Chapter 6S; Statistical Process Control

Control Chart

A graphical presentation of process data over time.

• 2 types of these charts → variables & attributes

Assigned Variation

This variation is caused by factors that can be clearly identified & possibly managed.

• SPC tools help detect and eliminate assignable causes of variation.

• This is the bad guy

  • Examples: Poorly trained employees, misadjusted equipment, machine wear, & new batches of raw materials.

Natural Variation

This variation is due to random causes and is inherent in the production process.

• Random and unexpected

  • Examples: Delivery driver, having heavy traffic due to an accident randomly, but typically no traffic.

Continuous Data

Quantitative information that can take any value within a given range and is measured rather than counted.

• Called variables

  • Example: Weight, height temperature, lead time, elapsed time, cost, sales.

• Use x-bar and R control charts together

Discrete Data

Numerical information that you can count using whole, distinct, and separate values.

• Called attributes

  • Example: A name can be spelled wrong 2 or 3 times, not 2.5 times.

  • Examples: # of broken seats in auditorium, # of paint imperfections in a vehicle.

• Use p or c control charts

An Attribute

A characteristic that is evaluated with a discrete response such as good/bad, yes/no, (smell, texture, taste, cleanliness, color, etc)

• Use p & c charts for control charts

A Variable

A characteristic that is evaluated on a continuous scale (length, weight, height, temperature, time, speed, etc).

• Use Mean (x-bar) & range (R) charts for control charts.

Benefits of SPC Tools

Benefits include:

• Improved product/service performance.

  • Ex: Sheraton Hotels takes care of service issues within 15 mins.

• Lesser need to inspect all units/transactions.

• Improved customer satisfaction and higher market share from improved quality.

Statistical Process Control (SPC)

A process which involves monitoring the manufacturing or service process using statistical tools to minimize defects/errors.

• Detects assignable variation and helps minimize them.

When is a Process in Control?

A Process is IN control, when there are no outliers in both of the charts.

When is a process OUT of control?

A process is OUT of control when there are outliers in either chart.

6 Sigma

Occurs when the control center limit, exactly matches the customer spec center limit.

• The LCL and UCL will not match the customer spec, but they will fall above LCL, and below UCL.

• Dispersion is VERY narrow.

3 Sigma

Occurs when the control limits exactly match the customer specs.

• Should show a normal or wider distribution.

DMAIC

Define, Measure, Analyze, Improve & Control

• Six sigma is based on this problem solving method.

Process Stability

A process is stable if there are no outliers within either of the charts.

• If there is an outlier the process is not stable.

Process Capability

The ability of a process to meet design specs.

• A process is capable is the customer spec limits meet or fall in between the control limits.

Specification Limits

These limits are given by the customers.

• Ex: Order must be ready in 5 minutes + or - 1.

Control Limits

These limits are computed statistically.