You are on page 1of 10

Int J Adv Manuf Technol (2009) 41:760769 DOI 10.

1007/s00170-008-1526-1

ORIGINAL ARTICLE

Taguchi-based Six Sigma approach to optimize plasma cutting process: an industrial case study
Joseph C. Chen & Ye Li & Ronald A. Cox

Received: 19 September 2007 / Accepted: 10 April 2008 / Published online: 18 June 2008 # Springer-Verlag London Limited 2008

Abstract This case study outlines the use of Taguchi parameter design to optimize the roundness of holes made by an aging plasma-cutting machine. An L9 array is used in a Taguchi experiment design consisting of four controllable factors, each with three levels. With two non-controllable factors included in the setting, we conduct 36 experiments, compared to the 81 parameter combinations (four factors, three levels or 34) required in a traditional DOE setting. Therefore, using the Taguchi method significantly reduces the time and costs of a quality improvement process. Conducted for two response variablesbevel magnitude and the smallest diameter deviation of the holethe Taguchi experiments gave the optimal combination A1B2C1D3 (small for tip size, 93 in/min for feed rate, 100 V for voltage, and 63A for amperage), which is verified with a confirmation run of 30 work pieces. All 30 cuts meet the quality requirement for subsequent assembly. Furthermore, statistical analysis indicates that the mean value and standard deviation of the confirmation run data
J. C. Chen Department of Agricultural and Biosystems Engineering, Iowa State University, 221 I. ED. II, Ames, IA 50011-3130, USA Y. Li (*) Department of Industrial and Manufacturing Systems Engineering, Iowa State University, 2019 Black Engineering, Ames, IA 50011-2164, USA e-mail: yeli@iastate.edu R. A. Cox Center for Industrial Research and Service (CIRAS), Iowa State University, Campus, 2272 Howe Hall, Suite 2620, Ames, IA 50011-2272, USA

are smaller than those before Taguchi parameter design is conducted. Keywords Taguchi method . Quality . Six Sigma . Process optimization

1 Introduction Many small- and medium-sized industries have implemented the two most popular process improvement methodologieslean manufacturing and Six Sigmawhich originated at Toyota and Motorola, respectively. Each methodology has its unique structure and tools. The central focus of lean manufacturing is to provide value by eliminating waste, which is defined as anything that is not value-added from the customer s perspective. The seven deadly wastes, as defined by this method, are overproduction, inventory, waiting, movement, transportation, defects, and over-processing. By continuously eliminating these wastes, the customer receives a high value product. When the variation of a part or a service does not meet the specifications of the downstream internal and/or external customers, the methodology and tools of Six Sigma can be implemented to improve the quality of the product or service. The latest structure of Six Sigma is defined as the define-measure-analysis-implementation-control (DMAIC) model. Each stage of the model provides tools for conducting Six Sigma quality improvements for any process or service. Figure 1 shows a summary of tools used in each stage of Six Sigma management. In order to reduce the variation of a process, many Six Sigma teams use design of experiments (DOE) methodology to find solutions that will align products with customer expectations. DOE is a statistical

Int J Adv Manuf Technol (2009) 41:760769


Phase 0: Define Scope and Boundary Define Defects Team Charter and Champion Estimated $ Impact Leadership approval

761

Phase I: Process Measurement Map Process and Identify Inputs and Outputs Cause and Effects Matrix Establish Measurement System Capability Establish Process Capability Baseline

Phase II: Process Analysis Complete FMEA Perform Multi-Vari Analysis Identify Potential Critical Inputs Develop Plan for Next Phase

Phase III: Process Improvement Verify Critical Inputs Optimize Critical Inputs via Taguchi

Fig. 3 Plasma table



Phase IV: Process Control Implement Control Plan Verify Long Term Capability Continuously Improve Process

