Here
are typical questions/comments on the Barringer Process Reliability
technique.
The issue:
Do we have a maintenance problem or
a process problem? Each side says the
other party is responsible for lost production.
We need a tool to decide what area to attack and set priorities for
corrective action. The Barringer Process
Reliability technique is a method for answering questions about losses in a
quantitative manner.
How to resolve the issue:
Barringer’s Weibull process
plots give visual displays about the categories of losses and who is
responsible for initiating corrective action.
Process reliability plots provide, on one page, vital signs of how well
the maintenance and production systems are functioning along with evidence of
who is responsible for solving the problems along with establishing problem
solving priority. Input data for the
analysis is daily production output of prime product.
The revenue stream from any process
is dependent upon daily production of prime product. Consistency of daily output is a precursor
for how smoothly the money machine functions as a key business index.
We use daily production output
of prime product and Weibull probability plots as a decision tool for identify
the types of problems and where corrective action should occur.
When production is in a state of
statistical control, the daily output makes a straight line on Weibull
probability plots. Steep straight lines
(small variability in output are synonymous with steep line slopes defined by the Weibull statistic beta, b) are desired. Steep trend lines illustrate good control of
the process (variability is due to common cause variation which are many in number
and common cause variations are difficult identify and correct). World class processes show small variations
in output which generates a steep slope measured by the probability plots
statistic called a shape factor beta.
World class betas are equal to or greater than 100!
Beta shape factor values can be
benchmarked for one-to-one comparisons of similar processes which avoids the
“My process is better than your process and thus I don’t need to make
improvements—I’m happy, just leave me alone”.
The characteristic output of the
process (defined by a Weibull statistic value called eta, h) is also a benchmark value. Eta is the single point determination of the
demonstrated maximum asset capability.
The demonstrated maximum asset capability near the name plate
(entitlement value) of the design point is desirably.
Loss of statistical process control
shows production tending toward lower output with a cusp(s) on the Weibull
plots. The cusps are due to things (special
cause variability is due to things you can see and understand), which
are often maintenance related. The first
cusp toward lower production defines a point estimate of the process reliability (high values are
desired).
The Barringer process reliability
technique identifies/quantifies hidden factory costs for removal. The hidden factory is in stealth mode, you’re
paying both fixed and variable costs but you’re not getting profitable
results. Hidden factories are wasteful
and paid for by the customer or the stockholder as they add no value for the
customer or the stockholder. Stealth
waste is under your nose, use the process reliability
technique to:
1.
Quantify
the value of lost production in the hidden factory (how big is it, where is it,
and what’s it worth to correct), and
2.
Set
a strategy to recover the losses by taking corrective action on a Pareto
priority basis to destroy the waste.
When was the first process reliability plot created?
In 1995, data for the first process
reliability plots used feed data delivered to a refinery crude tower. The Weibull plots demonstrated desirable steep
trend lines and desirable high reliability typical of a good process with few
losses.
Cusps on the trend lines were
directly related to known maintenance issues.
Losses were quantifiable gaps between the demonstrated production line
and the actual data points falling to the left (smaller output) of the
demonstrated production trend lines. The
gaps matched losses from the reliability issues. These problems are solved by maintenance
engineers and reliability engineers.
In 1996, nameplate lines were
applied to the process reliability plots.
Gaps between the demonstrated production lines and the nameplate line
quantified common causes losses between actual results and what could be
achieved from the asset. Reasons for
the gaps between the demonstrated production line and the nameplate line are
due to efficiency and utilization issues.
These problems are solved by management and six-sigma black belt
experts.
Where have process reliability plots been used most effectively?
The process reliability technique
provides a quick and qualitative assessment of production/maintenance problems
and quantifies the size and cost of the hidden factory which detracts from
profitability.
Management driven efforts to
increase profitability by attacking the most important issues for reducing
losses are most effective. The plots
reflect before/after conditions to monitor progress. Consider what makes reliability programs success and
unsuccessful.
Continuous processes have the
largest opportunities for reducing losses using these non-traditional tools.
The process reliability plots are
useful communication tools between management groups to explain and quantify
problems.
Who has used the process reliability tool most effectively?
Major
chemical plants and refineries have been most effective in motivating
improvements because of their need to improve profits from a top down
perspective.
The least effective users have been
in industries where use of new tools and new approaches are discouraged,
particularly when driven as a bottom-up engineering approach.
What’s the greatest single improvement on record at a single plant
using the process reliability technique?
$143,000,000/year is the largest
single project audited savings from a chemical plant.
Removal of an older process and
installation of a newer process resulted in doubling of the beta (line slope)
which cut losses of the hidden factory in half.
This project involved the Plant Manager pushing for the changes. The Plant Manager actually spent two days in
the Process Reliability training classes to emphasize to his organization how
important it was to use new tools to make new breakthroughs to improve profits
by eliminating losses.
Geographically, where are process reliability plots in use?
How many people have completed the process reliability training
classes?
Over 1500 engineers have been
trained in Process
Reliability to find the hidden factory along with assessment of the
reliability of the process.
Approximately 10% to 20% of the group (150 to 300 people) represents
expert users of the technique, and about 1% to 4% of the group (15 to 60)
represents super users of the technique.
For other Barringer process reliability articles and examples:
·
Use Periods Of Low Production
Output To Improve Process Reliability And Consistency
·
Special Cause Variations, Common
Cause Variations, and Process Reliability Plots
· Process Reliability Punch List
·
Production Reliability Example
With Nameplate Ratings
·
Key Performance Indicators From
Weibull Production Plots
·
Process Reliability Plots With Flat
Line Slopes
· Process Reliability Line Segments
·
Automating Monthly Weibull
Production Plots From Excel Spreadsheets
·
Papers On Process Reliability As
PDF Files For No-charge Downloads
- New
Reliability Tool for the Millennium: Weibull Analysis of Production Data
- Process
Reliability and Six-Sigma
- Process
Reliability Concepts
Comments:
Refer to the caveats
on the Problem of the Month Page
about the limitations of the following solution. Maybe you have a better idea
on how to solve the problem. Maybe you find where I've screwed up the solution
and you can point out my errors as you check my calculations. E-mail your
comments, criticism, and corrections to: Paul Barringer by clicking
here.
You can download a copy of this page as a PDF
file.
Return
to Barringer & Associates, Inc. homepage