Lean Six Sigma Green Belt Minitab Home Work

C. Assuming that this was a screening experiment, what do you do next and/or what do you tell your boss about both your existing data and your ideas for the future? (5)

2. Vermont invests about $4.5 million/year in anti-smoking efforts. Out of 100,000 VT high school students 17,600 smoke. In New Hampshire no money is spent on anti-smoking campaigns and 20,800 out of 100,000 HS students smoke. Using the Poisson distribution, can Vermont argue that their program has some positive effect? (15)

2. An election poll concluded that two candidates were tied at 50% of polled voters each. The number of people polled was 500. What is the statistical error at a 95% confidence level? Comment on the results. Hint: Use the continuous approximation to the binomial distribution. This type of result will be similar to when polls are reported on the news and the newscaster will say that the error is plus and minus 3%. (15)

IV: SPC

1. A. Discuss the differences between special cause and common cause failures or variation, giving at least one example of each. How might you see the difference in your measurements? (5) B. Discussion the difference between attribute and variables data. How do they work together in an effective SPC Program? (5) C. What is a Pareto Chart? How is it used in a continuous improvement program? (5)

2. The USGA announced that they are going to be very “tough” in enforcing the rule that the maximum coefficient of restitution (COR) for golf club drivers be 0.83. (some drivers were measured by them as 0.85, this COR gives a drive 5 yards or so advantage over one that is “in spec.” Assume that the lower spec limit is 0.70) The USGA will allow the upper spec limit on any manufacturer’s driver to be 0.83 as long as the Cpk is at least 2.0. Ping has developed a very repeatable process for making the driver heads that they believe can achieve a Cpk of 2 with the average COR being 0.828.

a) If this is true what must  be? (5)

b) Ping is encouraged as they believe their competitors is at least 0.004. If this is true, what COR average must the competitor aim for to have a Cpk = 2. (5)

c) For every 0.01 loss in COR the driver loses 2.5 yards. How much less yardage will the competitions driver hit a ball on average as compared to Pings? (5)

d) Ping manufactures 10,000 drivers. The data are in the spreadsheet in tab “Golf Club COR” . Analyze the data. Assume with subgroup size = 1. Did they achieve their goal? Comment on the data Re: normality, etc. Are there Shewhart violations? If so discuss them. State any assumptions. Copy and paste appropriate data and/or graphs. Use an upper spec of 0.83 and a lower spec of 0.82.(15)

V: DOE

A full factorial experiment is proposed to optimize stencil printing for a 0.3mm process. The assembly will be in Singapore, so management wonders if temperature and humidity control will be needed. The stencil aperture is 6x4x30 mils, +/- 10% of this volume is the proposed spec. The factors are Type 4 paste (1=ACME, 2=AJAX, 3=Kostco), stencil (efab, laser), humidity (30, 45, 60 RH), temperature (20, 25, 30 C). The two replicate data are in the worksheet “Solder Paste DOE.” Set up the experiment with the factors in this order and do not randomize runs. Analyze the data, look for interactions and non linear effects. Comment on the results. What are your recommendations? There is a rumor in the industry that the AJAX paste does not print well at 20C. What do the data tell you in this regard? Do you propose further experiments, what are they? (40)

VI: General

Explain Lean, Six Sigma, and the limitation of each as a separate discipline in a way that your 15 year old neighbor would understand. (10)

VII: Lean 1. You are a Six Sigma practitioner working for a small local shop that sells yard tractors. The company has been losing market share recently, and it has been determined that this is due to poor customer satisfaction with the in-store repair department. You have just been assigned a project to improve the process of repairing tractors returned to store.

1) What is the most probable primary metric (or key process output variable) for this process? (5) ___________________________________

2) How would you begin to determine possible sources of waste in this process? (Select the best answer) (5)

a. Conduct a DOE to determine sources of variation at the factory.

b. Map the value stream of the repair process.

c. Run a 2 sample T test on process time before and after improvements.

d. Look at what has been done in the past when sales have fallen and repeat those actions.

e. Ask the owner what he would do.

3) Create a rough process map of the repair process, and identify the NVA, BNVA and VA steps (can be simply “step 1 step ”, no fancy graphics required). Comment on which could be eliminated or reduced. (10)

2. Think of an organization you work with or have worked with. Discuss how 5 S and the 7 mudas could be used to improve the operation. If the organization has implemented these tools, discuss the process and benefits. (20).