Test 2 OMIS 430

Why is Product Design necessary?

- New products are constantly introduced—these must to be designed
- Old products are also reintroduced—these must to be redesigned

Operations Manager must ensure that the products produced (or services offered) are:

- Functional
- Cost effective (to be competitive)
- Reliable

Product design plays a big role in functionality, reliability, and cost effectiveness in production. Products must be designed for "Manufacturability" (i.e. ease of fabrication and assembly). This will ensure cost-effectiveness. Design motto should be "KISS" (i.e. Keep It Simple and Save).

Importance of Product Design

- Affects quality
- Affects costs
- Affects customer satisfaction
- Affects functionality of product
- Affects reliability of product

Reasons for Product Design (Economic, Competitive, Legal, Technological, etc.)

- To increase business (Design and offer new products for more business)

- To be competitive (Design and offer new products to stay in the game)

- Availability of new technology, materials, etc. (This was not possible before)

Reasons for Product Redesign

- Customer complaints/dissatisfaction
- Excessive accidents/injuries, excessive warranty/liability costs
- Low demand
- Availability of new raw materials or technology or processes
- To comply with new regulations, government changes, safety issues, etc.
- To improve quality
- Recent trends of:
1.)Increased emphasis on customer satisfaction
2.)Increased emphasis on improving productivity
3.)Increased emphasis on reducing manufacturing time
4.)Increased emphasis on environmental concerns (EPA)

Design Objectives

1.) Achieve customer satisfaction (functionality, cost-effectiveness, and reliability)
(Understand what the customer wants and design accordingly!)

2.) Design for manufacturability (make it easy to produce)

3.) Design for operations (make it easy to produce for your operation)

4.)Redesign to reduce costs, recycle, increase value, remanufacture (reuse)

Legal, Ethical and Environmental Issues

1.) Product liability (responsibility for injuries or damages)

2.) Uniform commercial code (product must be usable for its intended purpose)

3.)Products must be environmentally safe

No product design is usually possible without Research.

Many big corporations have Research and Development departments.

2 types of research

1.)Basic- Advancing the knowledge
2.)Applied- Some goal of commercial application is in mind

**Big Corp. have R & D**
-Converts basic research into an application

Product life cycle assessment

1.) All products go through a life cycle: introduction (incubation), growth, maturity, saturation and decline.

It is important that most critical design features are perfected in the early part of the life cycle (incubation or introduction) to minimize recall problems.

Demand,Incubation,Growth,Maturity, Saturation, Decline

"(Degree of) Standardization" (Absence or degree of variety)

- 0%
- High Variety
- Ex: Dentist Visit

- 100%
- Low Variety
- Ex: Automatic Car Wash

Standardization

helps make design simpler and cheaper, allowing us to be cost effective, but can be less appealing to the customers because of reduced variety.

Some companies go for "Mass Customization," that is "Producing basically standardized goods with some degree of customization."

Other Issues in Product Design/Service Design

1.) Product Life Cycle Assessment
2.) Degree of Standardization
3.) Reliability
4.) Modular Design
5.) Robust Design

Reliability

the ability of the product or system to perform its intended function under a prescribed set of conditions

a.) Ability of the product to perform its intended function (consistently)
b.) Important aspect of product design
c.) Important aspect of competitiveness (selling point)

Ways of improving reliability

1.) Better component design: Use better components in your products

2.) Better production/assembly techniques

3.) Better testing of products or services

4.) Using (redundancy) back-ups

5.) Better user education

6.)Better system design

Redundancy in Design

- use of back-ups

- Ex: Spare tire in your car

Modular Design

A form of standardization in which component parts are grouped into modules that are easily replaced or interchanged

Modular design offers easier diagnosis, repair, replacement, etc. but can be costly.