Fig. 1 DMAIC process improvement methodology

tool that studies the relationship between independent variables, the Xs (process variables), and their interactions on a dependent variable, the Y, which is considered the critical-to-quality (CTQ) of the product. The most commonly used DOE tool is the 2k factorial design, where k is the number of factors, each with two levels. Thus, in a three-factor design, there are eight treatment combinations, i.e., 23 or 222. Unfortunately, for most manufacturing processes, two levels of each factor may provide insufficient information about quality improvement. Often, three levels for each factor are needed. For example, when studying the feed rate of a turning operation, three levels could be evaluated (e.g., 0.005, 0.010, and 0.015 in. per revolution). Thus, for a threeFig. 2 Electric switchboard

factor, three-level design, there would be 27 treatment combinations, i.e., 33 or 333. In addition, each treatment combination should be run twice to achieve a more reliable statistical data analysis, bringing the total number of experiments to 54 (272). If the cost of each experiment is $100, the Six Sigma team will spend $5,400 to analyze a three-factor, three-level experiment design with two replicates. In many manufacturing processes, achieving a solution that meets customers specifications may require an evaluation of five or six factors. Costs will increase accordingly. For example, assuming a base cost of $100 per experiment, a six-factor, three-level experiment design with two replicates will cost $145,800 (3 6 or 729 combinations2=1,458). This is the cost of conducting DOE alone and does not include time spent on DOE, which may reduce the productivity of other jobs and delay resolution of the quality problem. In addition, during this time, the process will produce more defects and wastes. Based on the aforementioned analysis, resolving industrial problems cost-effectively and in a timely manner requires a

Fig. 4 Indicator light

762 Fig. 5 Illustration of bevel

Int J Adv Manuf Technol (2009) 41:760769

Bevel magnitude

more economical DOE approach. Taguchi parameter design, which is capable of providing the optimal solution with a reduced number of experiment runs, is one such approach. One small-sized, Midwest manufacturer has an aging plasma-cutting machine that produces defects and causes production delays. Though it is easier to replace older equipment with new machines, most small manufacturers have limited capital. The manager of this company approached the researchers for assistance in answering the following questions: 1. What is the optimal setting to produce the highest quality products? 2. Can the optimal setting lower the defect rate but maintain the required productivity? If the defect rate remains high after implementing optimal settings, the manager will have the justification he needs to invest in new equipment. To address this challenge, the researchers conducted a Taguchi parameter design study, using the results of that study to make recommendations. The procedure and process is demonstrated in the four phases outlined in Fig. 1.

conventional approaches used in off-line quality control, Taguchis philosophy is based on the belief that once quality is designed into both the product and process, only minimal inspection is necessary. Taguchi proposes a holistic view of quality related to cost, which extends the focus of quality beyond manufacturers at the time of production by integrating the customer and society as a whole. Taguchi defines quality as the (minimum) loss imparted by the product to society from the time the product is shipped. This economic loss is associated with losses due to rework, waste of resources during manufacturing, warranty costs, customer complaints and dissatisfaction, time and money spent by customers on failing products, and the eventual loss of market share. Taguchi methods provide an efficient and systematic way to optimize designs for performance, quality, and cost. These methods have been used successfully in Japan and the United States in designing reliable, high quality products at low cost in such areas as automotives and consumer electronics. Taguchi breaks down off-line quality control into three stages, concept design, parameter design, and tolerance design, which are summarized below: Concept design results in either a design concept or an up and limping prototype. In the initial phase, more than one design concept, each with its own set of pros and cons, can be presented. The ideal design concept will be the one that research shows best addresses customer needs and is inherently robust.

2 Overview of Taguchi parameter design Taguchi parameter design is one of several methods developed by Dr. Genichi Taguchi [1]. One of the
Fig. 6 Illustration of smallest diameter deviation

D normal

D smallest

Smallest diameter deviation: | Dsmallet - Dnormal | Dsmallest : Smallest diameter Dnormal : Nominal diameter Continuous curve: Actual plasma-cut hole Dashed curve: Nominal diameter hole, maximal roundness

Int J Adv Manuf Technol (2009) 41:760769 Fig. 7 Fish bone diagram

763

Parameter design is a critical production step. The nominal design features or selected process factor levels are tested and the combination of product parameter levels or process operating levels least sensitive to changes in environmental conditions and other uncontrolled factors (noise) is determined. Tolerance design is used to further reduce variation, if required, by tightening the tolerance of those factors shown to have a significant impact on variation. This stage utilizes loss function to determine whether spending more money on materials and equipment will result in a better product, thus emphasizing the Japanese philosophy of invest last not invest first.

