Premium
Experimental design for non‐statisticians
Author(s) -
Neuman Richard C.
Publication year - 1993
Publication title -
journal of vinyl technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.295
H-Index - 35
eISSN - 1548-0585
pISSN - 0193-7197
DOI - 10.1002/vnl.730150110
Subject(s) - computer science , simple (philosophy) , variable (mathematics) , industrial engineering , dividend , profitability index , product (mathematics) , design of experiments , operations research , econometrics , management science , mathematics , statistics , engineering , economics , mathematical analysis , geometry , epistemology , finance , philosophy
Many of us were taught that if more than one variable were changed at a time, the cause of some end result could never be determined. This is certainly true if the experiment is not planned to accommodate analysis of the data. The complex problems faced by modern scientists and engineers can not be solved efficiently using one‐at‐a‐time methods. Good experimental design promotes broader exploration of the variables studied, produces more results from fewer experiments, and validates results by comparing them against measured experimental error. This is done by deliberately varying all variables from run‐to‐run in systematic patterns. Timely product development and problems solving bears directly on profitability. A systematic approach can pay handsome dividends. The basis of good experimental design and several simple designs for everyday use are presented.