these clients. So everything is in place to create network, all of the routes, the ducts, the cables, and the fibers and so on are all in there. And this is done for the full area. So I'll just hide a bit of detail to show it also in the higher layers, this is being configured. So we have six clusters, all each represented by a different color, and then in the middle, a cabinet with a green dot. And all of these are served with feeder cables. So these feeder cables pass different distribution cabinets. I'll go back to the central office on the right upper side. And these are typically cables with a higher fiber count. So here we have a 48 fiber cable. That's again on the ground with a length of a bit over a kilometer. So all of the geographical data is present to start creating the lower level design of the network. And then I'll be showing the second part of the output of any calculation, which is the overview of the costs. So ComsoFiber generates a bill of material with any calculation. That is an Excel sheet containing some statistics and overview of the costs incurred to create this network. So the first tab in the Excel sheet contains, for example, the number of homes, 871, number of demand points. So we have a couple of NDUs, mainly single dwelling, but a couple of multi-dwelling units as well. And of course, the cost, total cost of this project is $1.3 million. And that is then subdivided into cost of deployment, cost of activation on the one hand, but also in the cost for the different layers. With a nice visualization in the pie chart. So in this case, the drop layer has the most costs. And then you can go into more details to see where these costs are coming from. So if I go to the overview tab here, the costs visible in this sheet are the input costs that we gave in the rules dialog or the architecture in the beginning of the demo. So the unit costs in this column show the actual input that we gave, but that is then combined with the volume. So for example, the first line is the number of poles that are being reused. There are 19 poles reused for a cost of $50 per pole, total cost of $950 and so on. So each of the cost items is listed here with the number of times that they occur, the unit price, and then the total cost incurred that way. Looking again at the drop layer that we were seeing in the previous sheet was quite high in cost. We see that the combination of the drop closures and drop boxes is incurring quite high cost. So based on these numbers, I can go back and look, is this configuration the right one? Can we lower the cost for this specific component to have a lower total project cost? Or is it the fact that the vendor is just quite expensive and we can do the same, but with the vendor that has a lower price. So each of these costs is quite visible here. Also for the higher layers, for example, distribution, we can see how many cables are being used areally, so cables not underground, and then cables that are underground with the different fiber counts, number of cabinets that are existing, number of cabinets that are new. This is all here with the amount of times this equipment is being used and the cost in the total cost column. So this is all very nice and allows you to dig into the costs that are caused with the current configuration. And given the speed of Comso Fiber, it's really easy to do other calculations and then compare. So going back to the overview here, and just to summarize, what we did is start from the GRS information, which was demand points and streets. Within a very brief time space, we were able to create additional data that we need for a network calculation. We ran that calculation itself in under a minute, get results both from a geographical point of view and a costing point of view. And this shows the power of Comso Fiber with the analysis you can do in a very short time. As mentioned in the beginning, this is a very small dataset. Comso Fiber is able to handle quite larger ones as well. And of course, this is a short demo with only the basic features being shown. There are quite a lot of other features in here as well. If you want to see more, don't hesitate to contact us and get more personalized demonstrations. Thanks for watching this demo and have a nice day.