O'Reilly logo

Achieving Lean Changeover by John R. Henry

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

141
Chapter 7
Execute
In an ideal world, after cleanup and setup are completed, it should be pos-
sible to restart the process and have it run immediately at full efficiency.
This is the vertical startup that was discussed in Chapter 1. This startup time
may be even greater than the time spent on cleanup and setup combined.
In a few extreme cases, the production run may finish before the process is
completely settled down, running at normal speed and efficiency.
There is a single cause of nonvertical startup: variability.
Some startup is caused by variability of the materials. Changeover
improvement does not usually directly address material variability. This is
usually the responsibility of the quality, materials, manufacturing, and other
departments involved in developing and maintaining material specifica-
tions. One issue is that if changeover is not done precisely, it is difficult to
determine where the issues lie. A finger-pointing game can develop with
each side blaming the other. Manufacturing may say that precise setup is
not possible because of material variability. The materials department may
say that materials are in spec, and that the problem is that manufacturing is
unable to set the machines properly. Each side is often at least partially right.
Manufacturing must get the changeover under control. Once they do, and
can show that they can perform them repeatably, time after time after time,
they will be able to demonstrate the delays caused by material variability.
These delays need to be noted and quantified. They should be quantified
in dollars if possible or in time if the cost per hour is not available. Once
quantified, a tool exists to justify reducing material variability.
Most startup is caused by variability of the changeover. Setup is generally
the main culprit but variability in cleanup sometimes causes startup delays
142 ◾  Achieving Lean Changeover: Putting SMED to Work
as well. If the variability can be driven out of changeover, startup times will
be dramatically reduced.
Execution is the key to driving out variability. Absent proper execu-
tion of the changeover, much of the value of the elimination, simplifica-
tion, and externalization will be lost. The improved changeover must be
executed exactly.
Proper changeover execution depends upon several factors. Well-written
and comprehensive standard operating procedures (SOPs) and checklists
are one. These were discussed in detail in Chapter 3. Proper supervision of
changeovers to assure that they are properly done is another. Quantifiable,
repeatable, exact adjustment is one of the most important factors and is the
focus of this chapter. Ideally, it should be possible to set everything to the
target set point and run. In reality, although much variability may be driven
out of the setup, some will still remain. SOPs must be written to give lati-
tude to adjustment so they can be tweaked as needed. When tweaking is
required, the actual run setpoint must be recorded in an equipment log or
database. Where the reason for the variance from the SOP is known, it must
be noted. If the variance becomes consistent, perhaps because of machine
wear, this should generate either a maintenance action to bring it back to
ideal or a revision of the SOP to reflect the new required settings
It is critical to measure and standardize the elements of changeover.
Equally critical is the measurement of the changeover itself. Every change-
over must be measured and recorded. This should be done graphically with
trend lines shown. Discrepancy reports need to be generated when change-
over takes significantly more time than normal. The discrepancy report must
be investigated and the reasons for delay corrected as necessary. A discrep-
ancy report should be also generated when the changeover takes less time
than normal. If a normal 4-hour changeover is finished in 3, it may be that
something was not done that should have been. This must be corrected. In
other cases it may be that the hour was saved because there were no delays
and everything just went right. This should be noted as well since it may
give clues to permanently improve the changeover.
One plant had different departments running almost the same processes
on the same equipment. Department 1 took 30–35 hours to do a major
changeover, mostly consisting of cleaning. Department 2 did the same
cleaning but normally took 20–25 hours to complete it. When this was first
discovered (after it had been going on for years), the first thought was that
Department 2 was not following the SOP or took shortcuts. Closer evalua-
tion showed that they were following the same process, they were just better

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required