Taguchi parameter design is an experiment-based process that uses the following steps to identify settings of design parameters that maximize performance characteristics (e.g., yield or productivity, etc.): 1. Identify initial and competing settings of design parameters, as well as important noise factors and their ranges. 2. Construct the design and noise matrices, and plan the parameter design experiments. 3. Conduct the parameter design experiments and evaluate the performance statistic for each test run of the design matrix. 4. Use the values of the performance statistic to predict new settings of the design matrix. 5. Confirm that the new settings truly improve the performance statistic. Considering multiple factors simultaneously, the Taguchi parameter design method uses orthogonal experimental combinations to shorten the product development cycle, which, in turn, saves time and money, Taguchi parameter design has been utilized in traditional manufacturing processes such as milling, turning, and drilling to determine

optimal combinations of parameters for better performance. Ghani et al. [2] applied Taguchi parameter design to optimizing parameters for end milling process. Low resultant cutting force and good surface finish were found with high cutting speed, low feed rate and low depth of cut. Zhang et al. [3] used a Taguchi parameter design application to optimize surface quality in a CNC face milling operation, where the best surface roughness (response) and signal-to-noise ratio were sought. Davim and Reis [4] studied cutting parameters of composite milling process using Taguchi-based experiments. Kirby et al. [5] discussed the application of Taguchi parameter design to optimize turning operations for best surface roughness. Palanikumar and Karthikeyan [6] used Taguchi methods to conduct experiments in turning composite material to achieve maximum material removal rate and minimum surface roughness. Taguchi parameter design has also been used to analyze optimal parameters for drilling process [7, 8, 9, 10]. Although many of the aforementioned projects were conducted primarily in laboratories, research can also be done in an industrial setting. The Iowa State University Center for Industrial Research and Service (CIRAS), an extension service unit of this Midwest land grant university,

A B C D

Factor Tip Size Feed Rate Voltage Amperage Noise Air Pressure Pierce Time

Units in/min volts (V) amperes (A)

Level 1 Small 83 100 43 Low 45 0.70

Level 2 Medium 93 105 53 High 60 1.40

Level 3 High 103 110 63

1 2

lbs/in2 (PSI) seconds (s)

Fig. 8 Test parameters

764 Fig. 9 Taguchi experiment table


Control Factors and Levels B C 1 1 2 2 3 3 1 2 2 3 3 1 1 3 2 1 3 2

Int J Adv Manuf Technol (2009) 41:760769


Noise Level Setting 1 2 3 4 45 45 60 60 0.7 1.4 0.7 1.4 N1 y11 y21 y31 y41 y51 y61 y71 y81 y91 N2 y12 y22 y32 y42 y52 y62 y72 y82 y92 N3 y13 y23 y33 y43 y53 y63 y73 y83 y93 N4 Avg Circularity y14 y24 y34 y44 y54 y64 y74 y84 y94

Noise
1-Air Pressure 2-Pierce Time

Run 1 2 3 4 5 6 7 8 9

A 1 1 1 2 2 2 3 3 3

D 1 2 3 3 1 2 2 3 1

has proposed this methodology to help industry to obtain optimum manufacturing processes to meet customer demands within a reasonable timeframe and budget. The following sections will present the application of Taguchi parameter design in helping a small business achieve the Six Sigma paradigm.

