In the RailNL case, we're trying to optimize the line planning of trains. The case contains two sets of CSV-files, one for the stations and connections in Noord- and Zuid-Holland, and one for the stations and connections of the Netherlands. The goal is to get for both areas a covering network with a maximum of maxLines and where the duration of every line is not longer than maxTime.
| Area | nStations | nConnections | maxTime | maxLines |
|---|---|---|---|---|
| Holland | 22 | 32 | 120 | 7 |
| Netherlands | 45 | 89 | 180 | 20 |
The quality of the solution can be measured using the following formula:
Where p is the percentage of the connections covered (between 0 and 1), #Lines is the number of lines used in the solution and TotalTime is the sum of the times of every line.
The code used is completely written in Python 3.8 and all packages needed to run this program can be found in requirements.txt. You can use pip to install these packages with:
python3 -m pip install -r requirements.txt
Or use conda:
conda install --file requirements.txt
To run this program you could use the function main.py that has four required input arguments and two optional. First the name of an area (-a) is needed. You could use "Holland" to calculate the lines for Noord- and Zuid-Holland and "Nationaal: to calculate the lines for the Netherlands. The second argument needed (-d) is the Maximum number of minutes that a train can ride on a line. The third argument (-L) is the number of lines the function needs to calculate. The final required input argument (-A) is the algorithm that is going to be used to create the lines. The algorithms that can be used are "random", "greedy", "hill_climber", "simulated_annealing".
The first optional option (-r) is the amount of runs an algorithm must be used. More runs might give higher scores. The second optional option (-i) is the number of iterations the Hill Climber uses in every run.
usage python3 main.py [options]
required options:
-a, --area Area run the algorithms for
-d, --duration Max duration for one line
-L, --lines Max no. of lines
-A, --algorithm Algorithm to run
optional options:
-h, --help Prints this message
-r, --repeat No. of repetitions
-i, --iterations No. of iterations per run
When filled in, it will look something like this:
python3 main.py -a "Holland" -d 120 -L 7 -A "random" -r 1000
When the function is started, it will return all the lines and number of minutes they take, a quality score of the lines combined, an output csv-file and an output map as png-file.
An example of terminal output of the above function is:
Click to see Terminal Output
Haarlem, Heemstede-Aerdenhout, Haarlem, Amsterdam Sloterdijk, Zaandam, Hoorn, Alkmaar, Den Helder
Duration 115 min
Den Haag Centraal, Gouda, Alphen a/d Rijn, Leiden Centraal, Heemstede-Aerdenhout, Haarlem, Beverwijk, Haarlem, Amsterdam Sloterdijk, Amsterdam Centraal
Duration 119 min
Alphen a/d Rijn, Leiden Centraal, Den Haag Centraal, Delft, Den Haag Centraal, Leiden Centraal, Alphen a/d Rijn, Gouda, Den Haag Centraal
Duration 115 min
Amsterdam Zuid, Amsterdam Sloterdijk, Haarlem, Amsterdam Sloterdijk, Amsterdam Centraal, Amsterdam Sloterdijk, Zaandam, Castricum, Beverwijk, Zaandam, Amsterdam Sloterdijk, Amsterdam Centraal
Duration 118 min
Rotterdam Centraal, Schiedam Centrum, Delft, Den Haag Centraal, Gouda, Rotterdam Alexander, Rotterdam Centraal, Schiedam Centrum, Delft, Schiedam Centrum, Rotterdam Centraal, Dordrecht, Rotterdam Centraal
Duration 119 min
Leiden Centraal, Heemstede-Aerdenhout, Leiden Centraal, Schiphol Airport, Amsterdam Zuid, Amsterdam Amstel, Amsterdam Zuid, Amsterdam Amstel, Amsterdam Centraal, Amsterdam Amstel, Amsterdam Centraal, Amsterdam Sloterdijk, Amsterdam Centraal, Amsterdam Sloterdijk
Duration 119 min
K-score 8337
sections traversed 27/28
7.4045667 s
Searching for map of Holland.
Map-Holland is created.
And an example of the output map will look like:
All the important folders and files are structured below:
- /code:contains all code.
- /code/algorithms: contains code to draw lines.
- /code/classes: contains all classes.
- /code/data_loader: contains code to load data.
- /code/visualization: contains code to plot results.
- /data: contains csv-files of RailNL case.
- /data/shapefile: contains shapefile to draw (parts of) the Netherlands.
- /output: contains (examples of) output-files.
- /output/plot: contains output-plots.
- Dani van Enk, 11823526
- Michael Faber, 6087582