- Maps: Folder with the illustration of optimum routes in map.
- Screenshots: Folder with screenshots of optimum routes' map plotting.
- Coordinates&CargoUnits.xlsx: File that contains all the necessary data.
- Links.txt: Useful links to related websites and repositories.
- OptimumRouting.py: Python code that implements optimum route searching.
- Dependencies.txt: File with all Python site packages that have been used.
This project aims to find the optimum, in terms of driving distance, route for a vechicle which collects specific cargo units from different load points in the city of Heraklion, Greece.
The vechicle has a limited cargo capacity of 50 units. Having a depot as start, it is called to collect cargo units through different load points. It is only allowed to load the whole (and not a portion) point's cargo, but there is the option of transfering of units from one point to another. So, they can be gathered in specific spots and be loaded in the vechicle after it has passed through all loading points. Regarding loading points, it must be noted that they do not only provide different units between each other, but also their own cargo changes chronically.
A set of loading points' coordinates and their cargo units in two different days is given in .xlsx file.
Python code has been developed to find both the minimum number of routes and the minimum length of those and plot them in Herklion's map figure. The algorithm that was used to find the optimum routes is Dijkstra's Algorithm. The packages that have been used are enlisted below:
openrouteservice library provides its users with an easy access to Openrouteservice routing data via its APIs. The user can obtain an API Key through a free registration in Openrouteservice, when helpful directions of API's usage and posibilities can be found here.
Installation using pip:
pip install openrouteservice
Note:
It is possible that if an amount of requests per period of time is exceeded, openrouteservice will raise a warning. For more information, consult openrouteservice-client GitHub.
osmnx package allows the analysis and processing of real-world networks and geospatial geometries through the usage of OpenStreetMap data. The full documentation of osmnx can be found here.
Installation using pip:
pip install osmnx
itertools package is a set of tools that provides useful iteration and algebra functions such as count() or map(). There is no need of manual installation as it comes pre-installed in Python environment. For more information about this package, see its documentation here.
The networkx library is related to the analysis and manipulation of complex networks attributes and characteristics. For more information, follow this link.
Installation using pip:
pip install networkx
folium package provides its users with the posibility of manipulating and visualizing data with the utilization of leaflet.js library's mapping dynamics. For package's documentation, click here.
Installation using pip:
pip install folium
pandas is a useful and flexible open source tool that is related to data analysis and manipulation. For its documentation, follow this link.
Installation using pip:
pip install pandas
numpy library is directed to provide numerical computing posibilities through Python. The documentation of `numpy' can be found in NumPy's official website.
Installation using pip:
pip install numpy
Optimum routes for both days are plotted in html files in Maps folder. Each day corresponds to 2 different routes:
- Original route, which represents the driving course of the vechicle from depot to loading points.
- Returning route, which occurs after the vechicle has gone through all loading points and starts transfering cargo units back to depot.
For instance, the original route of the first day is displayed in the image below. Each loading point is represented by a marker, in which each point's ID and position within the route (1st, 2nd, etc.) are shown.
 |
|---|
| 1st Day's Original Route |
 |
|---|
| 1st Day's Returning Route |