3 Current problem with an older plasma-cutting machine The case study in this paper is based on a problem presented by an electrical manufacturing company (Company A) located in Des Moines, Iowa. The major products of Company A are large electrical switchboards (Fig. 2). One step in the process of making the boards requires the use of a plasma table (Fig. 3) to cut holes for hardware such as an indicator light (Fig. 4). The current problem of the plasma cutting process is that some of the holes do not allow the fitting for the hardware. A close examination of the holes reveals that two reasons may keep the hardware from passing through the holes. First, the plasma cutting process is beveling the edge of the holes it creates (Fig. 5), which, in turn, obstructs the hardware. Meiji EMZ-5TR zoom stereo microscope was used to measure the magnitude (unit: 0.001 inch) of the bevels for analysis. Second,

the holes have poor circularity (roundness). Roundness is measured by finding the smallest diameter deviation, which is the difference of the smallest diameter of the actual hole from the nominal diameter (Fig. 6). Beveling and roundness problems that prevent hardware from fitting properly into the switchboard is a defect that requires rework. To minimize production costs due to this rework, a Taguchi analysis was undertaken to determine the optimal setting to produce holes with minimum bevel and minimum out-ofround diameter. Ideally, all of the holes cut by the plasma machine would have best round shape and enable the hardware to be inserted smoothly. To accomplish this project for Company A, a Six Sigma team consisting of operators, engineers, researchers and a manager was formed, and the Taguchi parameter design procedure was applied.

4 Six Sigma improvement process 4.1 Define 4.1.1 Factors and levels Before the Six Sigma team could conduct the Taguchi parameter design experiment, they needed to thoroughly

Fig. 10 Data of bevel (unit: 0.001)

Noise 1-Air Pressure 2-Pierce Time Control Factors and Levels Run A B C D 1 1 1 1 1 1 2 2 2 2 1 3 3 3 3 2 1 2 3 4 2 2 3 1 5 2 3 1 2 6 3 1 3 2 7 3 2 1 3 8 3 3 2 1 9

Noise Level Setting 1 2 3 4 45 45 60 60 0.7 1.4 0.7 1.4 N1 39 35 55 40 40 45 52 39 56 N2 31 25 19 47 52 54 48 47 51 N3 22 35 32 44 62 60 49 46 40 N4 42 35 37 82 87 62 49 52 45 Avg Bevel 33.5 32.5 35.75 53.25 60.25 55.25 49.5 46 48

S/N Ratio

8.96 5.00 14.91 19.38 19.97 7.63 1.73 5.35 6.98

2
80.33 25.00 222.25 375.58 398.92 58.25 3.00 28.67 48.67

-30.73 -30.31 -31.60 -34.94 -35.94 -34.91 -33.90 -33.30 -33.69

Int J Adv Manuf Technol (2009) 41:760769


Avg Bevel level 1 level 2 level 3 A B C D Optima combinations(Raw Data) 33.92 45.42 47.25 47.25 smaller the better 56.25 46.25 44.58 45.75 A1, B1, C2, D3 47.83 46.33 48.50 45.00

765

S/N Ratio level 1 level 2 level 3 A B C D Optima combinations(S/N Ratio) -30.9 -33.2 -33 -33.5 larger the better -35.3 -33.2 -33 -33 A1, B2, C1, D2 -33.6 -33.4 -33.8 -33.3

Fig. 11 Response of controllable factors to bevel and S/N ratio

done by determining normal settings for each factor and then setting one level lower (level 1) and one level higher (level 3), using the existing settings as level 2. For uncontrolled factors (noise factors), two levels were chosen, low and high. The levels and units of each factor are shown in Fig. 8. These levels were set to determine if changes to the factors impacted hole quality, and, if so, how much. In this way, the most significant factor was identified, and the optimal level for each factor determined. For the uncontrolled factors, the two levels helped determine if background noises produce a significant effect. This effect will be further analyzed with T-tests. 4.1.2 Experimental design Using these four controllable factors and two uncontrolled noise factors, we constructed an L9 Taguchi experiment table with the appropriate settings of each factor (Fig. 9). For each controllable parameter setting combination, four runs were conducted, each under a different noise factor setting combination. Each trial run is represented as yij, where i ranges from 1 to 9, denoting controllable parameter setting for experimental run, and j from 1 to 4, denoting noise factor setting. Thus, a total of 36 experimental runs are conducted as a setting shown in Fig. 9. The average value, standard deviation, variance and signal-to-noise (S/ N) ratio of the four runs under the same controllable parameter setting were calculated. The same experimental

