MGT 302 Mod 6
R Chart
-Variables control chart -data to collect: individual unit -nature of individual data: continuous numerical data (ex: in a sample of 5 laptops, each laptop is an individual unit. The weight of each laptop is individual data, X, which is continuous numerical data.) -data to plot: sample statistics R (sample range = largest - smallest individual data in the sample) (ex: R = highest laptop weight - lowest laptop weight in the sample)
Attributes Control Chart: C Chart
-data to collect: individual unit -nature of the individual data: each unit can have multiple defects (in a sample of 5 students, each student (individual unit) can have 0,1,2..or 100 wrong questions (defects) out of 100 questions -data to plot: C (the # of defects in the entire sample) (ex: total # of wrong questions from the 5 students in the sample, C could vary from 0-500) -plot a C chart when you are interested in the # of defects
Attributes Control Chart: P Chart
-data to collect: individual unit -nature of the individual data: each unit in the same is either defective or not (0 or 1 decision) (ex: in a sample of 5 students, each student (individual unit) either fails or passes an exam) -data to plot: p (proportion of defective units in the sample) (ex: p = # of students who failed the exam/# of individual units in the sample) -plot a P chart when you are interested in the # of defective units
X bar Chart
-variables control chart -data to collect: individual unit -nature of the individual data: continuous numerical data (ex: in a sample of 5 laptops, each laptop is an individual unit. The weight of each laptop is individual data, X, which is continuous numerical data.) -data to plot: sample statistics, X-bar (sample mean (ex: avg. of individual data, X, in the sample)) (ex: for a sample of 5 laptops, we plot the avg. weight of the 5 laptops in this sample, NOT the individual weight for each laptop).
If the lower control limit for the p chart is a negative value, then which of the following becomes the new value? 1 1.5 0 0.5
0
For the 9 weeks you randomly selected in the previous problem, the number of days in each week that the owner received complaint(s) is as follows: 2, 0, 1, 1, 4, 2, 0, 3, and 1 (assuming five days a week). What would be the LCL and UCL of the appropriate control chart? (Assuming z value of 3 is used to construct UCL and LCL).
0 and 0.93 (Since you are interested in the number of days owners received complaints, you are interested in the number of units defective. Therefore, you should plot a p chart. Each day you only care whether complaints were received and you don't care how many complaints were received. Therefore for each unit (i.e. day) this is a 0 (no complaint) or 1 (receiving complaints) decision. Each week represents one sample. The sample size is 5. To construct a p chart, first you need to calculate p value for each sample. Second, use the formula for p-chart to calculate UCL and LCL. Each of the nine numbers in this problem represents the number of defective units in each sample.)
Red Top Cab Company receives several complaints per day about the behavior of its drivers. You randomly sampled 9 weeks. The overall average number of complaints per week received from irate passengers over the 9 weeks is 6. What would be the LCL and UCL of your quality control chart if your interest is the number of complaints? (Assuming z value of 3 is used to construct UCL and LCL).
0 and 13 (Every claim represents one defect. Since you are interested in the number of complaints (i.e., the number of defects), a C chart should be used. Every day (i.e, each unit) can have one or multiple complaints or no complaints -- not a 0-1. Each week represents one sample. The number of weekly complaints is the C value for each sample. The overall average number of complaints per week is C-bar, namely, average of the C values over the 9 samples. Therefore, UCL = 6+3*sqrt(6), LCL = 6-3*sqrt(6) --> 0.)
The quality assurance manager is assessing the capability of a process that puts pressurized grease in an aerosol can. The design specifications call for an upper specification limit of 65 pounds per square inch (psi) and a lower specification limit of 55 psi. The process means is 61 psi and the standard deviation is 2 psi. What is the Cpk value of this process?
0.67
What are the two fundamental operational goals of Total Quality Management?
1. Careful design of the product or service (focus on the design quality) 2. Ensuring that the organization's systems can consistently produce the design (focus on conformance quality)
General Procedure for SPC
1. Periodically collect samples from a process and prepare the data (sample statistics) to be plotted. 2. Compute the center line (CL). 3. Compute the upper and lower control limits (UCL, LCL), 4. Draw the CL, UCL, and LCL on the chart. 5. Plot each sample data on the chart. 6. Interpret control charts (the state of the process - is the process in control? the nature of variations - whether assignable variations exist? the quality level - any change? desireable or not? actions - whether the process should be stopped for investigation)
What two things does Cpk imply?
1. The location of the process mean 2. The probability that the process will product defects (falling outside the specifcation range)
What are the two types of Control Charts?
1. Variable Charts (Continuous Numerical Data (i.e. data that are measurable)) 2. Attributes Charts (Categorical or Discrete Numerical Data (data that can only be counted))
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. What is the maximum # of defects per sample?
15 (3*5)
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. What is the maximum # of defects per bill?
3
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. What is the maximum # of defective units in the sample?
5
Factors affecting Cpk (Slide 58)
???
Statistical Process Control (SPC)
??? (SPC utilizes sample data to decide whether or not a process is operating as expected while the product or service is being produced).
