diff --git a/README.md b/README.md
index 12bf23a..2a906c7 100644
--- a/README.md
+++ b/README.md
@@ -2,6 +2,8 @@
 ## Requirements
 - [x] [JDK 17](https://www.oracle.com/java/technologies/downloads/#java17)
 - [x] Python 3
+- [x] Input file holding the network of the city (For example: merged-network.osm.pbf)
+- [x] Input GeoJSON file with all buildings (For example: summerschool_all_buildings.geojson)
 
 ## Goal
 Estimate city pollution coming from people driving in vehicles.
@@ -12,18 +14,21 @@ Make sure to have JDK 17 installed and clone this repository.
 ## Usage
 The file `main.py` is essentially a replica from MEB workflow. It constructs a city and then allows the user to apply characteristics to that given city.
 
-The difference is in the implementation of MatSimEngine. This engine should create the following files:
-- facilities.xml
-- network.xml
-- population.xml
-- config.xml
+The additional steps from the MEB workflow are:
+1. Generate Matsim required xml files.
+2. Run Matsim analysis.
+3. Visualize Matsim results.
 
-Facilities and network are quite simple as the facilities file can be built from our city object and the network file can be built using a matsim utility along with OpenStreetMaps data. The population file must be generated via other ways and the config file can be generated once the first 3 files are created.
+### 1- Generate Matsim Required XML Files
 
-Additionally, it may be interesting to include public transport in the simulation. This is a potentially simple task. Montreal offers the GTFS files for its public transport for free and the project [PT2MATSim](https://github.com/matsim-org/pt2matsim) allows you to create the MatSim xml files from the initial GTFS files.
+This step is done in `matsim.py`. The class was built to replicate the behavior of other HUB export factory classes. It takes in a city object, an output directory, and then generates all the xml files required by Matsim.
 
-## Visualization
+### 2- Run Matsim Analysis
+
+This step is done in `matsim_engine.py`. It simply runs the matsim.jar while providing the config.xml file generated in the previous step.
+
+### 3- Visualize Matsim Results
+
+This step is done in `matsim_visualizer.py`. It takes in a network and events xml file generated by the previous step and then displays it on a graph. It then shows in heatmap style the change in traffic over time.
 
-Given we want to be able to visualize the results similar to the way the Munich scenario did it: https://www.matsim.org/gallery/munich.
 
-The repository also has the code to visualize these results. It's a very basic implementation, but it basically uses matplotlib and nx to plot the network and how much traffic exists on a given link at a given time.
diff --git a/__pycache__/matsim.cpython-39.pyc b/__pycache__/matsim.cpython-39.pyc
index 625f92b..105f345 100644
Binary files a/__pycache__/matsim.cpython-39.pyc and b/__pycache__/matsim.cpython-39.pyc differ
diff --git a/__pycache__/matsim_visualizer.cpython-39.pyc b/__pycache__/matsim_visualizer.cpython-39.pyc
index 3c2c327..fdc48a0 100644
Binary files a/__pycache__/matsim_visualizer.cpython-39.pyc and b/__pycache__/matsim_visualizer.cpython-39.pyc differ
diff --git a/main.py b/main.py
index 8559450..a2a0fee 100644
--- a/main.py
+++ b/main.py
@@ -25,10 +25,10 @@ try:
   ConstructionFactory(construction_format, city).enrich()
   UsageFactory(usage_format, city).enrich()
 
-  Matsim(city, 'output_files')._export()
+  Matsim(city, 'output_files').export()
   MatSimEngine('output_files/Montreal_config.xml').run()
 
-  visualizer = MatsimVisualizer('output_files/Montreal/output_network.xml.gz', 'output_files/Montreal/output_events.xml.gz')
+  visualizer = MatsimVisualizer('output_files/Montreal/output_network.xml.gz', 'output_files/Montreal/output_events.xml.gz', 'output_files')
   visualizer.visualize()
 
 except Exception as ex:
diff --git a/matsim_visualizer.py b/matsim_visualizer.py
index 55fd178..1a11b8c 100644
--- a/matsim_visualizer.py
+++ b/matsim_visualizer.py
@@ -95,7 +95,7 @@ class MatsimVisualizer():
     traffic_data = self._cumulative_traffic[tick]
 
     edge_colors = [self._cmap(self.norm(traffic_data.get(link['id'], 0))) for link in self._links]
-    edge_widths = [10 + self.norm(traffic_data.get(link['id'], 0)) * 10 for link in self._links]
+    edge_widths = [1 + self.norm(traffic_data.get(link['id'], 0)) * 10 for link in self._links]
 
     plt.cla()
     nx.draw(self._G, self._pos, node_size=0, node_color='blue', width=edge_widths, edge_color=edge_colors,