understand the plasma-cutting process. A brainstorming exercise that examined all of the factors of the process provided the necessary information. Since these factors involve knowledge in different domains and could be either controllable or non-controllable, it was necessary to have all team members participate in the brainstorming. A fishbone diagram based on this exercise revealed all possible causes of defective plasma-cut holes from the perspectives of method, material, machine, operator and environment (Fig. 7). Through group discussion and ranking, four factors were identified as controllable factors (voltage, feed rate, amperage, and tip size) and two as uncontrollable noise factors (air pressure and pierce time). The Six Sigma team then determined ranges of the levels to explore for each factor. For controllable factors, this was

Factor A: Tip Size


-28 -29
S/N Ratio

S/N Ratio Ave Bevel

Factor B: Feed Rate (ipm)


-31 83 -32
Avg Bevel

S/N Ratio Avg Bevel

sm

md

lg

60

50 93 103 49 48 47 46 45 44
Avg Bevel Avg Bevel

-30 -31 -32 -33 -34 -35 -36

52

S/N Ratio

-33 -34 -35

44

36 -36 28 -37

Factor C: Voltage (Volts)


-32 -33
S/N Ratio

S/N Ratio Avg Bevel

Factor D: Amperage (Amps)


-32 -33 43 53 63

S/N Ratio Avg Bevel

100

105

110

54 52
Avg Bevel

52 50 48 46 44 42

50 -34 48 -35 -36 -37 46 44 42

S/N Ratio

-34 -35 -36 -37

Fig. 12 Response graphs for bevel magnitude

766 Fig. 13 Data of smallest diameter deviation (unit: inch)


Noise Level Setting 2 3 45 60 1.4 0.7

Int J Adv Manuf Technol (2009) 41:760769

Noise 1-Air Pressure 2-Pierce Time Control Factors and Levels Run 1 2 3 4 5 6 7 8 9 A 1 1 1 2 2 2 3 3 3 B 1 2 3 1 2 3 1 2 3 C 1 2 3 2 3 1 3 1 2 D 1 2 3 3 1 2 2 3 1

1 45 0.7

4 60 1.4 Avg Bevel 0.012 0.006 0.007 0.026 0.049 0.038 0.035 0.014 0.034

S/N Ratio

N1 0.002 0.001 0.006 0.012 0.037 0.023 0.037 0.009 0.041

N2 0.007 0.001 0.001 0.018 0.044 0.036 0.027 0.005 0.032

N3 0.018 0.006 0.003 0.021 0.060 0.040 0.042 0.027 0.021

N4 0.020 0.016 0.019 0.052 0.055 0.052 0.035 0.014 0.041

0.0087 0.0071 0.0081 0.0179 0.0104 0.0120 0.0062 0.0096 0.0095

2
0.000075 0.000050 0.000066 0.003200 0.000109 0.000143 0.000039 0.000092 0.000090

37.116 41.337 39.925 30.442 26.051 28.147 28.956 35.888 29.184

design table was used for two response variables, the bevel magnitude and the deviation of the smallest diameter of plasma-cut holes. 4.2 Measure and analysis This section provides details of the Taguchi-based experiments conducted for bevel magnitude and smallest diameter deviation. Data were collected and analyzed using the experimental design described in the previous section. 4.2.1 Taguchi experiment for bevel Figure 10 shows the experiment data for the bevel size. The response of each controllable factor on bevel magnitude and the signal-to-noise ratio are shown in Fig. 11. The quality characteristic is the-smaller-the-better; the formula to calculate S/N ratio is given below and its derivation can be found in [1]: " !#

