Datastor Company’s Quality problems and Their Solutions

Questions: Has DataStor Company had a quality problem (4 returned shipments in 20 days)? Can the problem with unaccepted shipments simply be caused by random variation? What evidence leads you to your conclusion? Attach supporting evidence from your data analysis. (Hint: you need to think about the following probabilities)? Answer: First of all, we have to draw the control chart for checking of the quality of the product. We have to use the xbar and R control chart for this purpose. We have to observe this xbar and R chart and we have to check whether any product below the lower control limit or above the upper control limit. The control chart for product of DataStor company is given as below: The above xbar shows that process is in statistical control. R chart also shows that the points are within statistical limit but there is a specific pattern and almost observations in the R chart is below the Rbar line. If the DataStor DS1000 hard drive production process at DataStor Company is in control, what percentage of the drives produced would be considered in nonconformance by Four-D? Solution: If the DataStor DS1000 hard drive production process at DataStor Company is in control, this means, all points are within 3sigma limits. Then we know that the probability or percentage of the drives produced would be equal to 0.3% approximately. If the DataStor DS1000 hard drive production process at DataStor Company is in control, how often would shipments be found unacceptable by Four-D? Solution: If the DataStor DS1000 hard drive production process at DataStor Company is in control, then there would be 3 in 1000 shipments found unacceptable by Four-D. What is the probability of four rejected shipments in the past twenty days assuming that the process has been in control all this time? Solution: If the process is out of control, then product will be rejected. But sometimes process is in statistical control but product or shipments will be rejected. This is due to specific pattern in xbar or R chart. The required probability is given as 0.003^4 = 0.000 approximately. Why were the defective products not detected before the shipments?How can the problem be fixed? Solution: The defective products are not detected before the shipments because total numeration of the product is not possible. Census inspection is very costly and for avoiding this cost, the method of random sample for quality check is selected in the company. So due to this reason, the defective products not detected before the shipments. If the problem with unaccepted shipments is due to an increase in drive nonconformances at DataStor, when were the low quality products produced (e.g., weeks,shifts)? What evidence leads you to your conclusion?Attach supporting evidence from your data analysis. Solution: For supporting this evidence, the R chart shows the specific pattern and thats why the process is out of statistical control. For statistically control process we need the random pattern of sample points within the control charts. We need to find out exact reason behind this by analysing the data. We draw the control chart for the PDQ on the basis of shifts. So improvements in shift pattern is necessary for gaining more quality for the product.