ASQ CSSBB - Certified Six Sigma Black Belt

The process capability index Ppk measures how well a process is performing relative to its specification limits while accounting for how centered the process mean is. It is calculated using the formulas:

First determine the specification limits from the tolerance 5.2 ± 0.3 :

USL = 5.2 + 0.3 = 5.5

LSL = 5.2 − 0.3 = 4.9

Next compute both capability sides.

Upper side:

Lower side:

Ppk takes the minimum of these two values because capability is limited by the nearest specification limit.

Thus, the correct answer is A (0.625) . The process mean is closer to the upper specification limit, which reduces capability on that side and determines the overall Ppk value.

The correct answer is D, Factor B low, because the main-effects plot indicates that the response is lowest at the low setting of factor B. In a main-effects plot, the average response is plotted at each level of each factor, and the team compares the height of those plotted means to determine which settings increase or decrease the outcome. The factor level associated with the lowest plotted point gives the lowest average response. In Six Sigma Analyze and Improve work, this is a fundamental DOE interpretation skill. Black Belts use main-effects plots to identify which factor settings move the response in the desired direction and to determine promising operating conditions for further study or optimization. The steeper the line, the stronger the main effect of that factor on the response. This question focuses only on identifying the lowest response setting, and from the plot that point occurs at factor B low. Therefore, the correct answer is D, because the low level of factor B corresponds to the minimum response among the settings shown.

===========

The correct answer is C. identify different types of variation . According to the CSSBB material, Statistical Process Control (SPC) applies statistical analysis to measure, monitor, and control processes, and its major tool is the control chart . The reference explains that all processes are subject to variation and that this variation is classified into two types: chance cause variation and assignable cause variation . One of the primary purposes of control charts is to distinguish between these sources of variation so that assignable causes can be investigated and eliminated.

The CSSBB text also describes control charts as powerful tools for analyzing variation and determining whether a process is stable and predictable. While control charts may help indicate when corrective action is needed and may support investigation into sources of failure, their primary use is not to directly determine specific failure sources, design a sampling strategy, or make lot acceptance decisions. Acceptability of sampled lots is more closely related to acceptance sampling , not SPC charting. Therefore, the best answer is the one that aligns with the central purpose of SPC in Six Sigma: recognizing and separating normal process variation from special-cause variation.

The correct answer is A. attribute data . In the CSSBB source, attribute charts are used for characteristics such as pass/fail, good/bad, number of defects, and number of defectives, and the text explicitly states that attributes are discrete, counted data . This fits the question exactly because error counts in the final product are count-based and therefore discrete rather than continuous. The CSSBB material also identifies attribute data as one of the primary types of data used in Six Sigma, alongside variables data and locational data.

A data collection plan is a method for organizing how data will be gathered, not a type of data. Stratified data refers to data separated into meaningful subgroups, such as by shift, machine, or supplier, but that does not define whether the data are discrete or continuous. Continuous data, by contrast, involve measurements on a scale such as weight, time, or length. Since the question specifically describes discrete counts of errors, the CSSBB-consistent term is attribute data . This is why the best and verified answer is A .

In any data set measured around its arithmetic mean, the algebraic sum of the deviations from the mean equals zero. This is a foundational statistical property used throughout Six Sigma and is especially important in the Measure Phase, where descriptive statistics are used to summarize process performance. Each value above the mean contributes a positive deviation, while each value below the mean contributes a negative deviation. Because the mean is the balancing point of the distribution, these signed deviations offset one another and sum to zero. This is true for a normal distribution and, more generally, for any data set when deviations are taken from the arithmetic mean. The importance of this property is that it explains why variation cannot be measured by simply summing deviations; instead, Black Belts use squared deviations, absolute deviations, variance, or standard deviation to quantify spread. That is why variance and standard deviation are measures of dispersion, while the direct sum of deviations is zero. Therefore, the correct answer is zero, which reflects the balancing property of the mean in statistical measurement and process-data analysis.

