9801303 Williamson This Small Business Research Innovation Research Phase II project focuses on the development of a state of the art parameter estimation software package. Many phenomena in science and engineering are modeled by systems of ordinary differential equations (ODEs). Often, although it is possible to write out the ODEs which govern the system, the coefficients or parameters which describe the rate of the various processes are not known. If the researcher has experimental data describing the operation of the system, it can be used to estimate the parameters. The process of numerically determining these parameters is known as parameter identification or estimation. Parameter identification problems are common in many areas, including chemical kinetics, thermodynamics, combustion, population modeling in biological systems, and pharmacology. Unfortunately, these problems can be extremely difficult for all but the simplest problems, and there are currently no sufficiently robust commercial or public domain codes. During Phase I, the firm demonstrated that the proposed domain decomposition-based algorithm was more robust than standard methods for parameter estimation. The objectives of Phase II are further design and implementation to improve the robustness and computational efficiency of the method to yield a viable commercial software package. Parameter estimation is common in many scientific and engineering disciplines, but there is no readily available, reliable software for non-specialist users. Thus, the market for a robust and relatively easy to use parameter estimation code is substantial. Since the potential user base is highly educated and able to pay relatively high prices for a package that conveniently solves their problems, the commercial potential of this project is substantial.
