| 摘要 |
The method first creates a BS database by randomly sampling experimental data items from an experimental data DB storing a plurality of experimental data items having n (n is an integer of 2 or larger) kinds of variates. It then identificates an n-dimensional sampling curve or surface model approximating a correlation of the n kinds of variates with reference to the BS database, and stores the model into a BS optimum solution database. The method then repeats the creation of the BS database and the identification of the sampling curve or surface model, thereby creating the optimum solution database. Finally, the method obtains a mean and variance of the optimum solution with reference to the BS optimum solution database and estimates the reliability of the optimum solution from these values.
|