determine the optimal combination of levels of the controllable factors that creates the minimum bevel. Figure 12 is the graphical depiction of the response and S/N ratio for bevel magnitude. Since the quality characteristic for beveling is the-smaller-the-better, the optimal setting combination for minimum bevel is A1-B1-C2-D3, which is interpreted as small tip size, a feed rate of 83 in/ min, a voltage of 105 V, and amperage of 63A. For the S/N ratio, we are looking for the largest value. Therefore, the optimal setting combination for S/N ratio response is A1B2-C1-D2, which means small tip size, a feed rate of 93 in/ min, a voltage of 100 V, and amperage of 53A. The arrows in Fig. 12 indicate the chosen level of each factor, and the directions of the arrows show the quality characteristic with upward denoting the-larger-the-better and downward thesmaller-the-better. 4.2.2 Taguchi experiment for smallest diameter deviation Similarly, Fig. 13 shows the experiment data for the smallest diameter deviation. The response of each controllable factor on the smallest diameter deviation and the S/N ratio are shown in Fig. 14. Figure 15 is the graphic display of the response to smallest diameter deviation. The quality characteristic is also the-smaller-the-better for the smallest diameter deviation. The optimal setting combination for minimum smallest diameter deviation is A1-B2-C1-D3, which means small tip size, a feed rate of 93 in/min, a

4 1 X h 10 log y2 n j1 ij

where, yij n S/N ratio, individual response for each trial run, the number of runs due to noise factors; in this case, n=4.

The bevel response table (Fig. 11) shows the mean response of the variable (bevel) from each controllable factor at each level in the Taguchi experimental design. Each level of the controllable factors has three response values in the orthogonal array (Fig. 10) and is calculated by averaging these three values. The response table for the S/N ratio (also in Fig. 11) was similarly obtained, using the S/N ratios in the orthogonal array. With the response values and the S/N ratio values, the response table can be used to

Fig. 14 Response of controllable factors to smallest diameter deviation

Int J Adv Manuf Technol (2009) 41:760769


Factor A: Tip Size
50 48 46 44 42 40 38 36 34 32 30 28 26 24 sm md lg
S/N Ratio Deviation

767
Factor B: Feed Rate (ipm)
40 38 0.000
Deviation S/N Ratio
S/N Ratio Deviation

0.050

36 34 32 30 28

-0.050

-0.100

-0.150 26 -0.200 24

0.080 0.075 0.070 0.065 0.060 0.055 0.050 0.045 0.040 0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 83 93 103

S/N Ratio

Factor C: Voltage (Volts)


50 48 46 44 42 40 38 36 34 32 30 28 26 24 100 105 110

S/N Ratio Deviation

Factor D: Amperage (Amps)


60 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 28 26 24 43 53 63

S/N Ratio Deviation

0.040 0.030
Deviation S/N Ratio

0.040 0.030
Deviation

S/N Ratio

0.020 0.010 0.000 -0.010 -0.020

0.020 0.010 0.000 -0.010 -0.020

Fig. 15 Response graph for smallest diameter deviation

voltage of 100 V, and an amperage of 63A. The S/N ratio responses render the same parameter setting combination. 4.2.3 Setting selection According to Figs. 12 and 14, bevel and smallest diameter deviation as well as their S/N ratios have different optimal setting combinations. These are summarized in the first four rows of Table 1. For example, bevel S/N ratio response gave level 2 (53A) for factor D (amperage), while all other responses (bevel, smallest diameter deviation and its S/N ratio) chose level 3(63A). However, Company A needs only one overall optimal setting combination for its plasma table. Therefore, the appropriate optimal setting combination is the level of each factor having the largest number of occurrences. For example, level 3 of factor D occurred

three times, while level 2 occurred once. Therefore, level 3 for factor D is chosen as the overall optimal setting combination. With this rule, the overall optimal setting for the plasma table is A1-B2-C1-D3, as indicated in the bottom row of Table 1. The setting A1-B2-C1-D3 means the small tip size, a feed rate of 93 in/min, a voltage of 100 V, and amperage of 63A. 4.2.4 T-test To examine the effect of noise factors on the response variables, a t-test was conducted for each noise factor. For
Table 2 T-test for air pressures effect on smallest diameter deviation Air pressure (PSI) Mean Variance Observations 45 0.019 2.4791 e-4 18 60 0.030 3.0446 e-4 18

