Six Sigma Tools (Control) Flashcards
Balanced Score Card
- A set of measures that gives top managers a fast but comprehensive view of the business
- The balanced scorecard allows managers to look at the biz from 4 important perspectives:
- Customer perspective: How do customers see us?
- Internal perspective: What must we excel at?
- Innovation & Learning perspective: Can we continue to improve and create value
- Financial perspective: How do we look to shareholders?
- The balanced scorecard also forces managers to focus on the handful of measures that are most critical
Dashboard
- A tool used for collecting and reporting information about vital customer requirements and/or your business’ performance for key customers.
- Provide a quick summary of the process and product performance.
Andon
- A Japanese word for lantern
- The board hangs over the aisle between production lines and alerts supervisors to any problem
- A typical tool to apply the Jidoka principle in Lean production
Deming on the variation in a process
- Variation is inevitable in industrial life
- Productivity and quality are linked, not traded off against one another
- Reduce 6Ms
- Common cause variation
- Special cause variation
Common cause variation
- Systemic issues, shared by numerous operators, machines, or products
Special cause variation
- Produce non-random variation in the system, usually confined to individual employees or activities
- Causes can be identified and eliminated.
Quality
- A process needs to be capable and in control to consistently produce to customer requirement
Capability
- The ability of a product, process, person or organisation to perform its specified purpose based on tested, qualified or historical performance, to achieve measurable results that satisfy established requirements or specification.
- Cp and Cpk are computing a single value to determine whether a process is capable.
Cp
- Compares the “natural tolerance” of the process (its natural variation) to the specifications: Cp = (engineering tolerance)/(natural tolerance) = (customer specification range)/ 6 sigma
- A Cp of 1 denotes a capable process at 3 sigma - but to allow for drift, a 1.33 is often used as the acceptable minimum (i.e. 4 sigma)
- The disadvantage of Cp is that it doesn’t account for process centering.
Cpk
- Compares the “natural tolerance” of the process (its natural variation) to the specifications for process centering: - Cpk = Z min - Zu = (upper specification - mean)/3 sigma - Zl = (mean -lower specification)/ 3 sigma
- We look at the difference between the mean and the upper and lower specifications. In a centered process, we’d expect these to be equal.
- At 3 sigma a Cpk > 1 is required (but +1.33 is preferred).
Capable Processes in Control
- When a process is capable, it can produce output that meets customer specifications.
- However, a process is only in control when it behaves as expected, that is it exhibits only random variation.
- When a process is capable and in control, the process is producing output that meets customer specifications, consistently.
Two criteria for capable and in control
- Is the process capable, that is does the output meet customer specifications? (calculate Cpk)
- Is the process in control, that is do the outputs meet customer specifications consistently? (variable: x bar and R charts; attribute: p-chart). (only when a process is in control can you know its true capability)
How do you identify whether a process is in or out of control?
Look at the control chart: if in control, the pattern should be random.
Attribute data (e.g. good or defective)
- Data that count items, such as the number of defective items in a sample
- Need just one chart, because mean determines standard deviation - p chart. - e.g. “good” or “defective”
Variable data (measurement)
- Data that measure a particular product characteristics such as length or width.
- Need two charts because mean and standard deviation are independent: x-bar and R chart
Control chart interpretation: points outside of limits
If there are sample means above UCL or below LCL, investigate for assignable causes.
Control chart interpretation: two-in-a row between 2&3 sigma
Two consecutive sample means between +2 and +3 sigma (or between -2 and -3 sigma), investigate for assignable causes
Control chart interpretation: two-out-of-three consecutive between 2&3 sigma
If two out of three consecutive sample means between +2 and +3 sigma, investigate for assignable causes
Control chart interpretation: 4-out-of-5 consecutive between 1&3 sigma
If 4 out of 5 consecutive sample means fall between +1 and +3 sigma (or -1 and -3), investigate for assignable cause.
Control chart interpretation: 5 in a row on one side of center line
If there are five sample means in a row above or below the center line, investigate for assignable cause.
Control chart interpretation: Trends
If there are 6 in a row steadily increasing or decreasing, investigate for assignable cause.
Control chart interpretation: 8 in a row between 2 and 3 sigma
Meaning none of the 8 within 1 sigma of the center line
Control chart interpretation: 14 in a row alternating
It is likely that one part of the tool is worn out unevenly, or 2 machines are used for the same part.
Control chart interpretation: 15 in a row between +/- 1 sigma
It is likely that the control limits need to be reset.
P-chart
- For attribute data
- p bar or the center line is the observed value of the average portion of defective/defects.
X bar (Mean chart)
- For variable data
- Aim to evaluate the stability of the sample mean.
- X bar is the average of mean
- A common practice to use 3SD -> 99.7% sample will fall between the UCL and LCL if the process is stable
- If the sample mean falls outside 3SD, it indicates special cause variations (e.g. machine malfunction, operator fatigue), otherwise the process is in control
R chart (Range chart)
- For variable data
- Center-line is the average of the range
- Aim to evaluate the stability of sample ranges
- The mean chart and the range chart must be viewed together. The former reflects central tendency, the latter measures dispersion within each sample. Neither chart is sufficient in its own to determine whether a process is in control.
