Chapter 9 Capability and Rolled Throughput Yield What

Chapter 9 Capability and Rolled Throughput Yield

What is capability analysis? And what can it tell us? Provides a measure of how well a process satisfies customer requirements. Will provide various ways to calculate defects per unit Allow us to calculate the rolled throughput yield Therefore uncovering the hidden factory (often consuming 30% of an org’s resources).

Measuring Capability • Measure the true quality levels of products and services. • Report the relationship between what is acceptable to the customer and what the process can naturally deliver. • Process Capability – is defined as the spread within which almost all of the values of a data distribution will fall. (Generally described as 6 or 3 ) – The main objective of 6 is to have no more than 3. 4 defects per million opportunities (DPMO). – The process must be brought into statistical control so that performance becomes predictable and capable of consistently meeting customer expectations.

What is a normal distribution? -6 -5 -4 -3 -2 www. barringer 1. com/jan 98 f 1. gif -1 0 1 2 3 4 5 6

So what is Capability? • The proportion of output that will be within product specification tolerances. Specification Range Usually set by customer LSL Lower Specification Limit USL Upper Specification Limit Process Range

How is capability measured? Cp>=1 Cp<1 Cp= 1. 33 process capable process not capable minimum value for process to be considered capable*. *Most organizations use 1. 33 as the minimum due to the shift that occurs over time We have a lower sigma for long-term than shortterm since long-term process capability is expected to shift over time by 1. 5. 6 -1. 5 = 4. 5 3. 4 DPMO answer to HW #1

Indices of Capability - Cp LSL Lower Specification Limit USL Upper Specification Limit C p= 1 Cp= 1. 5 Cp= 2. 0 Improvement Cp<1

Capability Example 9. 1 • CALCULATE the Cp – Process has a mean =. 738 – =. 0725 – LSL=. 500, USL=. 900 Is this process capable? Why?

Cp is missing a key element It only really makes sense when the process remains centered. • Cpk was created to address instances when the process mean is not centered relative to the specification limits. LSL Specification Range USL Process Range

Indices of Capability - Cpk LSL USL Potential Real Capability Cp= 2. 0 Cpk=2. 0 Cp= 2. 0 Cpk=<1 Cp= 2. 0 Cpk= 0 Cp= 2. 0 Cpk< -1

How can we calculate Cpk ? = process center = short-term standard deviation m = midpoint of specification MINITAB STAT QUALITY TOOLS Capability Analysis

What is capability analysis? And what can it tell us? Provides a measure of how well a process satisfies customer requirements. Will provide various ways to calculate defects per unit Allow us to calculate the rolled throughput yield Therefore uncovering the hidden factory (often consuming 30% of an org’s resources).

You can’t improve what you don’t measure? But what characteristics do Six Sigma metrics have? METRIC: 1. 2. 3. 4. 5. 6. Must have a scale Must have a goal, or standard Compensation or other forms of recognition must be attached to actual performance compared to goal. The metric is reported and reviewed vertically and horizontally throughout the organization on a regular basis. The metric must be able to be pooled horizontally and vertically within the enterprise. The metric must be strongly correlated with one or more of the 12 dimensions of quality at the business, operations, and/or process level of the organization.

12 Dimensions of Quality Metrics Aligned Owned Predictive Minimal Simple Correlated Contextual Reinforced Standardized Actionable Transformative Validated

What defects are we measuring? And what’s the difference? • • Unit = item, service, product, transaction delivered to customer. Defect = failure to meet customer req. or a performance standard Defective = unit that contains at least one defect Defect opportunity = opportunities for a unit to be defective • With DPMO, look up Long-term and Short-term Sigma in the Sigma Conversion Table (Table VIII in Appendix A)

DPU/ DPMO Example 9. 4 • Bank branch receives 300 mortgage applications per business day. • Report shows that these applications contained 50 defects. • Each application as 14 key fields (defect opportunities). Calculate the DPU, DPO and Sigma Level: Look up 11, 904. 76 in the Sigma Conversion Table (Table VIII in appendix) What is the range for the Short-term and Long-term Sigma?

DPU/ DPMO Example 9. 7 • Office receives 4500 engineering permit applications per day. • Audit shows that these applications contained 300 defects. • Each application as 12 key fields (defect opportunities). Calculate the DPU, DPO and Sigma Level: Look up 5, 555. 56 in the Sigma Conversion Table Short-term sigma = 4 - 4. 1 Long-term sigma = 2. 5 - 2. 6

Capability vs. Performance • For indicating the level of long-term performance that a process can achieve we use:

Equivalent Mean Offset LSL Short-Term Capability LSL Long-Term Performance USL 1. 5 USL

What is capability analysis? And what can it tell us? Provides a measure of how well a process satisfies customer requirements. Will provide various ways to calculate defects per unit Allow us to calculate the rolled throughput yield Therefore uncovering the hidden factory (often consuming 30% of an org’s resources).

What is meant by yield? • “Yield” can be viewed as the ratio of output to input: • A successful unit would mean it is free of any type on nonconformity (or defect). • EXAMPLE: Process 100 units, 90 are successful. • Called FTY (Final Test Yield)

But…yield is a flawed concept It doesn’t account for rework and replacement • We must look at a process map and measure IPY (In Process Yield) at each step. • RTY (Rolled Throughput Yield) is calculated by multiplying together they IPY. RTP = IPY 1 x IPY 2 x IPY 3 x…x IPYn • This could be considered a true assessment of process effectiveness.

Example of Yield vs. RTY START • Bank with 280 deposits a day. Each slip as 5 defect opportunities. They get a lot of complaints, but they can’t understand why since the # of defects per final confirmation is low. Customer gives deposit slip. Bank associate processes deposit Enters deposit in electronic system System generates deposit confirmation Customer obtains confirmation END FIRST: create a flow chart

Example of Yield vs. RTY START Customer gives deposit slip. Bank associate processes deposit Enters deposit in electronic system System generates deposit confirmation Customer obtains confirmation END Defs DPU IPY 5 5/280=. 0179 . 9823 6 6/280=. 0214 . 9788 3 3/280=. 0107 . 9893 4 4/280=. 0143 . 9858 2 2/280=. 0071 . 9929 . 0714 SECOND: Note the defects per step, and calculate the DPU, IPY for each. IPY = e-DPU Since there are multiple defect opportunities per unit.

Example of Yield vs. RTY START Customer gives deposit slip. Bank associate processes deposit Enters deposit in electronic system System generates deposit confirmation Customer obtains confirmation END Defs DPU IPY 50 50/280=. 1786 . 8364 6 6/280=. 0214 . 9788 3 3/280=. 0107 . 9893 4 4/280=. 0143 . 9858 2 2/280=. 0071 . 9929 . 2321 SECOND: Note the defects per step, and calculate the DPU, IPY for each. IPY = e-DPU Since there are multiple defect opportunities per unit.

What is capability analysis? And what can it tell us? Provides a measure of how well a process satisfies customer requirements. Will provide various ways to calculate defects per unit Allow us to calculate the rolled throughput yield Therefore uncovering the hidden factory (often consuming 30% of an org’s resources).

The Hidden Factory START • • The hidden factory is used to designate all work performed above and beyond to produce a good unit of output and is not explicitly identified. Can include the costs for rework, scrap, excess floor space and equipment. Estimated that 30% of an organization’s resources go to the hidden factory. Customer gives deposit slip. Rework Inspection Bank associate processes deposit Rework Inspection Enters deposit in electronic system Customer obtains confirmation Failure Analysis Rework Inspection END Failure Analysis Rework Inspection System generates deposit confirmation Failure Analysis