Six Sigma | Chapter 6 - DMAIC, Measure Flashcards
(96 cards)
In DMAIC, for Measure, what are the Inputs of the methodology?
- Project Charter
- Established metrics
- Problem Statement
- Roles & Responsibilities
In DMAIC, for Measure, what are the Tools of the methodology?
- Data Collection Tools & Techniques
- Measurement Scales
- Validation Techniques
- Statistical Distributions
- Data Mining
- Run Charts
- Detailed process maps and process charts
- Stakeholder Tools
- Process Costs
In DMAIC, for Measure, what are the Outputs of the methodology?
- Well-Defined Processes
- Baseline Process Capabilities
- Process parameters affecting CTQ
- Cost of Poor Quality (COPQ)
- Measurement Systems
In DMAIC, for Measure, what are the objectives of the measure phase?
In Measure phase, our objectives are:
- Data collection to understand present system better
- Validation and reliability of measurement system and key metrics
- Determining the process capability for present system
- Determining how progress and project success would be measured
In DMAIC, for Measure, between continuous and discrete data which one is considered variables and which one is considered attributes?
- Continuous Data is Variables
- Discrete data is Attributes
In DMAIC, for Measure, what is the definition of continuous data? What are some examples?
This is information that can be measured on a continuum or scale e.g.
- Weight of packages sent
- Customer wait time for every customer service call
- Average speed of cars traveling in a highway
In DMAIC, for Measure, how can you identify continuous data?
To identify continuous data, ask the question whether the data can be expressed to any desired level of precision or decimal point e.g. What is the distance covered by the car? This can be answered as 1 mile, 1.1 miles, 1.1.1 miles etc.
Any data that can be measured to any desired level of precision is called continuous data.
In DMAIC, for Measure, what are tools used with continuous data?
Data collection:
1. Demographic questions (e.g. age etc.)
2. Fill-in-the-blank questions (e.g. distance covered, temperature, humidity etc.)
Data analysis tools :
1. ANOVA etc.
In DMAIC, for Measure, what is the definition of discrete data? What are some examples?
This is a whole number (or count) of attributes like :
- Number of people buying a product
- Number of defects per 1000 events
- Number of satisfied customers
In DMAIC, for Measure, how can you identify discrete data?
To identify discrete data, ask the question whether the data can be expressed by 1 level of precision e.g. Is your age equal to or greater than 35 years? This can be answered as Yes or No. There is no further level of precision possible. So, the data type is discrete
In DMAIC, for Measure, what are tools used with discrete data?
Data collection:
1. Ranking
2. Rating
3. Yes/No questions etc. - covered in chapter 2: Stakeholders, Customers and financial measures
Data Analysis tools:
1. Chi-square
2. regression etc.
In DMAIC, for Measure, why is converting data required?
Data conversion helps in using appropriate tools to analyze data for example, if discrete data is plotted in control charts intended for continuous data, the results will be erroneous; similarly tools like chi-square can only be used for discrete data
In DMAIC, for Measure, how do you convert data from discrete to continuous data?
Discrete data are often caused by rounding off the data to too few levels or precision or decimal points.
To convert discrete data to continuous data, try to find out if the data can be expressed in greater level of precision - this can usually be achieved by asking for greater accuracy in data collection.
In DMAIC, for Measure, how do you convert data from continuous to discrete data?
Continuous data can easily be converted to discrete data by changing the data collection scale.
Consider a case study where we are trying to determine whether average heights of people in different countries are different. We conduct a survey of 100 people in 50 countries, where we ask every person their heights (the height measurements using this approach is a continuous scale).
To convert the data to discrete type, we can categorize each person’s height as follows:
Less than 4 Feet : Very Short
4 feet to 5 feet : Short
5 feet to 5.5 feet: Average
5.5 feet to 6.2 feet: Tall
Greater than 6.2 feet: Very tall
Please note that using a different measurement scale has helped us convert the continuous data type to discrete data type
In DMAIC, for Measure, what do we use to collect data?
Check Sheet(s)
In DMAIC, for Measure, why are check sheets important?
Check sheets are very important tools for data collection. Inputs gathered from check sheets can be used for creation of pareto diagrams, cause and effect diagrams etc.
In DMAIC, for Measure, what are the steps to create a check sheet?
Creating check sheet - steps involved:
- Determine the measurement objectives. Ask questions such as “What is the problem?”, “Why should data be collected?”, “Who will use the information being collected?”, “Who will collect the data?”
- Create a form for collecting data. Determine the specific things that will be measured and write this down the left side of the check sheet.
- Collect the frequency of data for the items being measured. Record each occurrence directly on the Check Sheet as it happens.
- Tally the data by totaling the number of occurrences for each category being measured
In DMAIC, for Measure, what are two other data collection techniques?
- Coding Data
- Gauging
In DMAIC, for Measure, what is coding data?
Data coding is used to get variable data required for control charts (control charts will be covered in chapter 8)
In DMAIC, for Measure, why is coding data useful?
Coding data enables the user to plot several parts from a given process into the same control chart.
In DMAIC, for Measure, how is coding data standardized?
The data is standardized by subtracting nominal or other target values from actual measurements.
Coding data is often standardized so that measurement units are converted to whole numbers (e.g. 0.022 miles will be recorded as 22)
In DMAIC, for Measure, what are measurement scales used for?
From a six sigma perspective, measurement scales help in categorizing data into different types so that they can be collected and analyzed separately.
In DMAIC, for Measure, what are the major measurement scales?
Major measurement scales include:
- Nominal
- Ordinal
- Interval
- Ratio
NOIR
In DMAIC, for Measure, what is a nominal measurement scale and an example?
Nominal
Here, items are assigned to groups or categories. There is no ordering of data (i.e data collected does not show that something is better than the other).
Nominal scales are therefore qualitative rather than quantitative.
Variables measured on a nominal scale are often referred to as categorical or qualitative variables
Examples: country or origin, sex (Male/Female), and religion