Table 1 Optimal setting combination Selection Criteria Bevel Bevel S/N ratio Smallest diameter deviation Smallest diameter deviation S/N ratio Overall optimal setting Combination A1 A1 A1 A1 A1 B1 B2 B2 B2 B2 C2 C1 C1 C1 C1 D3 D2 D3 D3 D3

df Difference in mean Std. Error (difference between the means) t Stat t Critical two-tail (alpha=0.005) P(T <=t) one-tail

34 0.011 0.0055 2.0358 2.7284 0.0496

Deviation

768 Table 3 T-test for pierce times effect on smallest diameter deviation Pierce time (s) Mean Variance Observations Df Difference in mean Std. Error (difference between the means) t Stat t Critical one-tail (alpha=0.005) P(T <=t) one-tail 0.7 0.023 2.94 e-4 18 1.4 0.026 3.18 e-4 18

Int J Adv Manuf Technol (2009) 41:760769 Table 5 T-test for effect of pierce time on bevel Pierce time (s) Mean Variance Observations Df Difference in mean Std. Error (difference between the means) t Stat t Critical one-tail (alpha=0.005) P(T <=t) one-tail 0.7 43.944 107.4673 18 1.4 48.056 290.9967 18

34 0.003 0.0058 0.6575 2.7284 0.5153

34 4.112 4.7050 0.8738 2.7284 0.3884

air pressure, a t-test determines if the two levels (45 psi and 60 psi) will have a significant effect on smallest diameter deviation. The hypothesis is stated as H0 : 45 60 H1 : 45 6 60 The t-test result is shown in Table 2. From the test result it can be seen that the Abs (t Stat)=2.0358 < t critical twotail (alpha=0.005)=2.7284, or equivalently P (T <= t) onetail=0.0496 > alpha=0.005. The statistical conclusion is that the difference in mean from the two air pressure levels is not significant, so the hypothesis Ho can not be rejected. Therefore, we can conclude that air pressure does not significantly affect the smallest diameter deviation. Similarly, three t-tests were conducted for the effects of pierce time to smallest diameter deviation (Table 3), air pressure to bevel (Table 4), and pierce time to bevel (Table 5). From Table 3, it can be seen that the Abs (t Stat)=0.6575 < t critical two-tail (alpha=0.005)=2.7284, or equivalently P (T <= t) one-tail=0.5153 > alpha=0.005. Therefore, pierce time does not significantly affect the small radius. From Table 4, it can be seen that the Abs (t Stat)=1.2667 < t critical two-tail (alpha=0.005)=2.7284,
Table 4 T-test for effect of air pressure on bevel Air pressure (PSI) Mean Variance Observations Df Difference in mean Std. Error (difference between the means) t Stat t Critical two-tail (alpha=0.005) P(T <=t) one-tail 45 43.056 110.5 18 60 48.944 278.5 18

or equivalently P (T <= t) one-tail=0.2139 > alpha=0.005. Therefore, the air pressure does not significantly affect the bevel magnitude. From Table 5, Abs (t Stat)=0.8738 < t critical two-tail (alpha=0.005)=2.7284, or equivalently P (T <= t) one-tail=0.3884 > alpha=0.005. Therefore, the pierce time does not significantly affect the small radius. 4.3 Implementation After the overall optimal setting combination for the plasma cutter was identified, a confirmation run was conducted. With this optimal setting (A1-B2-C1-D3), 30 cuts were made to test the smallest diameter deviation and the bevel magnitude. The hardware passed easily through all 30 cuts, meaning no rework was needed. In addition, as the results shown in Table 6 indicate, the mean value and standard deviation of the confirmation run data were smaller than those before Taguchi design was conducted. 4.4 Control The optimal setting combination was sent to Company As production department. The Six Sigma team is now trying to uncover other possible causes of unacceptable deviation so that other Taguchi experiments can be conducted for continuous improvement of the process. For example, if defects occur later following the optimal condition, the Six Sigma team will follow the DMAIC procedure (Fig. 1) to pursue the next cycle of process improvement. Taguchi

