Scribd
Upload
9 views

2C - Reacting To Quality Problem

The document discusses different types of quality problems based on their recurrence or frequency of occurrence: 1) One-time occurrences, where the cause is usually obvious and can be addressed without extensive investigation. 2) Low volume or occasional problems, which are harder to solve due to the lack of consistent data but can be investigated using tools like fishbone diagrams or comparing good and bad products. 3) Chronic recurring problems, which provide the most data to analyze and test potential causes and solutions through experiments since the problem happens regularly.

Uploaded by

Devarsh Kaka
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
9 views

2C - Reacting To Quality Problem

The document discusses different types of quality problems based on their recurrence or frequency of occurrence: 1) One-time occurrences, where the cause is usually obvious and can be addressed without extensive investigation. 2) Low volume or occasional problems, which are harder to solve due to the lack of consistent data but can be investigated using tools like fishbone diagrams or comparing good and bad products. 3) Chronic recurring problems, which provide the most data to analyze and test potential causes and solutions through experiments since the problem happens regularly.

Uploaded by

Devarsh Kaka
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 23

Structured Problem Solving

Reacting to Quality Problems: Recurrence


• 1-Time Occurrence
• Low-Volume/Occasional Problems
• Chronic Recurring Problems
Does this seem
like a reasonable
reaction to a
problem??
Week 2 Agenda
I. Product Non-Conformances
this is our starting point
II. Forming a Problem Statement
the 1st formal step in structured problem solving
III. Reacting to Quality Problems
Launching (or not) Structured Problem Solving
& Choosing the right Method
Response « Recurrence
Even outside the Quality profession, we often
consider recurrence (repeat occurrence) when
determining an appropriate response.
For example;
• First-time speeding offense  warning
• Second Offense  ticket
• Chronic Speeding  lose license
Recurrence
As recurrence (the rate of occurrence) increases, it signals
that the problem is more serious and potentially systemic,
or “built in” to the process.
• A process makes 1 bad part  minor issue, maybe just
a “mistake.” Keep making parts.
• A few bad parts in a day  need to review the work
method to understand cause. Keep running.
• Process makes mostly bad parts  there is a
fundamental flaw in their process. Stop work and resolve
before running any more parts.
Quality Problems

Recurrence

Low Volume Chronic


1-Time
or Occasional Recurring
1-Time Occurrence
For one-time failures, we often already know what went wrong.
• If your cell phone battery is dead, you might realize you
forgot to plug it in. We call this an “assignable cause.”
• Similarly, if an experienced machine
operator makes a mistake on a part, they
likely know what they did wrong, or can
figure it out with just a few quick checks.
• They might discover the cutting tool is
worn – and once that “assignable cause”
is fixed – the tool replaced – they can
continue with confidence.

Image source: https://www.ziprecruiter.com/blog/machine-operator-job-description-sample-template/


1-Time Occurrence
For one-time failures, we often already know what went wrong.
• In many industrial processes, there are a lot of things that
can go wrong.
• Tools wear out, parts shift in their
fixtures, operators get distracted,
someone forgets to make a program
adjustment or misreads a number off a
drawing – it’s a long list.
• Practically speaking, it isn’t possible to
track down the root cause in every case.
• Unless the parts are very expensive or
for some critical application (a part for a
space satellite) the problem-solving is left
to the operator.
Image source: https://www.ziprecruiter.com/blog/machine-operator-job-description-sample-template/
5Y for Critical 1-Time Occurrence
For more critical one-time failures: 5-WHY
• If the mistake is costly enough – the part is either very expensive or
there is zero tolerance for any errors (pharmaceutical, military,
nuclear, etc) – then there is reason to investigate more formally.
• Since this is a 1-time occurrence we
don’t have much historical data to
analyse. This rules out most statistical
methods.
• We also don’t want to “guess and test” –
since this might lead to another bad part
that we can’t afford.
• A common problem-solving tool we use
in this case is the 5-WHY

Image source: https://www.pharmaworldmagazine.com/containment-highly-potent-active-pharmaceutical-ingredients/


5Y for Critical Single Occurrence
For more critical one-time failures: 5-WHY
• The 5-WHY tool is well suited for single occurrences.
• The questions start with the person who ran the part or process
• “Why did the seal on this highly potent
pharmaceutical bottle come off?”
• If done properly, asking “why” in a respectful,
open and analytical way can lead to the
discovery of the root cause.
• The root cause can then be corrected, with a
high confidence in future product.
• Notice that this method tends not to use
statistical analysis, brainstorming, cause and Image source:
effect analysis, or any of the more sophisticated https://www.pharmaworldmagazine.com/contai
nment-highly-potent-active-pharmaceutical-
problem-solving methods we will explore next. ingredients/
Recurrence
Recurrence

Low Volume or Chronic


1-Time
Occasional Recurring

Do Nothing
(leave it to the
operator to resolve)

5-Why
Recurrence
Recurrence

Low Volume Chronic


1-Time
or Occasional Recurring

Do Nothing ?

