CQE Certified Quality EngineerExam Questions and Answers

In factorial design experiments, fractionating involves using a subset of the full set of experimental runs. A 242^424 full factorial experiment involves 16 runs, and fractionating this to 24−12^{4-1}24−1 results in 8 runs.

90% Fraction:

Incorrect, as this would imply a fractional design close to the full factorial, but it is not standard terminology.

75% Fraction:

The correct answer as 24−12^{4-1}24−1 implies a design that uses 816\frac{8}{16}168​ or 50% of the full factorial runs. However, in common terminology, a fractionated design resulting in fewer runs (half of the runs) is considered a 50% fraction. Therefore, this fractionation corresponds to a 50% fraction or half fraction.

Fraction:

Too vague, as it does not specify the fraction size.

Fraction:

Duplicates "fraction."

Therefore, the correct fractional reduction is to a 50% fraction, which is commonly used in design of experiments. The proper fraction designation in this case should be 50%, not 75%.

References: Design of experiments (DOE) principles.

Management commitment is the most crucial aspect of a successful quality improvement effort. Without the active involvement and support of management, quality initiatives are less likely to receive the necessary resources, attention, and prioritization. Management commitment includes setting clear quality goals, providing the required resources, and leading by example to foster a culture of continuous improvement. This ensures that quality improvement is integrated into the organizational strategy and receives ongoing support and reinforcement.

References:

Juran, J.M., & Gryna, F.M. (1993). Quality Planning and Analysis: From Product Development through Use. McGraw-Hill.

Crosby, P.B. (1979). Quality is Free. McGraw-Hill.

To calculate the 95% confidence interval for the proportion of nonconforming items in a sample, we use the formula for the confidence interval for a proportion: p̂ ± Z * √(p̂(1-p̂)/n), where p̂ is the sample proportion (80/2000 = 0.04), Z is the Z-score for a 95% confidence level (1.96), and n is the sample size (2000). Plugging in the values, we get the confidence interval as [0.0308, 0.0499]. Reference: "Statistics for Engineers and Scientists" by William Navidi, which explains confidence intervals and their calculations.

When developing a quality information system, the most important factor is understanding and addressing the needs of the end user. This ensures the system is functional, user-friendly, and meets the requirements for which it was designed. End-user needs dictate the system's features, usability, and overall effectiveness. Reference: ISO 9001:2015 - Quality management systems -- Requirements, which emphasizes the importance of customer focus in designing quality management systems.

A full factorial experiment involves testing all possible combinations of levels for each factor.

For a three-level (3 levels) and two-variable (2 factors) experiment, the number of runs is calculated as: 32=93^2 = 932=9

Therefore, the minimum number of runs required is 9.

References: "Design and Analysis of Experiments" by Douglas C. Montgomery discusses factorial design and the calculation of runs for different levels and factors.

Hoshin planning, also known as Hoshin Kanri or policy deployment, is a strategic planning process that aligns an organization's functions and activities with its strategic objectives. It involves setting idealistic goals and linking them to actionable work strategies through a systematic approach. This method ensures that all levels of an organization are working harmoniously towards common objectives.

The equation to calculate the upper control limit (UCL) for an average chart (X-bar chart) is X + D4 R. Here, X is the process mean, D4 is a control chart constant that depends on the sample size, and R is the average range of the samples. This formula helps in determining the upper threshold beyond which a process is considered out of control. Reference: "Introduction to Statistical Quality Control" by Douglas C. Montgomery, which provides detailed explanations on control charts and their calculations.

The constant failure rate per hour is calculated as the inverse of the mean time between failures (MTBF). Given an MTBF of 200 hours: Failure rate (λ) = 1 / MTBF = 1 / 200 = 0.005

Fault tree analysis (FTA) is a tool that begins at the point of failure and works back through the process to the components. FTA is used to identify the root causes of system failures by constructing a logical diagram of the various possible causes. This method helps in understanding the paths leading to a failure and identifying corrective actions. Reference: "Reliability, Maintainability, and Risk" by David J. Smith, which describes fault tree analysis and its application in failure analysis.

Severity in the context of software quality and risk management refers to the extent of impact a failure would have on a system or process.

It assesses the consequences of a failure, regardless of its likelihood or detectability.

