公开(公告)号：JPH1125182A

公开(公告)日：1999-01-29

申请号：JP15117598

申请日：1998-06-01

Applicant: IBM

Inventor： JEROME M KERTZBERG ,  MENACHEM REBANONI

IPC: G05B15/02 ,  B65G61/00 ,  G05B19/418 ,  G06F19/00 ,  G06Q10/06 ,  G06F17/60

Abstract: PROBLEM TO BE SOLVED: To execute an industrial experiment plan by deciding the ranking of manufacture process variables and the interaction, utilizing the decided ranking and using a biased experiment plan. SOLUTION: A manufacture process is provided with two variables, the experiment plan without bias as complete contrast by the variables is indicated as a reference number 10 and up to three levels are used for the respective variables. That is, in the case that on-going knowledge relating to the process specifies that a first variable X1 is provided with weight 10, a second variable X2 is provided with the weight 4 and the interaction between X1 and X2 is provided with the weight 6, 5 is defined for X1, 2 is defined for X2, 3 is defined for the interaction between X1 and X2 and rounding to the total of 10 points is performed. Thus, the industrial experiment plan is made possible.