In this project, we investigated the question of how to arrange given departments within a facility. The desired layout has a high efficiency for some production processes. The present problem focuses on the aspect of minimizing transportation costs between the departments. This can be mathematically modeled as a mixed integer optimization problem. As solving it at once using the standard solver for MIPs Gurobi has a very long running time, we implemented an algorithm that solves the optimization problem in two stages. In the first step, relative positions are determined for the departments. This is done by solving a nonlinear minimization problem. With these relative positions the MIP can be simplified enough to be solvable with Gurobi within a short time. We based this project on the book Facility Layout by Miguel F. Anjos and Manuel V.C. Vieira. For more information on the mathematical background, feel free to write us a message.
This program takes Excel files as an input where information about the facility, departments, and the dependencies between the departments are specified. Sample files are provided in the samples folder in Excel Files directory.
Likewise, this program also outputs Excel File with the determined x- and y-coordinates of the departments. These can be directly imported into the program visTABLE®. Two visualizations are also generated (of first and second optimization step). It can be used as an intermediate step to inspect the output of this program without having to import it into visTABLE®. Here is an example visualization:
This project was created in 2022 by Sarah Gräßle (Github), Julia Schmalzbauer (Github), Dario Schmid (Github), and Andreas Seger (Github) as part of a case study in discrete optimization at Technical University Munich, Germany. This project aims to solve the mathematical facility layout problem using a two-stage approach with nonlinear and linear optimization. For this purpose, we use scipy.optimize.minimize and our own implementation of gradient descent for the nonlinear part, and Gurobi for the integer program. This project was completed with the help of our supervisor at Technical University Munich, Dr. Michael Ritter, and Constantin Radzio at Voith Turbo GmbH, Garching bei München.
Choose a folder where you want to download the code to. Now download the code, which you can do in two different ways:
- Download FacilityLayout-main.zip and extract it to your chosen folder (recommended for non-programmers)
- Run
git clone git@github.com:darioschmid/FacilityLayout.gitin your chosen folder.
Now you should have a new folder called FacilityLayout or FacilityLayout-main in your previously chosen directory, containing files like README.md or requirements.txt.
Firstly, make sure you have Python 3 installed. If not, go to python.org and download the latest version. Generally any version above 3.0.0 should work.
Make sure to check the box "Add Python to PATH" and reboot after the installation is finished. Verify your Python installation by running python --version in your shell of choice (e.g. "Command Line" for Windows, "Terminal" for Mac and Linux).
Then install all packages listed in requirements.txt which can be achieved by running pip install -r requirements.txt, again in your shell of choice.
Put your three input Excel files into excel_files/import, namely:
Information.xlsxTransportflow.xlsxTransportmean.xlsx
Or modify the already existing ones. Sample files can also be found at excel_files/import/samples.
settings.ini provides a way to customize the behavior of this program. You can modify:
- Path for Excel input and output files
- Path for Visualizations
- Various settings for optimization process
Run main.py from your FacilityLayout/src or FacilityLayout-main/src folder, either by double-clicking on it, or running python src/main.py in your shell.
Please note that when double-clicking, the command line closes when the program finishes. If you want to see the output, open FacilityLayout.log.
In the following you will find troubleshooting steps if one of the above steps fails.
fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists:
Please make sure your command does not contain any typos, and the link is up-to-date.
Command 'git' not found or similar:
Please install git on your computer, or download the code here.
No such file or directory: 'requirements.txt':
Please make sure you are in the FacilityLayout or FacilityLayout-main folder in your shell. On Windows you can do that by navigating to that folder in the File Explorer and then type cmd in the top address bar and hitting enter. Then a command prompt window should appear.
python: can't open file 'C:\...\src\main.py': [Errno 2] No such file or directory:
Please make sure you are in the FacilityLayout or FacilityLayout-main folder in your shell. On Windows you can do that by navigating to that folder in the File Explorer, typing cmd in the top address bar, and hitting enter.
Command 'python' not found:
Can be the consequence of the following errors:
- When installing Python, you did not check the box "Add Python to PATH". Run the installer again, click on "modify", and tick "Add Python to environment variables", and restart your machine.
- You did not restart your machine after installing Python.
- Windows is blocking the
pythoncommand for the Microsoft Store version of Python. According to this thread go to Settings → Apps & featues → Manage App Execution Aliases → Disable Python - Despite checking the box in the installer, the Python executable was not added to the environment variables. Copy the path of the installation folder (Usually
C:\Program Files\Python310for Python 3.10), press Win+Pause → "Advanced System Properties" → Environment variables → click on "Path" in list → "Edit" → "New" → Paste Python installation path from previously. Reboot your machine. - The Python installation was not successful. Try uninstalling and reinstalling.