A technique is proposed for reducing the number of repetitious solu tions required to carry out a conventional parametric study of a theoretical problem. The technique is potentially applicable to a particular class of problems which involve the numerical solution of differential equations. An applicability criterion is presented, along with a method for predicting the errors which may be expected in problems which do not satisfy the criterion exactly.
Get full access to this article
View all access options for this article.
References
1.
2.
Fisher R.A.Statistical methods for research workersOliver and Boyd10th edition 1946
