This repository provides the code for applying ROBIST, a simple, yet effective, data-driven algorithm for optimization under parametric uncertainty. The methodology and numerical experiments are described in detail in an accompanying paper, available online here. The method was developed by Justin Starreveld, Guanyu Jin, Dick den Hertog and Roger Laeven.
The code is written in Python
, version 3.10.9. The dependency packages are listed in pyproject.toml
. The ROBIST algorithm is implemented in src/robist/robist.py
.
The recommended way to install robist
in your (virtual) environment is with
Python's pip
:
pip install robist
To be able to run the numerical experiments, make sure to clone the repository and install with the examples-requirements:
git clone https://github.com/JustinStarreveld/ROBIST.git
cd robist
pip install ".[examples]"
-
Toy Problem (abbreviated as tp). We compare ROBIST with the methods of Calafiore & Campi (2005) and Yanıkoglu & den Hertog (2013) in
examples/tp_experiments.py
. -
Altered Toy Problem (abbreviated as tp2). We analyze the performance of ROBIST in more detail in
examples/tp2_analysis.py
,examples/tp2_extra_experiments_1.py
andexamples/tp2_extra_experiments_2.py
. -
Linear CCP Problem (abbreviated as jiang2022). We compare ROBIST with the SAA-based methods of Ahmed et al. (2017) and Jiang and Xie (2022) in
examples/jiang2022_experiments.py
. -
Weighted Distribution Problem (abbreviated as wdp). We compare ROBIST with the scenario optimization methods of Calafiore & Campi (2005), Caré et al. (2014), Calafiore (2016) and Garatti et al. (2022) in
examples/wdp_experiments.py
. -
Portfolio Management Problem (abbreviated as pm). We compare ROBIST with the data-driven robust optimization approach proposed by Bertsimas et al. (2018) and the scenario optimization approach of Calafiore (2013) in
examples/pm_experiments.py
. -
Two-Stage Lot-Sizing Problem (abbreviated as ls). We compare ROBIST with the method of Vayanos et al. (2012) in
examples/ls_experiments.py
.
For more information on these experiments and the results we refer to preprint-paper.
We demonstrate ROBIST using the illustrative example as described in Section 2.2 of the paper.
Here we apply the algorithm to the following toy problem from Yanıkoglu & den Hertog (2013):
where
Suppose we have access to a data set of
First, we randomly split the data set into three equal-sized sets
We initialize the algorithm by optimizing for the expected/nominal scenario, i.e.,
We find that
We find that
Once the stopping criteria is reached, we use the validation data
The script used to create the figures in this illustrative example is examples/tp_illustrative_plots.py
.
Our code is not flawless. In case you have any questions or suggestions, please reach us at [email protected].
Was our software useful to you? You can cite us using:
@misc{ROBIST,
doi = {10.5281/zenodo.10143595},
year = {2023},
author = {Justin Starreveld, Guanyu Jin, Dick den Hertog and Roger JA Laeven},
title = {ROBIST: Robust Optimization by Iterative Scenario Sampling and Statistical Testing},
url = {https://github.com/JustinStarreveld/ROBIST}
}