Robust Design

design that results in products or services that can function over a broad range of conditions

(Phases and Development of Design) Design Process:

1.) Begins with (motivation) an idea*

2.) Product Specifications (Customer is the driving force)
(Marketing provides ideas/specifications from customers)

3.) Prototype development and Market Test

4.)Product introduction and Follow-up

Ideas can come from (3 ways)

1.) can come from customers, suppliers, etc. (Supply Chain based)

2.) can come from competition (reverse engineering), (Competition-based)
(Reverse Engineering: Dismantle and inspect competition's product for ideas)

3.) can come from R & D, that is, from Research (Research based)

Approaches to Design

1. Design for manufacturability
2. Design for assembly (easy vs. difficult to assemble)
3. Design for recycling
4. Design for remanufacturing (car parts, battery, alternator)
5. Robust design (most important today)

Improving the Design Process (For rapid New Product Development)

1.) Concurrent engineering: Bringing together design, marketing, manufacturing (operations) and testing people — allowing everyone to work closely early in the design phase. This helps speedier new product development. (Get new product out before the competition does).

2.) Computer Aided Design (CAD): Creating product designs using computer graphics resulting in improved efficiency of the designers and lower cost of the designs

Quality Function Deployment (QFD)

1.) This Integrates "voice of the customer" in product design process

2.)It Relates customer requirements (what) to their corresponding technical features or requirements (how)

3.)It is in the form of a matrix (also called House of Quality)

Service

Something that is done to or for the customer

Service Design

1.) Most services involve interaction with the customer at design or delivery stage. This makes service design delicate and difficult.

2.) It is generally preferred to minimize customer contact

3.) Service requirements vary from customer to customer.

4.) Also, many services are difficult to describe.

5.) Thus, many services must be highly customized, making design difficult.

6.) Another problem is that services cannot be inventoried.

7.) Response time can be very critical for service operations.

Product Bundle

A combination of goods and services provided to a customer

Many services are not pure services

Product reliability brings about perception of _______.

Quality

_________ should be an important focus in product design.

Reliability

Operations managers as well as designers are constantly seeking ways to improve ___________ as well as _____________ of their products.

Reliability, Functionality

Reliability (alternate definition)

is the probability that a component or a product (i.e. several components working together) or a system will function

Either:
1.) When activated or needed (in a single trial), (Ex: Will the car start now?)

2.) For a given length of time or service. (Ex: Will car last 5 years or longer?)

Quantifying/Computing Reliability: Reliability is computed as a _________?

Probability

2 ways to compute reliability as a probability

1.) probability that product will function when activated

2.) probability that it functions for a given length of time

Way I of computing reliability - Probability that product will function in one single trial

The method of computation depends on how individual components are arranged functionally in the assembly or system.

1.) Series type arrangement: uses Rule 1

2.) Parallel type arrangement: uses Rule 2

3.) Hybrid type: Arrangement that is neither purely Series type nor purely Parallel type
(A combo between the 2; Rule 3)

1. Series Type Arrangement

All components must function for product or system to function

Ex. 1.) Battery and starter in your car. Assume their reliabilities to be 90% and 80%.
The probabilities that these components will function (individually) are 0.8 and 0.9.

0.8 x 0.9 = .72

Ex. 2.) Now assume also that the reliability of the ignition switch is 90% (0.9).

Then, Probability that the car starts is

0.9 x 0.9 x 0.8 = 0.648 or 0.65

***As the number of components increases, the system reliability decreases***

Parallel Type Arrangement

All components do not have to function simultaneously for product or system to function b/c system uses back-ups or redundancy. Notice that the back-up is redundant until needed ex.) main unit fails

Ex.)Airlines have a system of two computers at the Gate to get passengers checked in. If one doesn't work, agent simply switches to the other with no loss of continuity.

Reliability of main computer = 90% (Computer 1 = .90)
Reliability of back-up computer = 80% (Computer 2 = .80)

Find system reliability — i.e. probability that all passengers are checked in.

System Reliability =.9+ (1 -.9) x .8
=.9 + (.10) x .8
=.9 + .08 = .98 (system reliability = 98%)

As the number of components in system increases, the system reliability will increase.

Hybrid Type Arrangement

Combination of Series type & Parallel type. Usually, you have many components in series arrangement, some of which have back-ups.
To compute system reliability here:
(1) Divide the system into sections so that each section has single component or components purely in series or purely in parallel. (2) Compute the individual section reliabilities.
(3) Using the section reliabilities, compute the system reliability. Sections will be purely in series or purely in parallel arrangement.

