Travel Time Penalty in Mins for the State of North Rhein Westfalen
The Background:
The last decade has seen an enhanced emphasis on promoting ÖPNV (Public Transport) as a step towards reducing the carbon footprint associated with the transportation sector (Zukunft des ÖPNV Study der Hans-Böckler-Stiftung). However, at the level of the individual consumer, using ÖPNV as opposed to a personal vehicle, comes with some concrete challenges:
The inflexibility associated with depending on ÖPNV, especially at night.
The higher costs associated with long-distance ÖPNV travel.
Unpredictable delays, especially in long-distance ÖPNV travel.
Longer travel time associated with ÖPNV within the city/town.
Relative inaccessibility of destinations that do not lie within the city boundaries with ÖPNV.
In this Application, we are trying to quantify the problem arising out of longer travel while using public transit in the State of North Rhein Westfalen, Germany. Additionally, we will also take a dab at quantifying the problem arising out of the reduced accessibility. But first things first.
The Map:
Below is a map of all the Gemeindes in the state of North-Rheinwestfalen of Germany. A Gemeinde is administrative unit that would correspond to a town or city. One can hover over a particular town and see the estimated travel time penalty. The travel time penalty is the estimated additional minutes that one needs when using Public transit versus a car, on average, when one is moving within the town/city. For example, one can see that the travel time penalty for Köln is 16.26 minutes ( not 16 minute and 26 seconds).
You can find the details of methodology in the section on Methodology below
Methodology
The calculation of the ÖPNV time penalty proceeds in three steps
Sampling of locations within the city/town.
Estimation of travel time between these locations for the two modes of transport.i.e when driving and when using ÖPNV (Public Transit)
Aggregating the differences in travel time to arrive at the average of time penalty in minutes
Step 1:
The points of interest for our analysis at the level of the town/city are individual post-boxes. In Post-boxes are generally located near population centers. Thus, we use the location of post-boxes as a easy proxy for population density. The coordinates of the individual post boxes are acquired from the Open Street maps API (OSM). OSM is also our source of all shape files used for the final visualization.
We use one fourth of the post_boxes of the town/Gemeinde as our points of interest. This is because the number of API calls increase in a quadratic fashion.
Step 2:
The travel time between our post-boxes are acquired from the Bing-Maps API service. The travel times are estimated for 11 AM on a weekday (specifically, 9 th of March,2022 for NRW). To ensure comparability between the different states ( an ongoing project) , the data acquisition for this step is always done for the specific time (11 AM on a weekday). This is not the most elegant solution. This is driven by the constraint imposed the usage of Bing Maps. The estimation of travel time using Public transit is possible only at the current local time. We hope that Bing Maps would add a feature for estimating travel time at a particular time point for public transit as well.
Step 3:
The difference between the travel time for driving and public transport gives us the travel time penalty (in minutes) for a particular route. The Travel time penalty for the whole city/town is then the average over all the routes.
Visualization:
The visulization is created using the Python packages Folium and Geopandas.