One-factor-at-a-time (OFAAT) and Design of Experiment (DOE)

19 May No Comments

 

“One-factor-at-a-time (OFAAT) and  Design of Experiment (DOE)” Please respond to the following:

Note: Online students, please select one of the two subjects to discuss.

Per the textbook, trying to understand factors that impact the outcomes of business process is an important aspect of improving business operations. Conventional wisdom plans experiment one-factor-at-a-time (OFAAT). Compare and contrast the main advantages and disadvantages of OFAAT and DOE and select the approach (e.g., OFAAT or DOE) that you would use in order to obtain effective business process. Provide a rationale for your response. 

In one-factor-at-a-time (OFAAT) each process variable is studied individually by varying the variable and measuring the response while holding all other variables constant.  Three assumptions are made 1) that the relationship between the variable and response is a complicated function with multiple peaks and valleys, 2) that the process variables function independently of each other, and 3) that the experimental variations associated with the response is small.  One advantage of OFAAT is that it is the most popular experimental methodology.  Another advantage is that no advanced statistical knowledge is needed in its execution and data analysis.  The process is simply repeated over and over until all factors have been varied.  Disadvantages of OFAAT include that it does not thoroughly examine all possible factor combinations and it prevents you from studying interactions among other factors.

 

Design of experiment (DOE), or a designed experiment, can be a more effective way to determine the impact of two or more factors on a response than the OFAAT because only one factor is changed at a time while all others remain constant.  There are advantages of DOE including the requirement of fewer resources, the estimates of the effects are more precise, the interaction between factors can be estimated systematically, and there is experimental information in a larger region of the factor space. DOE covers all possibilities and looks for factor interactions. 

 

The approach that I would use in order to obtain an effective business process is DOE and I base my decision on the following example:

 

Hewlett Packer’s first color ink-jet printer produced great prints on paper, but overhead transparencies were light with washed out colors and customers were not satisfied with the overhead transparencies.  Many top performers tried to improve the printer’s performance printing overhead transparencies by improving the inks but using OFAAT did not find any solutions.  DOE was applied and 46 experimental trials were performed.  Results of experimental trials that were never run were predicted in which the best inks were quickly predicted and the sweet spot inks were proven in the lab.  The sweet spot inks produced significant quality improvement, they were darker with more vibrant colors on overhead transparencies and customers were satisfied.  In addition to the improvement to the overhead transparencies they also doubled the number of colors that could be printed on paper.  It was determined that optimizing the interactions among the inks was critical to finding a solution.

Select one (1) project from your working or educational environment that you would apply the DOE technique for the work process. Next, analyze the overall manner in which you would utilize DOE to manage and improve the work process of the project in question. Provide a rationale for your response. 

In one-factor-at-a-time (OFAAT) each process variable is studied individually by varying the variable and measuring the response while holding all other variables constant.  Three assumptions are made 1) that the relationship between the variable and response is a complicated function with multiple peaks and valleys, 2) that the process variables function independently of each other, and 3) that the experimental variations associated with the response is small.  One advantage of OFAAT is that it is the most popular experimental methodology.  Another advantage is that no advanced statistical knowledge is needed in its execution and data analysis.  The process is simply repeated over and over until all factors have been varied.  Disadvantages of OFAAT include that it does not thoroughly examine all possible factor combinations and it prevents you from studying interactions among other factors.

 

Design of experiment (DOE), or a designed experiment, can be a more effective way to determine the impact of two or more factors on a response than the OFAAT because only one factor is changed at a time while all others remain constant.  There are advantages of DOE including the requirement of fewer resources, the estimates of the effects are more precise, the interaction between factors can be estimated systematically, and there is experimental information in a larger region of the factor space. DOE covers all possibilities and looks for factor interactions. 

 

The approach that I would use in order to obtain an effective business process is DOE and I base my decision on the following example:

 

Hewlett Packer’s first color ink-jet printer produced great prints on paper, but overhead transparencies were light with washed out colors and customers were not satisfied with the overhead transparencies.  Many top performers tried to improve the printer’s performance printing overhead transparencies by improving the inks but using OFAAT did not find any solutions.  DOE was applied and 46 experimental trials were performed.  Results of experimental trials that were never run were predicted in which the best inks were quickly predicted and the sweet spot inks were proven in the lab.  The sweet spot inks produced significant quality improvement, they were darker with more vibrant colors on overhead transparencies and customers were satisfied.  In addition to the improvement to the overhead transparencies they also doubled the number of colors that could be printed on paper.  It was determined that optimizing the interactions among the inks was critical to finding a solution.

 

http://mescal.imag.fr/membres/arnaud.legrand/teaching/2011/EP_czitrom.pdf

https://www.piersystem.com/clients/wsatag/72369.pdf




Click following link to download this document

One-factor-at-a-time (OFAAT) and Design of Experiment (DOE).docx