Sec .1 .98
Sec .2 .90 + (1-.90) x .90
.90 + (.10) x .90
.90 + .09 = .99
Sec .3 .95 + (1-.95) x .92
.95 + .05 x .92
.95 + .046 = .996
Thus, System Reliability is = .98 x .99 x .996 = .966

Way II of Computing reliability- Reliability is computed as a probability that the product will function for a given length of time

- Assumption is that the Product Life T (i.e. Time elapsed before failure) is a random variable.
- "T" may be an exponentially distributed or a normally distributed random variable.
- Probabilities that life "T" will exceed a specified time can be obtained from tables or by using the appropriate formulas.
- This probability will be the reliability.
- Probabilities that life "T" will not exceed a specified time (product will fail before time "T") can also be obtained from tables or formulas.
- Essentially, these probabilities are the areas under the exponential or normal curves.

Way II of Computing Reliability- CASE A

Product Life T is an exponential random variable with mean =MTBF
(MTBF = Mean Time before Failure or before failures)

Here, Reliability = Pr (no failure before T) will be equal to = e-T/MTBF

e = natural logarithm base, 2.7183
T = length of service before failure
MTBF = mean time between failures

And, Probability of Failure before time T is P (Failure before T) = 1- e^-T/MTBF

Way II of Computing Reliability- CASE B

Product Life T is a normally distributed random variable with mean = μ and Standard deviation = σ
Here, Reliability (i.e. product will not fail before time T) as well as probability that the product will fail before time T are obtained using Standard Normal Tables (Z tables).

Thus, Reliability = Pr (No failure before time T) = area to the right of T on the normal curve

And, Probability of failure before time T is: Pr (Failure before T) = area to the left of T on the normal curve

Convert T to Z and get the areas from the standard normal tables. (See your Class notes for this).
Remember Z = (T-μ)/σ

Such reliability (probability) calculations are useful in determining ____________________________.

Product Warranties

Failure Rate Profile for Products

The failure rate is high earlier because the defective products fail right away. The rate levels off random failure for a while and again climbs up when failures occur due to wear out. The curve is known as a "bath tub" curve.

Quality Revolution

began in Japan - Lead to today's TQM.

In the U.S., prior to 1970 --- Focus was on cutting costs and improving productivity, rather than on quality

Definition of Quality

Ability of a product or service to consistently meet or exceed implicit and explicit customer expectations

Actually, Quality is "What the customer says it is!"

Even today --- there is a significant gap between some US and Japanese products in terms of Quality, but the gap is closing fast.

Japanese Manufacturers: use defectives per million (units) \ Used to judge quality

United States Manufacturers: use defectives per hundred (units) to assess QUALITY

(I) Quality Gurus and Foundations of Modern Quality Management

Before quality revolution:

-Quality was responsibility of the QC Dept.
-Also, in mass production systems emphasis was on finding defectives at the end of the line (after assembly was complete). This was done to avoid high cost of system interruption or disruption. "This is how "Lemons" were built, because many defects were hidden from view, embedded in the product."
-Although, some defects were detected in the production process, efforts did not extend to design of products/processes to remedy the problems or to suppliers.
-Workers had little input if any at all.

Modern Quality Management

places emphasis on preventing mistakes rather than on finding and correcting them. Quality is responsibility of everyone in the organization (not just QC Dept). Inputs are sought from employees. Suppliers are treated as partners or part of the team.

Following "Quality Gurus" have contributed to the development of the modern quality management thru their teachings, views, philosophies, etc.

1.) W. Shewhart: - Father of statistical quality control. Control charts for quality

2.) W.E. Deming - - 14-point program
- Special (correctable) causes of variation
- Common (random) causes of variation
- Reduce variation to improve quality

3.) J.M. Juran: - defined quality as Fitness for use. Gave us
- Quality Trilogy: - Quality planning
- Quality control
- Quality improvement

4.) Phil Crosby - Proponent of "Quality is free" and "Zero defects."
- Our goal should be zero defects.
- Do it right the first time.

5.) K. Ishikawa - Cause and effect diagram (Problem solving tool)
- Quality circles (Getting workers involved in quality improvement)

6.) G. Taguchi: Gave us the important concepts of
- Robust design (Designing products so they function regardless of who uses them, how and in what environment)
- Quality (Taguchi) loss function
- The function tells us that quality is drastically lowered as deviations (that is variability) increase

7.) Ohno and Shingo: Gave us "Kaizen" - Concept of continuous improvement
Kaizen was very successfully implemented at Toyota

Today, emphasis has shifted from simply enforcing quality control (QC) to a ________ __________ to Quality.

