pom project
DOE PROJECT
Report
Submitted to:
Dr. B. Baliga (Course Instructor)
1Mechanial and Manufacturing Engineering student, St Cloud State University, St. Cloud, MN
St. Cloud State University
St. Cloud, MN
November 16, 2011
1 Introduction
Design of experiments, DOE, is an important tool in engineering because it helps us get to the desired solution of our problem by using the shortest amount of steps while still keeping a very high standard of safety and quality. The purpose of this project was to become familiar with DOE, and with the Taguchi method of predicting results, and also to be able to apply it for real life situations. The project consisted in selecting 4 factors at 3 levels that would affect our results. The degrees of freedom were then calculated and an appropriate orthogonal array (L9) was then chosen. After performing 4 trials for each combination given in the L9 orthogonal array, the results were interpreted using Minitab. The corresponding Means and Signal to Noise ratio were found and plotted in Minitab. ANOVA tables were also made in Minitab to see which factors affect the results more and which ones can be seen as just an error. From the signal to ratio plots I was able to determine which combination would give me optimal results. Using the contour plots of my 4 factors vs. SNR, and using the predict tool found in Minitab I was able to predict where my tennis ball would land. Finally test trials were done to see how accurate my results were and how well the predict tool actually works.
2 Experimentation and Results
A tennis ball and a rubber band were randomly chosen and a catapult was used to perform the experiment ( Fig. 1). The catapult was aligned with the edges of the table to make sure that it was shooting straight. Since even the slightest movement could affect our results dramatically, clamps were used to hold the catapult firm against the table. 4 factors at 3 levels were chosen and the results are shown in
Report
Submitted to:
Dr. B. Baliga (Course Instructor)
1Mechanial and Manufacturing Engineering student, St Cloud State University, St. Cloud, MN
St. Cloud State University
St. Cloud, MN
November 16, 2011
1 Introduction
Design of experiments, DOE, is an important tool in engineering because it helps us get to the desired solution of our problem by using the shortest amount of steps while still keeping a very high standard of safety and quality. The purpose of this project was to become familiar with DOE, and with the Taguchi method of predicting results, and also to be able to apply it for real life situations. The project consisted in selecting 4 factors at 3 levels that would affect our results. The degrees of freedom were then calculated and an appropriate orthogonal array (L9) was then chosen. After performing 4 trials for each combination given in the L9 orthogonal array, the results were interpreted using Minitab. The corresponding Means and Signal to Noise ratio were found and plotted in Minitab. ANOVA tables were also made in Minitab to see which factors affect the results more and which ones can be seen as just an error. From the signal to ratio plots I was able to determine which combination would give me optimal results. Using the contour plots of my 4 factors vs. SNR, and using the predict tool found in Minitab I was able to predict where my tennis ball would land. Finally test trials were done to see how accurate my results were and how well the predict tool actually works.
2 Experimentation and Results
A tennis ball and a rubber band were randomly chosen and a catapult was used to perform the experiment ( Fig. 1). The catapult was aligned with the edges of the table to make sure that it was shooting straight. Since even the slightest movement could affect our results dramatically, clamps were used to hold the catapult firm against the table. 4 factors at 3 levels were chosen and the results are shown in