Flashcards 2
Lean Inventory Levels
Maintaining neither too much or too little
Extra effort that is put into a product or service that adds little or no value
Over Processing
Producing more than is needed
Over Production
Five Steps associated with 5S
Sort, Set In Order, Shine, Standardise, Sustain
In 5S the concept of Sustain
Assures new methods and procedures will be followed in the future and become part of standard operating procedures
The statistics used in maintaining a process
Mean & variation (standard deviation)
Process Variation
The natural phenomenon where successive units generated by a process are not all the same, the way around a process mean. Dependant on the process, variation can be large or small.
Multi Vari Chart
A simple way to measure process variation and compare several processes at once. The range is depicted by a vertical line and mean by a short horizontal line placed on the vertical line.
Vertical lines in a multi-vari chart
Represent the range of the process output values from lowest to highest
X-Y Chart
For prioritising the influence of process outputs on process outcomes. Input and output variables are identified. The importance on the outputs is estimated and the extent of the relationship between inputs and outputs are also estimated.
A weighted impact is then calculated
Control Chart
A chart that monitors process output by displaying the means of a sample. Each sample is the same size and samples are taken periodically.
Upper and Lower Control Chart limits
Horizontal lines in a control chart specifying whether the process is in or out of control.
Definition of a stable process
Where output varies at the target level for a significant period of time and where sample means do not exceed boundaries defined by upper and lower control limits.
Difference between the lowest and highest number in a dataset
Range
A statistical measure of variation that takes into consideration every member of a dataset is called ….
Standard deviation
Describe a normal distribution
Data clusters near the centre and tapers on the other side. One side is a mirror image of the other
What’s another name for a bell-shaped curve
Normal distribution
What is the area under a normal distribution curve
One hundred % of the observations
Mean of a normal distribution
Divides the observations in half. Fifty % fall below the mean and fifty above it. The normal distribution is symmetrical on both sides of the mean.
% of observations that fall between the mean and three standard deviations on either side of the mean?
99.7%
What % of the outcomes fall beyond plus and minus 6 standard deviations from the mean?
A negible amount. Near perfection
% of observations that fall between plus and minus two standard deviations from the mean of a normal distribution
95%
Should every process strive for Six Sigma results
No, it’s not cost effective. 4/5 Sigma is fine for most uses
Choosing the correct sigma standard requires the balancing of … and ….
Cost and quality. Higher the level of sigma the higher will be the quality and cost.
DPU
Defects per unit. Metric used to measure quality. Calculated as the number of defects in a sample divided by the number of units sampled.
RTY
Rolled Through Yield
Represents the likelyhood that a process will complete all of the required steps without failures or rejects
FTY
First time yield, the total number of parts of units finally divided by the total number that originally started the process
Why sample?
To gain insight into a population where time or money prevent exploring every datapoint in the population
Population and samples
Samples are subsets of populations. Many samples can be taken from the same population
Sample results
When samples are taken from a population every sample means will differ. In other words we cannot expect that two sample means taken from the same population will be the same
Population means and sample means
When many samples of the same size are taken from a population, the sample means will vary. A plot of these sample means is called the Distribution Of Sample Means
Distribution Of Sample Means
when many samples of the same size are taken from a population the samples themselves have a distribution. The mean of many many samples taken from this population will tend to coincide with the mean of the population.
Central Limit Theorem
When sample sizes are large the distribution of sample mean takes the shape of a normal distribution. This concept plays a major role in the field of statistics
Central Limit Theorem and the shape of the population
Regardless of the shape of the population from which many samples are taken, the central limit theorem holds that the shape of the distribution of sample means will be normal.
Sampling & Six Sigma
Sampling is at the very core of Six Sigma. Samples taken at regular intervals and are of a specific size, these are plotted on a control chart. They behave according to the distribution of sample means and are disbursed normally around the centreline or target of the control chart.
Control Chart Center Line
Represents the target or average process
What are the Upper and Lower control limits when samples larger than 25 are taken and limits are set @ the three sigma level
UCL = Target +3 * σ/squareroott (n)
standard deviation,
LCL = Target -3 * σ/squareroott (n)
Limits set at the three sigma level
When a process is in control, samples outside the three sigma limit will be rare. In fact 99.7% of these samples will fall within these limits. Process is out of control if out of these limits.
FMEA =
Failure Mode & Effects Analysis
Objective of FMEA
Identify problems before they happen and design a process in such a way that these problems are less likely to occur
RPN =
Risk priority number
Three terms in the risk priority number
Severity (of problems if it occurs)
Frequency - expected frequency of occurrence
Difficultly of detection before it can do damage
How is RPN calculated
RPN = Severity * Occurrence * Detection
Define Poka Yoke
Mistake proofing, making it difficult for workers, staff and customers to make a mistake
What is process capability
the ability of a process to meet it’s output objectives
Measuring process capability
Measuring the ability of a process in relationship to process expectation Process capability of 1 means that the process is just capable of meeting it’s expectations. Values greater than 1 mean it’s more than capable. Values less than 1 mean it’s less than capable.
Process stability
A stable process is one which the process mean does not change over time
Purpose of a control plan
Focuses on the management of the overall quality process and as such goes beyond the establishment of control charts. It is formal process and documented in writing
Seven Elements of a Control Plan
Measurement & Specification Input, Output Process Process Execution Performance reporting Documentation Corrective actions Process owner
Control Plan - Measurements and Specficiations
Defining measurable objectives for the process and ensuring that they meet customer expectations
Control Plan - Input Out Process
A part of the plan that defines the inputs (X) and outputs (Y) of the process. Often includes SIPOC diagram and flow charts.
Control plan - Process execution
Designing the process and controls to ensure successful delivery of product and service
Control plan, performance reporting
Establishing periodic reviews or audits of the control plan
Control plan, document info
The control plan must be well documented to facilitate effective management of the plan
Control Plan, corrective actions
Specifying what action will be taken should the sample results suggest that the process is no longer in control
Control plan - process owner
Identifying the responsible party in the event changes need to be made or the process found to be out of control.