Strategic Approach

(II) Basics of Quality

Insights on Quality management: Quality is not a tacked on special feature but is an integral part of product or service. To achieve quality, one must understand the basics of quality consisting of the following:

1.) Dimensions of Quality
2.) Determinants of Quality
3.) Costs of Quality
4.) Consequences of Quality

Dimensions of Quality

Quality does not pertain to a single aspect of a product;
i.e. different aspects may be important to different people, such as:
a) Performance: Main characteristics of the products/service
b) Special Features: Extra characteristics
c) Reliability: Consistency of performance
d) Durability: Useful life of product/service
e) Service after Sale (Serviceability): Handling of complaints, checking on customer satisfaction
f) Conformance: Overall how well a product/ service corresponds to customer expectations
g) Aesthetics: Appearance, feel, smell, taste
h) Perception: Perceived quality i.e. reputation of the organization

Determinants of Quality

A. Quality of Design: Refers to inclusions or exclusions of certain features in a product or service. Design decisions must consider customer wants, production/service capabilities, safety, liability, costs. (Use surveys, focus Groups to find customer wants)
B. Quality of Conformance: The product may be well designed but is it well made? This is affected by capabilities of equipment used, motivation and attitude of employees, their skill, training, etc.
C. Ease of Use: Ease of Use and good "instructions of use" will ensure that product will be used as intended and for the intended purpose; in turn ensuring customer satisfaction and reduced liability.
D. Service after Delivery: Product may not perform as intended. Then, company must provide:
REPAIR/RECALL
ADJUSTMENT
REPLACEMENT BUY BACK
RE-EVALUATION OF SERVICE
"We stand behind our product or service," may convince some they are getting quality.

Consequences of Poor Quality

A.) Loss of Business---switch to competitor
B.) Liability---improper design or improper assembly can cause accidents, harm--LAWSUITS
C.) Loss of productivity---rework costs, wasted material, labor, etc.
D.) Other consequences ---- costs associated with scrap, rework, replacement and repair, warranty, discount for inferior quality. Also, inspection & transportation costs (Costs to remedy a quality problem).

Costs of Quality

A) Failure Costs:

Internal failures are those detected during the production process while external failures are those detected after delivery to the customer.

External failures result in warranty work costs, handling of complaints, replacements, liability
and loss of good will.

B) Appraisal costs - Cost of inspection, testing etc to ensure quality and uncover defects (labor + equipment) i.e. Q.C. costs

C) Prevention Costs - Costs of preventing defects from recurring (QA). This includes planning and administrative costs, working with vendors, all training, QC procedures, customer assessment, etc.

Ethics and Quality Management (Role of Ethics)

Excellence is = Quality + Ethics.

(III) Major Quality Awards

Baldridge Award: Given by U.S. govt. to recognize and promote quality achievements in U.S. businesses

European Quality Award: Prestigious award for organizational excellence

Deming Prize: Given by Japanese govt., for anyone, focuses heavily on Statistical Quality Control.

This is given to firms that distinguish themselves with excellent Quality Management Programs.

(IV) Quality Standards and Quality certifications

1.) ISO-9000: (9000-9004) European
2.) ISO-14000 European
3.) MIL-SPEC U.S. Military Specifications Q90-Q94
4.)Z-8101 Japanese

ISO-9000: (9000-9004). These are European standards.

- These are a series of international standards that outline the requirements for "Quality Management Systems" and Quality Assurance worldwide. Compliance will help improve efficiency, improve productivity, and reduce costs.
- Requires companies to document everything they do that affects the quality of goods and services
Critical to international business

ISO-14000 (European)

These are a set of international standards for assessing a company's environmental performance

MIL-SPEC (U.S.) Military Specifications Q90-Q94

Very first quality standards developed
Used as basis for development of ISO standards

Z-8101 (Japanese)

Many businesses today seek ISO certification to improve their ability to compete globally.

