Group Members: Francisco Marmolejo, Maria Margariti, Diego Lopez Ibarra For our final assignment we decided to create a system where one can select their area of interest (anywhere in the world), which will provide them with the amount of solar hours in that area. Then they are ask to select the solar panel that is suitable for their needs. Thus, the final outcome would be a calculation of the amount of energy that they will be able to obtain with that specific solar panel in their area. We are in the process of establishing this system and this is our current progress. The project started by creating a spreadsheet with the different types of solar panels and their characteristics. This file was then created into a cvs file and was brought into the workspace with a “CsvLoader.” Then we created a “Scatter2” graph that would illustrate some of the important characteristics of the panels. This was done by getting the information from the csv file and applying a “GetSubListFromPositions“, which gathered the information from columns 2,7,8 in the spreadsheet and illustrated the graph with the y-axis = column 2 (Price), the x-axis = column 7 (watts), and the size of circles defined by column 8 (price per watts). This information was accompanied with another component, “GetElementFromList” , which was defined by getting information from column 1 (name of model). Thus, providing additional information when selecting a circle, one would see not only the price, watts, price per watts, but also the model that the other data corresponds to. Our next step was to illustrate the countries where these solar panels are fabricated. We selected the component “GetElementFromList“, which help us get the data from column 9 (countries where solar panels were fabricated). This information was then linked to “JoinStringList” that combined the information in one string and linked them with the “,”. Next, we linked the component “ReplaceSubString” that allowed us to replaced the “/” with “,” where there was more than one country to a model. Next, we linked the “SplitString” component to separate each word that was separated by “,” therefore having each word be isolated. Next, we decided to eliminate all the repeating elements from the list and find out how many different countries were fabricating the solar panels. We did this by linking the “RemoveRepetitions” component which then gave us the final number of countries. Finally, to visualize the final list we had to link the component “StringListToCountryList” to the “ColorMapVisualizator” illustrating the 12 countries that we have as fabricators of solar panels. The last part of our process till now was to link the the data achieved in the last step in obtaining the 12 countries and having a way of zooming in. This was accomplished by linking the data from the “StringListToCountryList” to “ListElementSelector“, which gave us a module to select 1 of the 12 countries. Next, we linked it to the “GeocoderGoogleMaps” that produces the coordinates to find the selected country in a map. Next, we applied the “GetElementFromTable” component that gets the coordinates from the table and links this information to the “GoogleMapsVisor” that illustrates the coordinates.