High severity indicates significant detrimental effects, such as critical system failures or safety hazards.

References: The ASQ and FMEA (Failure Mode and Effects Analysis) methodology define severity as the measure of the seriousness of the effect of a failure mode.

In the provided FMEA table, the potential failure mode "No solder" has the highest severity rating (S = 9), making it the most critical failure that could affect the customer by causing an open circuit. Higher severity indicates a more significant impact on the customer, thus making it the most likely to be seen by the customer.

The correlation coefficient measures the strength and direction of a linear relationship between two variables. A correlation coefficient of -1.0 indicates a perfect negative correlation, meaning that as one variable increases, the other decreases.

References:

Moore, D.S., McCabe, G.P., & Craig, B.A. (2017). Introduction to the Practice of Statistics. W.H. Freeman.

Montgomery, D.C. (2009). Introduction to Statistical Quality Control. Wiley.

Under a defect classification system, a nonconformance that could cause a product's usability to be substantially reduced would be classified as major. Major nonconformances are significant issues that affect the product's functionality, performance, or customer satisfaction but do not pose a safety risk or cause complete failure. Reference: ISO 9001:2015, which includes guidelines for classifying nonconformances based on their impact on product quality and usability.

Calculate the degrees of freedom for error:dfError=Total df−Treatment df=24−4=20\text{df}_{\text{Error}} = \text{Total df} - \text{Treatment df} = 24 - 4 = 20dfError​=Total df−Treatment df=24−4=20

Calculate the sum of squares for error:ssError=Total ss−Treatment ss=643−333=310\text{ss}_{\text{Error}} = \text{Total ss} - \text{Treatment ss} = 643 - 333 = 310ssError​=Total ss−Treatment ss=643−333=310

Calculate the mean squares:

MSTreatment=ssTreatmentdfTreatment=3334=83.25\text{MS}_{\text{Treatment}} = \frac{\text{ss}_{\text{Treatment}}}{\text{df}_{\text{Treatment}}} = \frac{333}{4} = 83.25MSTreatment​=dfTreatment​ssTreatment​​=4333​=83.25 MSError=ssErrordfError=31020=15.5\text{MS}_{\text{Error}} = \frac{\text{ss}_{\text{Error}}}{\text{df}_{\text{Error}}} = \frac{310}{20} = 15.5MSError​=dfError​ssError​​=20310​=15.5

Calculate the F-statistic:

F=MSTreatmentMSError=83.2515.5≈5.37F = \frac{\text{MS}_{\text{Treatment}}}{\text{MS}_{\text{Error}}} = \frac{83.25}{15.5} \approx 5.37F=MSError​MSTreatment​​=15.583.25​≈5.37

If a value assigned to a unit is non-numerical, it is described as qualitative. Qualitative data describes attributes or characteristics that do not have a numerical value. Examples include descriptions such as color, type, or category.

References:

Montgomery, D. C. (2017). Design and Analysis of Experiments. John Wiley & Sons.

ASQ (American Society for Quality) - Types of Data: ASQ Glossary

Statistical Process Control (SPC) was originated by Walter A. Shewhart. Shewhart introduced the concept of control charts in the 1920s while working at Bell Laboratories. His pioneering work laid the foundation for the field of quality control and process management.References:

Statistical Method from the Viewpoint of Quality Control by Walter A. Shewhart.

ASQ Quality Press: The Certified Quality Engineer Handbook.

The Process Decision Program Chart (PDPC) is a tool used to anticipate potential problems that could interfere with project implementation and to develop countermeasures to prevent or mitigate those problems. It is a structured method for mapping out every conceivable event and contingency that can occur in the course of a project, thus allowing project managers to plan for and avoid disruptions effectively.References:

ASQ Quality Press: The Quality Toolbox.

Total Quality Management by Dale H. Besterfield.

The total cost of external failures includes warranty, customer complaints, and product liability. Summing these costs:

Warranty: $2,000

Customer complaints: $6,000

Product liability: $1,000

Total external failure cost = $2,000 + $6,000 + $1,000 = $9,000

The Plan-Do-Check-Act (PDCA) cycle specifies the use of a feedback cycle. It is a continuous loop of planning, doing, checking, and acting, aimed at continuous improvement and problem-solving in quality management.

Standardized work involves documenting and following the best practices for performing tasks on the production line.