Quality and Supply Chain:

Businesses understand that achieving quality in "Operations" alone is not enough. They are recognizing the importance of the entire "Supply Chain" in achieving quality starting from suppliers to logistics to distribution. It also involves customer perceptions, identifying problems, correcting these problems, getting suppliers involved and monitoring vendor quality efforts.

(V) TOTAL QUALITY MANAGEMENT (TQM)

Quality is a strategic decision that allows you to compete globally. But quality is not easy to achieve. Organizations need a formal quality management system (TQM) to achieve quality. But, what is it?

TQM- is an organization wide effort to achieve quality
TQM- is philosophy involving Quest for quality, involves everyone, and customer satisfaction in the driving force

The three basic fundamentals of TQM are:

1.) Customer Focus
2.) Process Improvement
3.) Total Involvement

TQM Approach

1.) Find what the customer wants (surveys, focus groups, interviews, etc.).
2.) Design product/service to exceed customer expectations, make it easy to use, and make it easy to produce
3.) Design production to do it right the first time. Strive to mistake-proof the process.
4.) Improve the process. Never stop trying to improve.
5.) Extend these concepts to suppliers and distribution. Make suppliers partners.

Other important elements of TQM include: TQM also involves the following.

1.) Continuous Improvement: (Kaizen in Japanese)
2.) Competitive Benchmarking: Identify the best and model your efforts after them.
3.) Employee Empowerment: Give responsibility for improvements and authority to make changes along with resources to the workers
4.) Team Approach: Use teams to solve quality problems.
5.) Decisions: Base decisions on facts rather than opinions.
6.) Knowledge of Tools: Train employees to use quality tools.
7.) Supplier quality: Include suppliers as partners in quality assurance and improvement efforts.
8.) Champion: Become a TQM champion: promote value & importance of TQM throughout company.
9.) Control quality at source: Make each worker responsible for the quality of his/her work at his/her workstation

This will transform a traditional company into a TQM company.

Six Sigma

Six Sigma has become a vital component of TQM today. Six Sigma is a business process for improving quality, reducing costs and increasing customer satisfaction. Statistically, it means no more than 3.4 defects per million, in any process, product or service. It is a program designed to reduce occurrence of defects in design, production, service and delivery, to lower costs, save time and improve customer satisfaction. It involves providing strong leadership, improving process, reducing variation, using statistical methods and designing a structured improvement strategy.

However, note that problems will always arise.

Thus, six Sigma can help but there are many obstacles to implementing TQM

As a result, TQM is not without criticism .

However, sustained continued "Problem Solving" can help TQM succeed.

Problem Solving

Thus, problem solving is one of the very basic procedures in TQM. For continuous improvement to occur, the quality problems that occur regularly must be solved. The PDSA (Plan-Do-Study-Act) cycle (also called Shewhart cycle or Deming wheel) can be of big help in any "Process Improvement" A number of other quality tools can also be of great help in problem solving and process improvement. These tools aid in data collection and interpretation (i.e. problem solving), and thus help in decision-making.

Everyone in the organization must be trained to use ______ tools so as to speed up transition of a company to a ______ company.

TQM

Commonly employed eight TQM Tools

1.) Check sheets: For recording/organizing (tallying) data to identify a problem
2.) Flowcharts: A visual representation of the steps in the process
3.) Scatter diagrams: A graph that shows the degree and direction of relationship between two variables
4.) Histograms: A chart of an empirical frequency distribution
5.) Pareto analysis: Technique for classifying problem areas according to degree of importance and focusing on the most important
6.) Control charts: Useful for monitoring the process to see if it is in control. These are statistical plots of time ordered values of sample statistic
7.) Cause-and-effect (CNE) diagrams: Used to search for the cause(s) of a problem; also called Fishbone diagrams
8.) Run Charts: Tool for tracking results over a period of time

In addition, quality tools for generating ideas are:

1.) Brainstorming: Generates free flow of ideas in a group (pp 403)
2.) Quality Circles: Groups of workers who meet to discuss ways of improving products or processes
3.) Interviewing: Technique for identifying problems and collecting information
4.) Benchmarking: Process of measuring performance against the best in the same or another industry
5.) 5W2H Questioning Approach: Asking questions about a process that include what, why, where, when, who, how, how much for improving the process

Pareto Analysis/Diagram/Charts (80-20 Rule)

A method of organizing errors, defects or problems in a bar chart form to help focus or concentrate problem-solving efforts. Pareto principle says that relatively few (20%) of the causes, factors, variables; items, etc. are responsible for majority (80%) of the problems, effects, etc. Resources are always limited.

