Dakota 7.0

The Dakota project delivers both state-of-the-art research and robust, usable software for optimization and uncertainty quantification (UQ). Dakota enables outer-loop analyses, such as optimization, UQ, sensitivity analysis and calibration. Its flexible interfacing structure lets you connect a wide range of simulation codes to Dakota, from black-box executables to Python functions. Build one interface to your code, called an analysis driver, and every Dakota method can use it.

Five Dakota method families (parameter studies, sensitivity analysis, uncertainty quantification, optimization and calibration) all connect through one analysis driver to your simulation.

Fig. 1 One analysis driver connects your simulation to every Dakota method.

Start here

Get help

Ask questions, report bugs and request features on Dakota’s GitHub Discussions page. Help Resources also lists video screencasts and training materials.

Cite Dakota

If you use Dakota in your work, please cite it as described in Citing Dakota.