===========

The correct answer is B. Selecting the intended response variable . In the CSSBB DOE planning material, one of the earliest planning steps is to set objectives and select process variables , including both inputs and outputs. The source specifically defines the response variable as the variable that shows the observed result of an experimental treatment and notes that good DOE practice includes selecting all relevant responses during planning.

This is clearly part of pre-experimental planning , because before running any trials, the team must decide what outcome will be measured and optimized. Without a clear response variable, the experiment cannot be structured properly, analyzed correctly, or linked to project objectives. Determining critical values is usually associated with later statistical decision-making, not the basic front-end design of the experiment. Increasing the number of experiments while reducing sample size is not a standard first-step planning rule, and ensuring that only one person conducts each experiment is not a general DOE principle. CSSBB guidance instead emphasizes defining objectives, understanding the process, choosing factors and responses, confirming the measurement system, and then selecting the design. Therefore, selecting the intended response variable is the best answer.

The design described is a Plackett-Burman design. This screening design is a two-level fractional factorial approach used to study many variables efficiently in a relatively small number of runs. A defining feature is that it can study k = N−1 factors in N runs, where N is a multiple of 4. This makes it especially useful in Six Sigma Analyze and Improve work when the team needs to screen a large number of potential factors quickly to identify the few that are likely to matter most. It is not a full factorial design because full factorial experiments require many more runs as the number of factors increases. It is not a Latin square design, which is used to control for two nuisance variables. Resolution IV is a classification of certain fractional factorial designs, but the formula given in the question specifically identifies the Plackett-Burman structure. For Black Belts, this design is valuable when resources are limited and the first objective is factor screening rather than detailed interaction modeling. Therefore, the correct answer is B, Plackett-Burman design.

The correct answer is A. A reduction in work-in-process . In the CSSBB materials, TOC uses three primary system measures: throughput , inventory , and operational expenses . Inventory is defined as all the money the system has invested in purchasing things it intends to sell, or in practical terms, money that is tied up inside the system rather than converted into sales. Reducing work-in-process (WIP) directly lowers the amount of material and partially completed work sitting in the process, which reduces the amount of money trapped in inventory or investment.

This is fully aligned with TOC thinking because bottlenecks and poor flow tend to increase queue sizes and WIP. The CSSBB source stresses balancing flow with market demand rather than maximizing local utilization, which helps prevent excess buildup of inventory. While reducing cycle time can be beneficial, it is not as directly tied to decreasing inventory as reducing WIP. Increasing repeat business or total sales relates more to throughput generation than to reducing internal investment. Therefore, the TOC element most clearly designed to decrease inventory or investment is a reduction in work-in-process .

The correct answer is C. 2.50 . The CSSBB financial-measures material gives the widely used formula:

ROI = Net Income / Investment

To apply that formula correctly, the net income or project gain is the savings minus the investment :

Net income = 700,000 − 200,000 = 500,000

Then divide by the investment:

ROI = 500,000 / 200,000 = 2.50

So the project returns 2.5 times the invested amount , which matches option C . This is consistent with Six Sigma project financial evaluation, where costs and benefits are compared to determine whether the project is worthwhile and how strongly it contributes to profitability. The CSSBB source also emphasizes that financial benefits are used to evaluate, prioritize, select, and track Six Sigma projects.

Option D would result only if someone incorrectly divided total savings by investment without subtracting the investment itself. Under the CSSBB ROI definition using net income, the verified answer is 2.50 .

The correct answer is B. xₘ − x . The CSSBB source defines bias as the difference between the output of the measurement system and the true value . It is often referred to as accuracy . The formal definition in the instructor text is the average measured value minus the value of the standard or true value.

Using the notation in this question, where xₘ is the measured value and x is the true value, the bias is simply:

Bias = xₘ − x

This gives both the size and the direction of the bias. If the result is positive, the gage overestimates. If it is negative, the gage underestimates. That is important in Six Sigma measurement system analysis because the team needs to know not only how far off the measurement system is, but also whether it consistently reads high or low. The ratio in option A is not the definition of bias, option C is a relative or percent-type expression, and option D removes the direction of the error. Therefore, the verified CSSBB answer is B .