Statistical Quality Control (SQC)
A number of different techniques designed to evaluate how well the process meets the specifications set by customers or design. Specifications are often described as a range.
When is a process capable?
A process is capable when the mean and standard deviation of the process are operating such that the UCL and LCL are acceptable relative to the USL and LSL. (UCL and LCL are the same as USL and LSL or within USL or LSL.)
What will you be able to conclude when the Cpk value is 0? The process is not capable to meet the design specification The probability that the process will produce defects is 50% Shifting the process mean towards the center of the specification range will improve the Cpk All of the above
All of above is correct
What are the four types of costs associated with the cost of quality?
Appraisal costs Prevention costs Internal failure costs External failure costs
Which of the following is variation that is caused by factors that can be clearly identified? Normal variation Common variation Random variation Assignable variation
Assignable variation
When the product or service can have more than one defect per unit, which of the following is the most appropriate chart? Mean chart Range chart P chart C chart
C chart
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. If you are interested in the # of defects, which chart will you use?
C chart
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. What data would you plot if you're interested in the # of defects?
C for each sample
Which quality term refers to the degree to which the product or service design specifications are met? Dimension Conformance Engineering Design
Conformance
What dimension of quality is SQC evaluation?
Conformance quality
What is the range of Cpk?
Cpk can be any number, but the higher the Cpk, the more capable the process is.
What term refers to the inherent value of the product in the marketplace? Performance quality Conformance quality Reliability Design quality
Design quality
Which type of quality cost is the most costly?
External failure costs are the most expensive ones.
True or False: A low defect rate indicates a high design quality.
False
True or False: Statistical quality control focuses on evaluating design quality.
False
True or False: The best way to control quality is by inspection.
False
True or False: The process capability index is calculated to determine whether the process is in control.
False
Which of the following statement is NOT correct about conformance quality? A high conformance quality implies a low defect rate. High conformance quality indicates high precision and consistency. High conformance quality suggests low design quality. Statistical quality control evaluates conformance quality.
High conformance quality suggests low design quality.
Which of the following is not an example of when to use a c-chart? Inspecting the diameter of holes drilled into a pipe Inspecting a roll of fabric for snags Inspecting a piece of lumber for knotholes Inspecting a car's paint finish for blemishes
Inspecting the diameter of holes drilled into a pipe. (Diameter is continuous numerical data. Hence variable charts not attribute charts are required here.)
Which type of quality cost deals with scrap, rework and repair? Internal failure costs Appraisal costs External failure costs Prevention costs
Internal failure costs
Who is responsible for design quality?
Mainly the cross-functional product design team
Who is responsible for conformance quality?
Mainly the manufacturer's responsibility. Improving process technology (ex: using robots or more precise equipment) will help.
Expenditures related to achieving product or service quality include all of the following except: Manufacturing costs Failure costs Prevention costs Appraisal costs
Manufacturing costs
Suppose you are interested in the weight of ipad air produced in a manufacturing process. The CL, UCL, and LCL for the X-bar and R charts are as follows:X-bar chart: CL = 1, UCL = 1.03, and LCL = 0.97R-chart: CL= 0.05, UCL = 0.08, and LCL = 0.02A new sample is collected. Assuming the sample size is large enough. The weight for each ipad air in this new sample is: 1.05, 1, 1.01, 0.99, and 0.98. Based on the new sample and the information on the control limits, what conclusion would you be able to make?
No assignable variations have been detected yet and therefore no action should be taken now. (First, you need to calculate sample mean and sample range of this particular sample. The decision is based upon sample mean and sample range not the individual X value.)
What are the two types of Attributes Control Charts?
P Chart C Chart
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. What data would you plot if you are interested in the # of defective bills?
P for each sample
Plots of sample ranges indicate that the most recent value is below the lower control limit. What course of action would you recommend?
Precision (or dispersion) is not in control; investigate for the cause of the assignable variations and then adopt the cause to bring the process under control. (Out of control does not mean the quality is worse. This is a R chart as sample ranges were plotted. Lower R is better.)
With 99.7% confidence, UCL = ?
Process Mean + ( 3 * Standard Deviation)
With 99.7% confidence, LCL = ?
Process Mean - (3 * Standard Deviation)
What is the measure of process capability?
Process capability index: Cpk
Which of the following conclusion is NOT correct if we find seven consecutive points decreasing in a C chart? Quality is getting worse The process is out of control The process has assignable variation The process does not operate as expected The process has shifted in a nonrandom way
Quality is getting worse
What are the two types of Variable Control Charts?