Job aids that outline step-by-step instructions ensure consistency and quality in task performance.

This practice helps in maintaining uniformity and reducing variability in the production process.

References: Lean manufacturing literature, such as "The Lean Manufacturing Pocket Handbook," describes standardized work as a core principle for ensuring efficiency and quality.

In value stream analysis, inspection is typically classified as non-value added. This is because inspection does not directly transform the product or add value from the customer's perspective. Instead, it is an activity that checks for defects or issues, which ideally should be prevented by improving the processes.

References:

Rother, M., & Shook, J. (1999). Learning to See: Value Stream Mapping to Add Value and Eliminate MUDA. Lean Enterprise Institute.

Liker, J.K. (2004). The Toyota Way: 14 Management Principles from the World's Greatest Manufacturer. McGraw-Hill.

The precision of the Coordinate Measuring Machine (CMM) is estimated by its standard deviation in repeated measurements, which is given as 2.497 mm. This value represents the precision of the measurements as it reflects the spread or variability in the measurement data.References:

Introduction to Statistical Quality Control by Douglas C. Montgomery.

ASQ Quality Press: The Certified Quality Engineer Handbook.

In Failure Mode and Effects Analysis (FMEA), the focus should be on implementing corrective actions on items that have high Risk Priority Numbers (RPN) or high severity values. The RPN is calculated by multiplying the severity, occurrence, and detection ratings, and it helps prioritize which failure modes need immediate attention.

When measurements are taken using a reference scale, such as a pair of calipers, the resulting data are referred to as continuous data. Continuous data can take any value within a given range and are measured on a scale that can be subdivided. This type of data is crucial for precise quality control and statistical analysis in manufacturing and engineering.

A material review board (MRB) gets its authority to act from executive management. The MRB is typically composed of representatives from various departments, including engineering, quality, and production, and is responsible for reviewing and making decisions on non-conforming materials. The authority given by executive management ensures that the MRB has the power to enforce quality standards and make necessary decisions.

Robustness in statistical conclusions often refers to the ability to draw valid conclusions even when certain assumptions are violated. One of the critical factors for robustness is high statistical power, which refers to the probability of correctly rejecting the null hypothesis when it is false.Reason: High statistical power increases the likelihood that the test will detect a true effect if it exists, thereby supporting the validity of the conclusions.

The process performance index (Ppk) measures the performance of a process relative to the specification limits, accounting for the mean and variability of the process. It is an indicator of how well the process can produce output within the specification limits.

References:

Montgomery, D.C. (2009). Introduction to Statistical Quality Control. Wiley.

ASQ. (n.d.). Process Performance (Pp, Ppk). Retrieved from https://asq.org/quality-resources/process-performance

P

Error-proofing, or "poka-yoke," is a technique used to prevent errors by designing processes, equipment, or fixtures to avoid mistakes.

Fixture design involves creating jigs, fixtures, or tooling that ensure parts are correctly positioned and assembled, thereby preventing errors.

Housekeeping, equipment calibration, and double sampling do not inherently prevent errors by design.

References: The concept of error-proofing is detailed in quality engineering literature, such as "Poka-Yoke: Improving Product Quality by Preventing Defects" by Shigeo Shingo.

Quality planning focuses on identifying and meeting the needs of the customers. It involves defining quality standards and determining how to achieve them to ensure customer satisfaction.

The correct test statistic for a paired-sample t test is used when comparing the means of two related groups. The formula for the paired-sample t test is:

t=dˉsd/nt = \frac{\bar{d}}{s_d / \sqrt{n}}t=sd​/n​dˉ​

Where:

dˉ\bar{d}dˉ is the mean of the differences between the paired observations.

sds_dsd​ is the standard deviation of the differences.

nnn is the number of pairs.

The chart provided is a control chart used to monitor process variation over time.

Control charts have upper control limits (UCL) and lower control limits (LCL) that represent the threshold of acceptable variation.

Points X and Y, which lie outside the control limits, indicate special causes of variation. These are deviations caused by specific, identifiable factors rather than inherent process noise.

Region Z, where data points are consistently below the mean and approaching the LCL, suggests a trend indicating a shift due to special causes.

References: According to ISO 9001 and the ASQ, control charts are utilized to detect special causes of variation, which signal that a process is out of control and requires investigation.

