https://github.com/cyang-kth/maximum-coverage-location
A Python library for solving maximum coverage location problem
https://github.com/cyang-kth/maximum-coverage-location
facility-location gis integer-programming location-services optimization
Last synced: 3 months ago
JSON representation
A Python library for solving maximum coverage location problem
- Host: GitHub
- URL: https://github.com/cyang-kth/maximum-coverage-location
- Owner: cyang-kth
- License: mit
- Created: 2019-11-22T01:31:55.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2019-11-22T12:06:28.000Z (over 5 years ago)
- Last Synced: 2025-03-18T04:51:20.795Z (3 months ago)
- Topics: facility-location, gis, integer-programming, location-services, optimization
- Language: Jupyter Notebook
- Homepage:
- Size: 649 KB
- Stars: 60
- Watchers: 2
- Forks: 15
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE.TXT
Awesome Lists containing this project
README
## Maximum coverage location problem (MCLP)
This repository provides a Python implementation of solving a classical instance of
the **maximum coverage location problem** described in Church 1974.
The problem is defined as: given **N** points, find **K** circles with radius of **r**
to cover as many points as possible.
- Example 1: Select 20 circles with radius of 0.1 to cover 300 points (uniform distribution)
(M is the number of candidate sites and C is the number of points covered)
- Example 2: Select 20 circles with radius of 0.2 to cover 300 points (moon distribution)
### Problem formulation
The method randomly generates a set of candidate sites within the region
of the input points. The problem is then solved by integer programming.
The mathematical formulation is given below:
### Demo and usage
```
from mclp import *
import numpy as np
Npoints = 300
from sklearn.datasets import make_moons
points,_ = make_moons(Npoints,noise=0.15)
# Number of sites to select
K = 20
# Service radius of each site
radius = 0.2
# Candidate site size (random sites generated)
M = 100
# Run mclp
# opt_sites is the location of optimal sites
# f is the number of points covered
opt_sites,f = mclp(points,K,radius,M)
# Plot the result
plot_result(points,opt_sites,radius)
```
Check the jupyter-notebook [demo.ipynb](demo.ipynb).
To run the example interactively, inside the project directory type the command
```
jupyter-notebook
```
### Requirements
- Python 2.7
- Scipy, Numpy (available as part of [Anaconda](https://www.anaconda.com/))
- [Shapely](https://github.com/Toblerity/Shapely)
- [Gurobi](https://www.gurobi.com/), commercial software (free for academic usage)
It is recommended to use Anaconda directly, where the packages can be installed with `pip` or `conda`.
```
pip install shapely
conda config --add channels http://conda.anaconda.org/gurobi
conda install gurobi
```
### Contact
Can Yang, Ph.D. student at KTH, Royal Institute of Technology in Sweden
Email: cyang(at)kth.se
Homepage: https://people.kth.se/~cyang/
### Reference
- [Church R, ReVelle C. The maximal covering location problem. Papers in regional science, 1974, 32(1): 101-118.](https://link.springer.com/article/10.1007/BF01942293)
- http://examples.gurobi.com/facility-location/