Overview
NOTE: As of September 2008, Surfpack is no longer under active development. Surfpack will continue to be maintained as part of the Dakota software toolkit.
Surfpack is a general-purpose software library of multidimensional function approximation methods for applications such as data visualization, data mining, sensitivity analysis, uncertainty quantification, and numerical optimization. Surfpack is primarily intended for use on sparse, irregularly-spaced sets of data points, where the data do not lie on a regularly-spaced grid. Surfpack generates output data files that can be readily imported into many standard data plotting and visualization software tools.
Surfpack generates mathematical models based on user-supplied data, and then uses these models for predicting (both via interpolation and extrapolation) general trends in the data. For example, a geologist could use Surfpack to create an approximate “map” of an underground water source, based on moisture content data from a few irregularly-spaced drill sites located within a city. Surfpack is a general-purpose mathematical and statistical software package, with broad applications to both technical and non-technical fields of study.
The Surfpack software package is written in a combination of C++ and FORTRAN77 programming languages. The C++ portion of Surfpack provides the primary data management and user interface capabilities, while the FORTRAN77 portion of Surfpack covers the bulk of the function approximation calculations. There are five main function approximation types in Surfpack. These are (1) low-order polynomial regression, (2) kriging interpolation, (3) multivariate adaptive regression splines, (4) a simple artificial neural network, and (5) a basic radial basis function method.
See the Surfpack manuals and reference materials for more information on the capabilities of this software package.
Surfpack Manuals and References
Giunta, A. A., Swiler, L. P., Brown, S. L., Eldred, M. S., Richards, M. D., and Cyr, E. C., “The Surfpack Software Library for Surrogate Modeling of Sparse Irregularly Spaced Multidimensional Data,” in Proceedings of the 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, AIAA Paper 2006-7049, Portsmouth, VA, 2006.
Related References
- Giunta, A. A., McFarland, J. M., Swiler, L. P., and Eldred, M. S., “The promise and peril of uncertainty quantification using response surface approximations,” Structure and Infrastructure Engineering: Maintenance, Management, Life-Cycle Design & Performance, special issue on Uncertainty Quantification and Design under Uncertainty of Aerospace Systems, Vol. 2, Nos. 3-4, Sept.-Dec., 2006, pp. 175-189.
- Giunta, A. A., and Watson, L. T., “A Comparison of Approximation Modeling Techniques: Polynomial Versus Interpolating Models,” Proceedings of the 7th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, St. Louis, MO, Sept. 1998, pp. 392-404 (AIAA Paper 98-4758).