Joint optimization of mean and standard
Standard Deviation in the Business World
Joint optimization of mean and standard deviation using response surface methods The importance of statistical information cannot be stressed enough in the business world, but even with our advanced computing systems, information can be entered wrong, or the deviation to large or small. The purpose of this article is to investigate the potential, and problems, with the dual response system (DRS), response surface methodology (RSM), and robust parameter design (RPD). In this study, the author explores the use of each system and the inherent problems that arise when a business chooses to use optimize the mean while keeping the standard deviation below a specified value. This one shot approach is acceptable, but this approach resists easy analysis. To eliminate some of these problems, many business are using the DRS to obtain more flexible information access. One of the approaches is to use a nonlinear multiobjective programing technique that uses the NIMBUS software and Solver in an Excel spreadsheet to acquire simultaneous solutions to the mean and standard deviation functions. The author suggests that through the use of specific algorithms, business can eliminate the DNS problems, and achieve a standardization of reporting. The findings of this study are not so much to introduce an overall fix for the DNS problem, but to inform the reader about a number of mathematician who are working to introduce a “one size fits all” solution to the global optimal solution in reporting the mean and the standard deviation targets (Onur, Necip 2003).
Reference:
Koksoy Onur, Necip Doganaksoy. (2003). Journal of Quality Technology. Joint optimization of mean and standard deviation using response surface methods: http://search.proquest.com.ezproxy.apollolibrary.com/docview/214494399
Joint optimization of mean and standard deviation using response surface methods The importance of statistical information cannot be stressed enough in the business world, but even with our advanced computing systems, information can be entered wrong, or the deviation to large or small. The purpose of this article is to investigate the potential, and problems, with the dual response system (DRS), response surface methodology (RSM), and robust parameter design (RPD). In this study, the author explores the use of each system and the inherent problems that arise when a business chooses to use optimize the mean while keeping the standard deviation below a specified value. This one shot approach is acceptable, but this approach resists easy analysis. To eliminate some of these problems, many business are using the DRS to obtain more flexible information access. One of the approaches is to use a nonlinear multiobjective programing technique that uses the NIMBUS software and Solver in an Excel spreadsheet to acquire simultaneous solutions to the mean and standard deviation functions. The author suggests that through the use of specific algorithms, business can eliminate the DNS problems, and achieve a standardization of reporting. The findings of this study are not so much to introduce an overall fix for the DNS problem, but to inform the reader about a number of mathematician who are working to introduce a “one size fits all” solution to the global optimal solution in reporting the mean and the standard deviation targets (Onur, Necip 2003).
Reference:
Koksoy Onur, Necip Doganaksoy. (2003). Journal of Quality Technology. Joint optimization of mean and standard deviation using response surface methods: http://search.proquest.com.ezproxy.apollolibrary.com/docview/214494399