Method for improving a manufacturing process

Abstract

A factorial experiment is conducted on a manufacturing process to generate a response matrix. The responses are used to calculate individual contrasts in a document as well as replicates effects. The contrast sums are also calculated and displayed in the document. The largest of the contrast sums are identified, and effects associated with those contrast sums are tested for significance using an end count method. The information from the process transformed into “significant effects” information is used to adjust process variables to improve the manufacturing process by avoiding the effect or imparting it to a measurable response of the process.

1 - 16 . (canceled)  
   
   
       17 . A method of adjusting a manufacturing parameter comprising the steps of: 
 a) determining a plurality of input variables that may affect the parameter;    b) designing an experiment having test runs;    c) setting the input variables as required by the experiment and recording the results of the experiment;    d) calculating a plurality of contrasts;    e) calculating a plurality of estimated effects;    f) identifying a test effect, from the plurality of estimated effects, to test for significance;    g) determining whether the test effect is an interaction of at least two variables of the plurality of input variables;    h) if the test effect is an interaction of the at least two variables, removing an estimated effect of each of the at least two variables;    i) testing whether the test effect satisfies a criterion of significance; and    j) displaying whether the test effect satisfies the criterion of significance.    
   
   
       18 . A method according to  claim 17  further comprising the steps of: 
 k) restoring the effects removed in step h;    l) if the test effect satisfied a criterion of significance, removing the test effect, where the test effect is removed in a manner that minimizes the estimated effect of the at least two variables if the test effect was an interaction;    m) recalculating the contrasts and the estimated effects;    n) determining if a remaining effect of the plurality of estimated effects should be tested for significance;    o) if a remaining effect should be tested for significance, iterating steps g through n;    p) adjusting at least one of the plurality of variables.    
   
   
       19 . A method according to  claim 18  wherein: 
 for each test run, the result is stored in a cell that is identifiable by reference to the variables of the plurality of input variables associated with the test run; and    if the test effect is an interaction, step 1 comprises the following steps determining whether the interaction is a spike interaction; 
 if the interaction is a spike interaction, subtracting the test effect from the results stored in the cells that are associated with the interaction and that manifest the spike interaction;  
 if the interaction is not a spike interaction, subtracting one half of the test effect from the results stored in the cells that are associated with the interaction;  
   
   
   
       20 . A method according to  claim 19  wherein: 
 the cells are located in a memory of a computer.    
   
   
       21 . A method according to  claim 18  wherein: 
 the experiment comprises a full factorial experiment.    
   
   
       22 . A method according to  claim 21  wherein: 
 step i comprises conducting an end count.    
   
   
       23 . A method for improving a manufacturing process comprising the steps of: 
 identifying a subject parameter;    identifying a plurality of variables that may affect the subject parameter including at least a first variable and a second variable;    determining a first plurality of values for the first variable and determining a second plurality of values for the second variable;    conducting a plurality of test runs wherein a test run includes the steps of: 
 assigning one of the first plurality of values to the first variable;  
 assigning one of the second plurality of values to the second variable,  
 conducting the manufacturing process,  
 measuring the magnitude of the subject parameter, and  
 recording the magnitude of the subject parameter in a one of a plurality of cells that are identifiable based on the plurality of variables;  
   calculating an estimated effect on the subject parameter of an interaction involving the first variable and the second variable;    determining whether the estimated effect is significant;    if the estimated effect is significant determining whether the interaction is a spike interaction;    if the estimated effect is a spike interaction, subtracting the estimated effect from those cells of the plurality of cells that are associated with the interaction in such a manner that an estimated effect of the first variable on the subject parameter and an estimated effect of the second variable on the subject parameter are minimized;    determining whether an effect of one of the first variable and the second variable is significant.    
   
   
       24 . A method for improving a manufacturing process wherein a value of a measurable response of the manufacturing process is indicative of an improvement to the process and wherein a plurality of process variables that may affect the value of the measurable response have been identified, the method comprising the steps of: 
 identifying a negative level and a positive level for each process variable of the plurality;    identifying a plurality of interactions wherein the plurality of interactions includes an interaction that corresponds to each possible combination of variables in the plurality of process variables;    conducting a full factorial experiment including a plurality of test runs wherein, for each test run, each of the plurality of process variables is set at either its negative level or its positive level and wherein each possible combination of variable settings is set for at least one of the test runs;    recording the responses of the full factorial experiment and calculating individual contrasts for each of the plurality of process variables and each of the plurality of interactions;    testing the significance of effects associated with the contrasts, wherein before each effect is tested, the previously tested effect is removed from the responses if found significant, and when the effect to be removed is an interaction effect it is removed to achieve the smallest remaining estimates for the lower order effects; and    adjusting the level settings of the process variables as a function of a least one of the significant effects.    
   
   
       25 . The method of  claim 23 , wherein when the effect to be removed is a spike interaction effect, it is removed only from a response in which it was observed.