1 IE 361 Exam 2 Fall 2009 I have neither given nor received unauthorized assistance on this exam. ________________________________________________________ Name Signed Date ________________________________________________________ Name Printed2 This exam consists of 20 multiple choice questions. There is a single best answer for each question. Circle EXACTLY ONE response for each question on this answer sheet. 1. A B C D E 2. A B C D E 3. A B C D E 4. A B C D E 5. A B C D E 6. A B C D E 7. A B C D E 8. A B C D E 9. A B C D E 10. A B C D E 11. A B C D E 12. A B C D E 13. A B C D E 14. A B C D E 15. A B C D E 16. A B C D E 17. A B C D E 18. A B C D E 19. A B C D E 20. A B C D E3 1. "Special cause variation" a) is exactly that kind of process variation that control charting is meant to detect. b) must be eliminated by appropriate process intervention in response to "out of control" signals before it makes sense to think of a process as physically stable. c) is also known as "random" variation. d) All of responses a) through c) are correct completions of the sentence. e) Exactly 2 of responses a) through c) are correct completions of the sentence. 2. In a case where individual measurements x are to be plotted on a control chart (the plotted statistic is just Qx= ) • standards given control limits should be based on engineering specifications. • one is essentially operating an x chart with 1n=. • retrospective control limits should be set at the average measurement plus or minus three times the grand sample standard deviation of the measurements. • engineering specifications for x can be derived from the plotted values as long as no out-of-control signals are issued. a) None of the above is a correct completion of the sentence. b) Exactly one of the above is a correct completion of the sentence. c) Exactly two of the above are correct completions of the sentence. d) Exactly three of the above are correct completions of the sentence. e) All of the above are correct completions of the sentence. 3. Every morning before a day's work begins in a testing lab, the same standard specimen is tested on a particular laboratory instrument. The single measured value produced is then plotted on a control chart with control limits set at measurement(true value for the standard specimen) 3σ± (the value of measurementσ was derived from repeated measurement of the standard specimen in a single session). This practice • is a form of "measurement process" monitoring. • is primarily effective at detecting a change in the instrument's bias, but can be expected to also eventually detect a degradation in instrument precision. • could be made more practically useful by replacing standards-given charting with retrospective charting. • is intended to detect a change in behavior of the instrument. a) None of the above is a correct completion of the sentence. b) Exactly one of the above is a correct completion of the sentence. c) Exactly two of the above are correct completions of the sentence. d) Exactly three of the above are correct completions of the sentence. e) All of the above are correct completions of the sentence. 4. Large sample sizes in control charting a) allow relatively quick detection of relatively small process changes. b) are relatively expensive. c) are essential to guarantee high rates of conformance to engineering specifications. d) Exactly 2 of a), b), and c) are correct completions to the sentence. e) All of a), b), and c) are correct completions to the sentence.4 Below are sample means and standard deviations from 10 samples of size 6n=. Use the information from these samples to answer questions 5 through 7. Sample 1 2 3 4 5 6 7 8 9 10 Sum x 4.20 4.62 3.27 6.27 4.96 7.31 6.77 6.99 5.58 8.41 58.38s 1.64 1.26 .72 1.19 2.11 2.10 2.03 1.87 1.96 1.93 16.81 5. Consider standards given control charting for both and xs, with standards 5.0μ= and 2.0σ=. a) The x chart and s chart both produce out-of-control signals. b) The x chart produces out-of-control signals, but the s chart does not. c) The x chart produces no out-of-control signals, but the s chart produces signals. d) Neither chart produces out-of-control signals. 6. Consider retrospective control charting for both and xs . a) The x chart and s chart both produce out-of-control signals. b) The x chart produces out-of-control signals, but the s chart does not. c) The x chart produces no out-of-control signals, but the s chart produces signals. d) Neither chart produces out-of-control signals. 7. Ignore any lack of stability that you found indicated by the s values above and use these values to make an estimate of σ. Based on this estimate, in turn set an upper control limit for future ranges (R) based on samples of size 4n = . (Remember that the samples represented in the table were of size 6n =.) This RUCL is closest to a) 3.84 b) 7.90 c) 8.30 d) 8.54 e) 8.97 8. Below are some statements about various kinds of "limits" encountered in a course like IE 361. How many of them are true? • Statistical tolerance limits change when engineering specification limits change. • 95% confidence limits for a process mean are inside 95% prediction limits for a new individual value from the process. • Standards given control limits for an 5n= sample mean can also be used to make 99% predictions for a new individual value from the process. • Control limits concern the establishment of process stability, while the use of statistical prediction and tolerance limits assumes that stability is present. a) All 4 are correct. b) Exactly 3 are correct. c) Exactly 2 are correct. d) Exactly 1 is correct. e) None is correct.5In a project intended to eventually improve the operation of a large heating, ventilation, and air conditioning (HVAC) system, the numbers of alarms issued by 2CO sensors in the system are recorded for 10 consecutive weeks and are reproduced below. Use this scenario and these data to answer questions 9 through 11. Week 1 2 3 4 5 6 7 8 9 10 Sum Sensor Alarms 4 5 4 10 6 8 7 7 7 6 64 9. The data in the table a) provide no evidence of change over time (variation) in the 2CO sensor alarm rate in the 10 week period. b) provide clear evidence that the system is not operating consistently at a designed-for-rate of 2 sensor alarms per week. c) are