Lean Six Sigma Black Belt Sample Paper 2: Measure 1. An operational definition: is a clear, concise and unambiguous definition of the CTQ provides a unified understanding of the data for all involved before the data are collected includes the formula (if any) that is used to calculate the metric All the above2. One of the following statements is NOT TRUE: For long term capability, the variation is generally higher than that with short term capability. Short-term capability is calculated from the data collected from stable processes, monitored for approximately 20 to 30 subgroups Measurement system reliability does not affect process capability assessment For such a short interval, the variability is often relatively small due to a focus on specific equipment, a set of trained people, homogeneous material, the same measuring system, etc.3. September 23, 1999, after a 286-day journey from Earth to Mars, NASA lost the $125 million Mars Climate Orbiter, after the spacecraft entered orbit 100 kilometers closer than planned. NASA used the metric system, while its partner Lockheed Martin, the organization that designed the navigation system, used English units. This is a classic example of the absence of____________ an operational definition a measurement system due diligence risk analysis4. Measurement value= True value True value + Measurement error True value + Systematic error True value + Calculation error5. In case of “Data Collection”, which of the following has the advantage of high data integrity and opportunity to seek clarification? Mystery shopping Online surveys Focus groups All the above6. Which of the following is not correctly matched? Repeatability: Variation in measuring equipment when measured by one appraiser in the same setting at the same time Bias: Difference between absolute value and true value Stability: Accuracy of measurement at various measurement points of the measuring range in the equipment Reproducibility. Variation in measurement when measured by two or more appraisers’ multiple times7. Formula for DPO is - a. Defects/Units *1000000 b. Defects/Opportunities*1000000 c. Defects/Opportunities d. Defectives/Opportunities8. The correct sequence for conducting Gage R&R study: Equipment calibration>sample identification for study> data collection> calculation Sample identification for study> data collection> calculation> equipment calibration Sample identification for study> data collection> Equipment calibration> calculation Equipment calibration> sample identification for study> calculation> data collection9. All the statements below are false, EXCEPT: In both Pp and Ppk, the lower the value, the better. Cp is always equal to or smaller than Cpk A Ppk below 1, indicates the process average is outside one of the engineering specifications The Pp and Ppk values will be equal if the process is perfectly centered.10. The following is not an error in Gage R&R study: It is recommended to even pick samples outside the specification limits Randomize the samples during measurement Involve appraisers who perform the measurements in the process on a regular basis All the above11. In case of automated data collection with no human intervention in the measurement, operational definition is not required MSA is not required Logical validation can be performed MSA is impossible12. Gage R&R study for a measurement system gave the following outcome, make a suitable inference: Measurement System is accepted but with caution Measurement System is accepted Measurement system is rejected More data is required13. All the statements below are correct, EXCEPT: Cp, Cpk ≥ 1.33 shows that the process is capable A Cp, Cpk value of 1.0 means that the process barely meets the specification A Cp, Cpk value less than 1.0 means that the process is producing units outside engineering specifications Abnormally high Cp, Cpk (> 3) shows that the specification is very tight14. One of the following does not warrant an action: Process in statistical control and meets specifications Process not in statistical control but meets specifications Process in statistical control but does not meet specifications Process not in statistical control and does not meet specification15. While performing Attribute Agreement Analysis (AAA) which of the following is TRUE? AAA can only be performed if the standard is known To assess the ability of an appraiser to consistently evaluate the same sample, each appraiser should rate each sample at least twice in random order. More replicates (Trials) for the same samples is a better choice while conducting AAA rather than more samples To ensure that the data collection order does not influence the results, each appraiser should evaluate all samples in a non-random order.16. The correct sequence for activities given below is: Measurement System verification> Process stability evaluation> Rational subgrouping> Process capability evaluation Process stability evaluation> Rational subgrouping> Measurement System verification> Process capability evaluation Measurement system verification> Rational subgrouping> Process stability evaluation > Process capability evaluation Rational subgrouping> Measurement System verification> Process capability evaluation> Process stability evaluation17. Which of the following combinations is incorrect? Nominal (Binary) data from appraisers – Fliess’ Kappa statistics Ordinal data from appraisers - Kendall’s coefficient of concordance Gage R&R- % Team accuracy None of the above18. A document requires 10 essential entries for further processing. In a sample of 100 such documents, defects (missing entries/incomplete entries/incorrect entries) were evaluated. 20 defects were found in 20 documents. What is DPMO, DPU and Yield respectively? 20000, 1, 98% 20000, 0.2, 98% 0.02, 0.2, 98% 20000, 1, 90%19. If your data is not normal, then the most appropriate sequence to perform process capability is -i. Transform the non-normal data to normal data using Box-Cox transformation or Johnsons transformationii. Identify the non-normal distribution that fits your data and perform capability analysis for non-normal dataiii. One must investigate the reason for non-normality to check for erroneous data/extreme values i -> ii -> iii iii -> i -> ii iii -> ii -> i i -> iii -> ii20. Choose the CORRECT one: Z.benchST : Overall standard deviation Z.benchLT : Within standard deviation Zshift: Difference between Z.benchST & Z.benchLT Z.benchST = Z.benchLT + 1.5, always21. What is the probability of observing at least one defective in a random sample of size 10, drawn from a population that has been producing, on the average, 10 percent defective units? 0.65 0.35 0.73 0.2722. Errors in invoice’s follow Poisson distribution with DPU of 0.5. What the probability that an invoice will have zero errors? 0.5 0.6 0 123. Given process average = 1.65, standard deviation =0.05, assuming statistical control and a normal population, what proportion of the population will meet specifications of 1.65 ± .05? 68.27 % 99.73 % 31.73 % 100.0 %24. If Z_{LT}=1, what will be Z_{GOAL} if we target 50% reduction in defects? 0.5 1.41 2 1.525. A process that is normally distributed, USL = 24 and LSL = 18. This process has a mean of 20, short-term standard deviation of 1 and long-term standard deviation of 1.5. Calculate Cp, Ppk 1, 0.67 1, 0.44 0.67, 0.44 0.67,0.67