5-Why
Occasional or Low Volume Occurrence
When the same problem appears every once in a while, then it can’t be
discounted as a “one time occurrence” and we need to put more effort
into investigating the root cause.
Some examples;
• A few parts are rejected for the same dimensional non-conformance
on some, but not all production lots.
• A very small percentage (<<1%) of pop bottles are underfilled
• Some custom-designed brackets made to order for an airplane
cockpit panel are out of tolerance

Image source: https://www.fergusonpackaging.com/listings/complete-beverage-bottling-line-4/


Occasional or Low Volume Occurrence
These kinds of problems may actually be the hardest to solve.
• The process normally makes good parts.
• Since there are very few bad pieces, there is not a lot of data to use
in an investigation
• If the problem does not occur in every production lot, or happens
only occasionally, it is hard to see if a process change has made
any difference. You can’t predict when a bad part would have
occurred.
• If the process does not run very often, it is hard to run a design of
experiments (DOE) since this requires multiple production lots which
could take months or longer to complete.
Recurrence Recurrence

Low Volume Chronic


1-Time
or Occasional Recurring

Do Nothing Fishbone

5-Why Is/Is Not

Trial & Error

8D if Critical
Fishbone Diagram
Problems that have low recurrence rely more on expert opinion
than experimentation.
A widely used tool that relies on
expert opinion more than testing is:
• Ishikawa Diagram
(aka Fishbone or
Cause & Effect Diagram)

You will have encountered the Ishikawa/Fishbone diagram in other


courses, and may have used it in the workplace or in assignments.
It is very effective at collecting ideas and working through the potential
causes without running any parts or trials.

Image source: https://asq.org/quality-resources/fishbone


Is/Is Not
Another simple tool that could be applied to occasional or low volume,
low recurrence problems is:
• Is/Is Not

Like the Fishbone brainstorming tool, this tool does not require
experiments or a lot of data.
The team simply compares the bad product to good product, to find
clues that might point to a potential cause.

We will practice using this tool later in the course. It can be very useful
as part of a structured problem solving approach as well, so is worth
learning.
Trial & Error (aka Guess & Test)
Trial & Error is NOT a structured problem solving methodology
- But it’s still very useful!
Don’t discount the value of Trial & Error, at least at the beginning of a
problem solving process:
• Quick & Easy – you might get lucky!
• Let’s operators and “experts” try out their ideas
• If a trial fails, you still learn something about the problem

It is natural and reasonable to run some trial and error when a problem
arises. Sometimes it works. The trick is to know when to stop guessing
and start using a more structured approach. One good indicator is
when the next trial is going to be expensive or time-consuming. Spend
the time and effort to do it “right” instead.
Recurrence
Recurrence

Low Volume Chronic


1-Time
or Occasional Recurring

Do Nothing Fishbone These aren’t technically


problem-solving
“Methods” – they are just
5-Why Is/Is Not
“Tools” we can use in this
situation. We will focus on
Trial & Error problem-solving tools
after success week.

8D if Critical
Recurrence Recurrence

Low Volume Chronic


1-Time
or Occasional Recurring

Do Nothing Trial & Error ?

5-Why Fishbone

Is/Is Not

8D
Chronic Recurring Problems
From a problem-solver’s perspective, this is the kind of problem we
like. (Don’t tell that to the boss!)
• If the problem appears regularly, we can watch the process and see it
happen
• We can collect a lot of data to compare good and bad parts
• We can run experiments and quickly see if the problem goes away
• When we implement a solution, we can clearly demonstrate it works

Process Change

All Good! Many Bad!

Image source: https://www.fergusonpackaging.com/listings/complete-beverage-bottling-line-4/


Chronic Recurring Problems
There are several Structured Problem Solving Methods that can be
used for Chronic, Recurring Problems.

• 8D
• DMAIC
• Shainin Red X
• Statistical Engineering Algorithm
• Kaizen
• etc
Recurrence Recurrence

Low Volume or Chronic


1-Time
Occasional Recurring

Do Nothing Trial & Error 8D

5-Why Fishbone DMAIC

Is/Is Not Shainin Red X

Statistical
8D Engineering
Algorithm

Kaizen

You might also like