The effectiveness of corrective action can best be assessed by measuring the results of the corrective action. This involves evaluating whether the action taken has successfully addressed theroot cause of the problem and has led to a measurable improvement in the quality metrics or process performance.References:

ASQ Quality Press: The Certified Quality Engineer Handbook.

Total Quality Management by Dale H. Besterfield.

To determine which problem area should be targeted for improvement first, we need to consider both the dollar savings and the probability of success. Calculating the expected savings (dollar savings multiplied by the probability of success) for each problem area gives us a clear picture:

A: 100,000 * 0.15 = 15,000

B: 92,000 * 0.30 = 27,600

C: 84,000 * 0.20 = 16,800

D: 72,000 * 0.33 = 23,760

Problem Area B has the highest expected savings (27,600), making it the best target for initial improvement efforts based on the given data. References:

Quality Management for Organizational Excellence by David L. Goetsch and Stanley Davis.

ASQ Certified Manager of Quality/Organizational Excellence Handbook.

A scatter diagram is used to assess the relationship between two variables. In this case, it can be used to analyze the relationship between storage temperature and the overall length of the metalcomponent. A scatter diagram helps in visualizing any correlation between the two variables. Reference: "The Quality Toolbox" by Nancy R. Tague, which provides descriptions and uses for various quality tools including scatter diagrams.

Short-run Statistical Process Control (SPC) is particularly useful in situations where many different parts are made in small lots. This is because traditional SPC methods may not be effective when production runs are short and the volume of data for individual parts is limited. Short-run SPC techniques allow for effective monitoring and control of processes that produce a variety of parts in small quantities by standardizing and coding data to detect shifts and trends over multiple short runs.References:

Montgomery, D. C. (2009). Introduction to Statistical Quality Control.

ASQ Quality Press: The Certified Quality Engineer Handbook.

To determine the cumulative percent of nonconformities by the end of the third operation using the given data, we sum the percent of nonconformities for the first three operations:

1st operation: 55.0%

2nd operation: 23.3%

3rd operation: 11.7%

Cumulative percent of nonconformities by the end of the third operation = 55.0% + 23.3% + 11.7% = 90.0% Hence, the correct answer is 90.0%.References:

ASQ Certified Quality Engineer Handbook.

Quality Management for Organizational Excellence by David L. Goetsch and Stanley Davis.

Design Failure Mode and Effects Analysis (DFMEA) is a systematic method for evaluating a design to identify potential failure modes, their causes, and effects, and to prioritize the actions to mitigate these failures.

Identify Personnel Responsible for Implementation:

While important for action plans, it is not the primary focus of DFMEA.

Predict Failure Frequencies:

DFMEA does consider failure frequencies, but the primary goal is to address and mitigate failures.

Minimize Measurement Variation:

This is more relevant to process control and not the primary focus of DFMEA.

Mitigate Failure Modes:

The main objective of DFMEA is to identify potential failure modes and implement actions to reduce or eliminate the risks associated with these failures.

This involves analyzing the design to foresee where it might fail and taking preventive measures to improve reliability and safety.

References: Quality engineering and risk management texts, specifically those discussing the principles and objectives of DFMEA.

Concurrent engineering is an approach to product development where multiple phases of the design process are carried out simultaneously to reduce time-to-market and improve product quality.

Concurrent engineering is a systematic approach to integrated product development that emphasizes the parallelization of tasks (i.e., performing tasks concurrently), which contrasts with the traditional sequential engineering design method.

A Matrix Diagram is a tool used to display the relationship between different factors. It helps in identifying and analyzing the presence and strength of relationships among two or more sets of items. This tool is valuable in quality management as it can help identify the factors that are most influential in a process, thereby assisting in prioritizing areas for improvement.References:

ASQ Quality Press: The Quality Toolbox.

Total Quality Management by Dale H. Besterfield.

A laser scanner is a noncontact measurement tool that uses laser technology to measure the dimensions and shape of an object without physically touching it. This is useful for precise measurements where contact might damage the part or alter the measurement.

References:

Jain, R.K. (1994). Engineering Metrology. Khanna Publishers.

ASQ. (n.d.). Measurement System Analysis. Retrieved from https://asq.org/quality-resources/measurement-system-analysis