R Chart X bar Chart
Which of the following is defined as the difference between the highest and lowest numbers in a sample? Mode Median Range Standard deviation
Range
Process Limits
Range of variation that a process is able to maintain with a high degree of certainty (i.e. confidence). They reflect the status of an in-control process. (ex: LCL: Lower Control Limit; UCL: Upper Control Limit)
Which of the following is an example of internal failure cost associated with quality? Quality training Scrap Customer complaints Material inspection
Scrap
The quality assurance manager is assessing the capability of a process that puts pressurized grease in an aerosol can. The design specifications call for an upper specification limit of 65 pounds per square inch (psi) and a lower specification limit of 55 psi. The process means is 61 psi and the standard deviation is 2 psi. Which of the following statement is NOT true? The process width (i.e. the difference between UCL and LCL) is 12. The current process is not capable. Shifting the means of the current process to 60 would decrease the Cpk value. Decreasing the standard deviation of the current process would increase Cpk value.
Shifting the means of the current process to 60 would decrease the Cpk value. (The center of the specification limits is 60. Shifting process means from 61 to 60 (i.e. towards the center) will increase Cpk.)
A process has a Cpk value less than 0. Which of the following conclusion can be made? There is more than 50% of chance that this process will produce defective items. There is 100% of chance that this process will produce defective items. There is 50% of chance that this process will produce defective items. There is a 0% of chance that this process will produce defective items.
There is more than 50% of chance that this process will produce defective items. (When Cpk is less than 0, it means either the process mean is greater than USL or smaller than LSL. In either case, more than 50% chance products will fall outside the specification limits.)
True or False: For continuous data, you need to construct both X bar-chart and R-chart.
True
True or False: R charts are useful for detecting any nonrandom changes in variations, standard deviations, precision, dispersion of the process' output. X-bar charts focus on detecting any nonrandom changes in the means of the process' output.
True
True or False: The concept of quality at source indicates that the person who does the work should take responsibilities ensuring conformance quality of his or her output.
True
True or False: The optimal design quality level of a product depends on the target customer.
True
True or False: The standard deviation (Sp) of the C chart is the square root of the C-bar.
True
True or False: When assignable variations are identified, one should manage to eliminate the assignable variations because the process is no longer in control and action is always called for in order to bring the process to control.
True
When is the process in control?
When common variation is THE type of variation present in a process, the process is in control or stable. This means that output of the process follows an expected distribution (i.e. the process behaves as expected).
Quality Management Philosophy
build quality into the product in the first place and use statistical means for controlling and improving quality
Internal Failure Costs
costs of defects incurred within the system (ex: scrap, repair, rework, downtime)
External Failure Costs
costs of defects that pass through the system (ex: warranty costs, out-of-warranty repairs/replacements, customer complaints, product liability, transportation losses)
Appraisal Costs
costs of the inspection, testing, and other tasks to ensure that the product or process is acceptable (ex: materials inspection, supplies inspection, reliability testing, laboratory testing)
Variations
deviation in the output of a process
Assignable Variation
deviation in the output of a process that can be clearly identified and managed; source of variation is identifiable causes (Honda Video Ex: the supplier that provides a certain feeder that is looser than the ones from another supplier)
Common Variation
deviation in the output of a process that is random and inherent in the process itself; source of the variation is a large number of unidentifiable and random causes that occurred within a process (causes are difficult to trace; each cause contributes only a small part of the observable variation)
High design quality
high level of standards (specification)
The (higher/lower?) the Cpk, the fewer areas of the distribution fall outside the specification range, the fewer chances of producing defects, and the more capable.
higher
Choice of control charts is based upon the nature of the data pertaining to ___________ units in a sample.
individual
Design Quality
inherent value of the product in the marketplace; refers to the features of the product/service relating directly to design issues; determine the needs of target customers and set design specifications accordingly for design quality (ex: iPad Air has a higher design quality than iPad2 due to the new technology)
High conformance quality = ______ defect rate
low
Total Quality Management
managing the entire organization so that it excels on all dimensions of products and services that are important to the customer
You have developed criteria which say that a bill can be defective in any of 3 ways: incorrect total, incorrect payee address, and incorrect price per unit. You inspected a sample of 5 bills. If you are interested in the # of defective bills, which chart will you use?
p chart
Defect
part that doesn't fall within specification limits (ex: a 42oz M&M bag with only 35oz has a conformance quality problem)
Specification Limits
range of variation that is considered acceptable by the designer or customers (ex: LSL: Lower Spec Limit, USL: Upper Spec Limit)
Control charts plot __________ ____________.
sample statistics
What is the pattern of an assignable variation?
samples exhibit different kinds of variabilities; output of the process deviated from the expected distribution; these distributions are different and are no longer identical; the process has changed in a nonrandom way
What is the pattern of a common variation?
samples exhibit the same kind of variabilities; individual output values vary, but all follow a certain distribution; hence, one will be able to predict how likely certain values of the output will occur
High conformance quality
small variations, high consistency and precision, a low defect rate
Prevention Costs
sum of all costs to prevent defects (ex: quality training, reliability consulting, pilot production runs, systems development)
Process Capability
the ability of a process to meet its specification
Conformance Quality
the degree to which the product or service design specifications are met; meet the design specifications, regardless of whether the design standards (specifications) are high or low
Quality at Source
the person who does the work takes responsibility for making sure that his or her output meets specifications (i.e. conformance quality)
The process is capable when Cpk ____ 1.
≥ (greater than)