Docstoc

Statistical Thinking and Applications

Document Sample
Statistical Thinking and Applications Powered By Docstoc
					Chapter 11


 Statistical
Thinking and
Applications   1
Key Idea

Raw data collected from the field do not provide
the information necessary for quality control or
improvement. Data must be organized, analyzed,
and interpreted. Statistics provide an efficient
and effective way of obtaining meaningful
information from data, allowing managers and
workers to control and improve processes.
Statistical Thinking

 All work occurs in a system of
  interconnected processes
 Variation exists in all processes

 Understanding and reducing
  variation are the keys to success
    Sources of Variation in
    Production Processes
                                             Measurement
                 Operators     Methods
Materials                                     Instruments


        INPUTS        PROCESS            OUTPUTS


Tools                                          Human
                 Machines    Environment       Inspection
                                               Performance


                                                       4
Variation

    Many sources of uncontrollable
     variation exist (common causes)
    Special (assignable) causes of variation
     can be recognized and controlled
    Failure to understand these differences
     can increase variation in a system


                                            5
Key Idea

A system governed only by common causes is
called a stable system. Understanding a stable
system and the differences between special and
common causes of variation is essential for
managing any system.
Problems Created by
Variation
   Variation increases unpredictability.
   Variation reduces capacity utilization.
   Variation contributes to a “bullwhip”
    effect.
   Variation makes it difficult to find root
    causes.
   Variation makes it difficult to detect
    potential problems early.
Importance of
Understanding Variation

                        time
                   PREDICTABLE




               ?   UNPREDECTIBLE
Two Fundamental
Management Mistakes
1.   Treating as a special cause any fault,
     complaint, mistake, breakdown, accident
     or shortage when it actually is due to
     common causes
2.   Attributing to common causes any fault,
     complaint, mistake, breakdown, accident
     or shortage when it actually is due to a
     special cause
Note to Instructors

   The following slides can be used to
    guide a class demonstration and
    discussion of the Deming Red Bead
    experiment using small bags of M&Ms,
    from a suggestion I found on a TQ
    newsgroup several years ago. The
    good output (“red beads”) are the
    blue M&Ms, with the instructor playing
    the role of Dr. Deming.
We’re Going into Business!!!

We have a new global customer and
have to start up several factories. So I
need teams of 5 to do the work:

1   production worker
2   inspectors
1   Chief Inspector
1   Recorder
Production Setup

1. Take the bag in your left hand.

2. Tear a 3/4” opening in the right
   corner. (only large enough for
   one piece at a time)
Production Process

1. Production worker produces 10
pieces and places them on the napkin.
2. Each inspector, independently,
counts the blue ones, and passes to
the Chief Inspector to verify.
3. If Chief Inspector agrees, s/he tells
the recorder, who reports it to me.
   Do it right
    the first
      time!




                       Take Pride in
                       Your Work!
Be a Quality Worker!
Lessons Learned

   Quality is made at the top.
   Rigid procedures are not enough.
   People are not always the main
    source of variability.
   Numerical goals are often
    meaningless.
   Inspection is expensive and does
    not improve quality.
Statistical Foundations

 Random variables
 Probability distributions

 Populations and samples

 Point estimates

 Sampling distributions

 Standard error of the mean
Important Probability
Distributions
   Discrete
    – Binomial
    – Poisson
   Continuous
    – Normal
    – Exponential

                        17
Central Limit Theorem

   If simple random samples of size n are taken
    from any population, the probability distribution
    of sample means will be approximately normal
    as n becomes large.




                                                    18
Sampling Methods

 Simple random sampling
 Stratified sampling

 Systematic sampling

 Cluster sampling

 Judgment sampling
Key Idea

A good sampling plan should select a sample at
the lowest cost that will provide the best possible
representation of the population, consistent with
the objectives of precision and reliability that
have been determined for the study.
Sampling Error

 Sampling error (statistical error)
 Nonsampling error (systematic
  error)
 Factors to consider:
    – Sample size
    – Appropriate sample design
Statistical Methods

 Descriptive statistics
 Statistical inference

 Predictive statistics
Statistical Tools
Excel Tools for Statistics

   Tools…Data Analysis… Descriptive
    Statistics
   Tools…Data Analysis…Histogram
Key Idea

One of the biggest mistakes that people make in
using statistical methods is confusing data that
are sampled from a static population (cross-
sectional data) with data sampled from a
dynamic process (time series data).
Enumerative and Analytic
Studies
   Enumerative study – analysis of a
    static population
   Analytic study – analysis of a dynamic
    time series
Design of Experiments

   A designed experiment is a test or series
    of tests that enables the experimenter to
    compare two or more methods to
    determine which is better, or determine
    levels of controllable factors to optimize
    the yield of a process or minimize the
    variability of a response variable.
   DOE is an increasingly important tool for
    Six Sigma.

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:180
posted:2/18/2010
language:English
pages:27
Description: The Scope and Language of Operations Management by James R. Evans.