Table 6 Confirmation run data Before Taguchi design Smallest diameter deviation Bevel =0.024 =0.0173 =46 =14.0672 Conformation run =0.006 =0.0032 =35.2 =4.5166

34 5.888 4.6491 1.2667 2.7284 0.2139

Int J Adv Manuf Technol (2009) 41:760769

769 Acknowledgement This project was partially funded by the Iowa Center for Industrial Research and Service, through a grant from the Department of Commerce NIST Manufacturing Extension Partnership.

parameter design for optimal condition could then be executed again.

References 5 Conclusion Six Sigma and lean manufacturing are powerful strategies for transforming an enterprise and escalating its competitiveness. The fact that Six Sigma and lean manufacturing can successfully save time and cut costs are important considerations, especially for small- and medium-sized enterprises with limited resources. This paper presented the application of the Taguchi method to optimize the roundness of the holes cut by an aging plasma-cutting machine. Using the orthogonal array in the experiment design for four factors, the Taguchi method reduced the experiment set-up from 81 parameter combination settings (34) in DOE to an L9 setting. With two noise factors included in the Taguchi experiment design, 36 total experiments are conducted. The optimal setting combination received from Taguchi experiment design is A1B2C1D3 (small for tip size, 93 in/min for feed rate, 100 V for voltage, and 63A for amperage), which maintains the existing feed rate for productivity and improves quality of products. The optimal setting combination gave no defects from the 30 plasma-cut holes in the confirmation run. In addition, the recommended setting combination was well received by Company A, and the problem with defects situation has been much improved. With the reduced time and cost, Taguchi experiment design has again demonstrated its effectiveness in achieving Six Sigma and lean paradigm.
1. Fowlkes WY, Clyde CM (1995) Engineering methods for robust product design using Taguchi methods in technology and product development. Addison-Wesley, Reading 2. Ghani JA, Choudhury IA, Hassan HH (2004) Application of Taguchi method in the optimization of end milling parameters. J Mater Process Technol 145(1):8492 Jan 1 3. Zhang JZ, Chen JC, Kirby ED (2007) Surface roughness optimization in an end-milling operation using the Taguchi design method. J Mater Process Technol 184(13):233239 Apr 12 4. Davim JP, Reis P (2005) Damage and dimensional precision on milling carbon fiber-reinforced plastics using design experiments. J Mater Process Technol 160(2):160167 Mar 20 5. Kirby ED, Zhang Z, Chen JC, Chen J (2006) Optimizing surface finish in a turning operation using the Taguchi parameter design method. Int J Adv Manuf Technol 30(1112):10211029 October 6. Palanikumar K, Karthikeyan R (2006) Optimal machining conditions for turning of particulate metal matrix composites using Taguchi and response surface methodologies. Mach Sci Technol 10(4):417433 Dec 1 7. Deng C-S, Chin J-H (2005) Hole roundness in deep-hole drilling as analysed by Taguchi methods. Int J Adv Manuf Technol 25(5 6):420426 March 8. Tsao CC (2007) Taguchi analysis of drilling quality associated with core drill in drilling of composite material. Int J Adv Manuf Technol 32(910):877884 April 9. Bagci E, Ozcelik B (2006) Analysis of temperature changes on the twist drill under different drilling conditions based on Taguchi method during dry drilling of Al 7075-T651. Int J Adv Manuf Technol 29(78):629636 July 10. Tsao CC, Hocheng H (2004) Taguchi analysis of delamination associated with various drill bits in drilling of composite material. Int J Mach Tools Manuf 44(10):10851090 August

You might also like