Therefore,
- Remember that all things are not equally important
- Identify key causes, factors, problems, etc.
- Focus your resources, efforts, and attention on the important few

Cause-and-Effect (CNE/Fishbone/Ishikawa) Diagrams

Most of the quality problems are caused by Materials, Methods, Manpower, and Machinery or equipment. Each of these categories can be focused on using brainstorming to identify the real or root cause of a specific quality problem.

Be careful when identifying root cause of a problem. For example, identifying "Bad weather" as a root cause of "Materials not getting delivered to customers on time" is inappropriate.

Businesses employ different approaches for controlling & assuring quality.

The three approaches (from least to most progressive) are:

1. Inspection before and after production (Acceptance sampling) 2. More Inspection during production & corrective action (Process control)
3. Building quality into process - fixing problems (Continuous improvement)
The level of inspection is the maximum in the first approach and is minimal in the third approach.

Role of Inspection

-Helps detect bad products before they reach customers.
- Does not correct the defects in the products or the deficiencies in the process responsible for these defects.
- It neither changes the product nor increases its value.
- Deming said that inspection is basically undesirable.
- A perfect process will not require any inspection of the output.
- Goal of POM manager should be to achieve this perfect process and eliminate inspection. However, in reality inspection may only be minimized.

Inspection is:

- Basically Undesirable
-Does not correct defects
- Does not add value
-Must be minimized

Basic Questions and issues regarding inspection

a.) How much to inspect and how often?
- as little as possible (This will determine your costs)
- as infrequently as possible (Use source inspection*)

b.) When to inspect? When should we inspect?
- when benefits exceed the costs However, remember there is no ideal/optimal level of inspection. Goal should be zero inspection.

c) Where to inspect?
- raw materials, purchased parts - finished goods
- before a costly operation
- before an irreversible process
- before a covering process (Ex: painting, assembly) - source inspection* (see below)
- On-site inspection (Ex: house must be inspected at the site)
Sometimes you need "On-site inspection" as opposed to Centralized insp.

Many operations managers like and require "______ _____________"

Source Inspection

What is Source Inspection?

where employees self-check their own work and that of the employees preceding them so that the immediate customer (the next step in the process) receives only quality output.

Quality is best assured by controlling the process and continuously improving it rather than through inspection.

"Control Process".
1) define
2) measure
3) compare
4) evaluate
5) correct
6) monitor

Statistical Process Control (SPC)

Statistical evaluation of process output during production.

SPC involves ________________ taking small size ______________ (measurements) from the process and comparing sample results with a predetermined standard.

Periodically
Samples

SPC (continued):

If the sample results are not acceptable the process is stopped and corrective action is taken. If the sample results are acceptable, the process is allowed to continue. The corrective action is only taken when the process exhibits excessive variability in the output i.e. when the process is considered out of control.

Every process exhibits variability. Dr. Deming showed that:
Common/random/chance causes of variation cause natural or normal variation.

Special/assignable/correctable causes cause abnormal (special) variation.

Normal/Natural Variation

in process output is due to inherent variability of the process components, due to chance or is created by countless minor, known or unknown factors

Abnormal/Special Variation

is a variation whose causes can be identified. For example, factors such a Tool wear, operator fatigue, lack of training, new supplier, etc. are assignable sources of variation.

If the variation is only due to natural or common causes, the process is said to be "____ ___________". The process that is in-control has a better chance of producing output that meets specifications or the customer expectations

In Control

Variation due to special or correctable causes will be excessive, throwing the process
_____ ____ _________. In this case, process should not be allowed to continue. It is stopped and assignable causes are identified and eliminated to bring the process back in control.

Out of Control

________ provides a statistical signal when assignable or correctable causes are affecting the process and the process is going "Out of control."

