Metadata-Version: 2.1
Name: worstcase
Version: 0.3.1
Summary: Worst case analysis and sensitivity studies using Extreme Value and/or Monte Carlo methods.
Home-page: https://github.com/amosborne/worstcase
License: MIT
Author: amosborne
Author-email: amosborne@users.noreply.github.com
Requires-Python: >=3.9,<3.10
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Requires-Dist: Pint (>=0.17,<0.18)
Requires-Dist: networkx (>=2.5.1,<3.0.0)
Requires-Dist: pyDOE (>=0.3.8,<0.4.0)
Requires-Dist: treelib (>=1.6.1,<2.0.0)
Project-URL: Repository, https://github.com/amosborne/worstcase
Description-Content-Type: text/markdown

# worstcase

## Overview

`pip install worstcase`

Worst case analysis and sensitivity studies using Extreme Value and/or Monte Carlo methods.

This package coexists alongside far more capable uncertainty analysis and error propagation packages such as [uncertainties](https://pypi.org/project/uncertainties/) (first-order propagation), [soerp](https://pypi.org/project/soerp/) (second-order propagation), and [mcerp](https://pypi.org/project/mcerp/) (Monte Carlo propagation).

This package is designed for engineering applications where worst case analysis computations are often done using the Extreme Value method over single-valued functions while falling back to the Monte Carlo method when the worst case is known to not exist at the extremes. The Extreme Value method is implemented as a brute-force search; the Monte Carlo method is implemented with Latin Hypercube Sampling over a uniform distribution.

## Usage

See the example usage [here](https://github.com/amosborne/worstcase/blob/master/examples/readme.ipynb).