SPC

An important tool in SPC is "_________ __________." Control charts are used to detect presence of assignable causes of variation and to decide if the process is out-of-control at any time

Control Chart

Control Charts

Control charts are time ordered plot of sample statistics used to distinguish between normal (random) and abnormal (assignable cause) variability. All Control charts have Center Line and Upper and Lower Control Limits within which the process output is considered acceptable (i.e. process in control).

UCL and LCL Control Limits are dividing lines between normal and abnormal variations above and below the "Center Line". Process output is examined periodically to check if it is within these established limits.

Thus, control charts allow you to identify whether variation present is normal/random or special variation. In general, we can conclude that the process is in control if only random variation is present. Alternatively, we can conclude that process out of control if special variation is present. Still error is possible.

Types of Control Charts

Type I Error: Concluding that the process is not in control when it actually is.

Type II Error: Concluding that the process is in control when it is actually not.

Types of Control Charts:

(A) Control Charts for Variables Data (X-bar chart and R-chart)
(B) Control Charts for Attribute Data (p-chart and c-chart).

Variables generate data that are measured. Attributes generate data that are counted or recorded.

Control Charts for Variables Data

1. Mean Chart (X-bar chart) is for monitoring mean of the process output
(This tells us if changes have occurred in the process mean)
2. Range Chart (R-chart) monitors range (variability) of the process output
(This tells us if changes have occurred in the range of the output)

These charts should be used together to check and keep process in control.

Mean Charts

The construction of the Mean (X-bar) Control Charts is done as follows:

1. Take several samples of size "n" of the process output and find Mean and Range for each sample (i.e. find all sample means and sample ranges).
2. Determine Grand mean of the process output X. (i.e. Mean of the sample means). This will be the "Center Line" [C.L.] of the Mean control chart.
3. Determine the upper and lower control limits (UCL and LCL). (Limits are in multiples (2 or 3) of the std. deviation of the sampling distribution of the sample statistic).
4. Draw UCL and LCL lines above and below the Center Line. i.e. Grand mean X line which represents the process mean.
5. Plot the process sample means for next batches.
6. If within UCL and LCL, the process is in control.

Case A: Process Standard deviation is known (i.e. σ is known)

In this case, 95% UCL = X + 2*σ/ n
and 95% LCL = X - 2*σ/ n
while 99% UCL = X + 3*σ/ n
and 99% LCL = X - 3*σ/ n ......................

Case B: Second approach: Process Standard deviation σ is not known.

In this case, the "Average range" (R-bar) is computed from the range values of all samples which then is used as a measure of process variability, the process standard deviation. Grand mean is still the Center Line (C.L.)
Then, the UCL and LCL are computed using the following formulas:
UCLX = X + A2*R and
LCLX = X - A2*R....................................

R = Average range of all samples.
X = Grand mean (Mean of all the Samples Means). This is the C.L.
A2 = Factor found in Table 10-3 on pp 431 for your sample size.

This approach assumes that the process range is in control.

Range Charts: (Used to monitor process variability)

The construction of the Range Charts (R-chart) is done as follows:

1. Take several samples of size "n" of the process output and find range for each sample.
2. Determine mean range (R) of the process output. (i.e. Mean of the sample ranges). This R-bar is the Center Line of the R-chart.
3. Determine the upper and lower control limits UCL and LCL. (Limits are determined by looking up appropriate values of factors D3 and D4 from Table 10-3, pp 431 for the UCL, LCL formulas (see below))
4. Draw UCL and LCL lines above and below the average process range line (i.e. R line).
5. Plot the process sample ranges for the next batches.
6. If within UCL and LCL, process variability is in control.

Specifically, the UCL and LCL for the Range chart are given by formulas:

UCLR = D4*R
LCLR = D3*R

When the Process Mean and the Process Variability is in control, THE PROCESS is said to be ____ ____________.

In Control

_______ and _______ charts must be used together to check if the process is in control since each provides a different perspective on the process

Mean
Range

Once both of these control charts are set up, they can be used to monitor the process and for deciding when to interrupt the process and search for assignable causes of ____________.

Variation

Control Charts for Attributes Data

Used when process characteristic is counted rather than measured.
Ex: # of accidents, # of customer complaints
There are two types of charts for attribute data:

a.) p-chart: used for monitoring proportions

b.) c-chart: used for monitoring counts
These charts are conceptually similar to Mean and Range charts but the formulas are slightly different.