Manufacturing Strategy in the US
for Low Volume Demand
(PRESENTATION GIVEN AT EMCWA TECHNICAL
CONFERENCE, INDIANAPOLIS, 2004)
Jerry Kauffman, Norman Kopp and Scott Ward
Windings Incorporated
Abstract: There are plenty of cost effective
options for high volume manufacturing; most
likely, production will be automated. For
low volume motor manufacturing-ten thousand
units per year or less-there are a lot of
considerations while exploring the options.
What appears to work well for high volume
does not for low volume. Evaluating trade-offs,
capital investment and utilization is one
of the considerations. Additionally, training
in the short-term and long-term is important.
Magnet wire is not a robust material; it
reflects its handling easily and problems
show up in the motor tests. Training is
critical to handling magnet wire effectively.
Process flexibility needs to be considered
as well. After decades of Just-in-Time,
Flow Manufacturing, Theory of Constraints
and Lean, it may not be easy to establish
dedicated cells. Changeover and set-up reduction
are extremely important.
Exploring those considerations in this
paper, we'll look at the trade-offs of those
considerations in an attempt to optimize
value for the customer: quality, delivery,
technology, service (responsiveness) and
price. Various options will maximize one
value factor over another. The organization
must decide which are the most important-they
cannot all be equal.
I. INTRODUCTION
Automation and information technology has
led to great productivity gains in terms
of generating revenue compared to an investment
in direct labor. Worldwide there has been
a loss of several millions of manufacturing
jobs; worldwide includes the next great
threat-China. Productivity in all nations
on high volume production has meant less
people are employed in manufacturing.
What if your market is not large? The product
has a specific application and is different
than all other products (e.g. motors)? Does
automation make sense when the capital investment
is large, and on a per-piece basis could
make the motor's price non-competitive?
When the volume is not great, overseas production
can cause the lead-times to be excessive
also. For the last twenty years, there has
also been a move toward designing components
to fit particular applications, rather than
trying to have the application utilize stock
(catalog components). Therefore low volume
manufacturing has become a desire in the
US marketplace.
Developing a high mix, low volume manufacturing
operation, we can follow the Cause-Effect/Ishikawa
(Fishbone) Diagram elements:
materials,
machines/equipment (capital investment),
methods (processes),
manpower (training)
management (flexibility in scheduling, material
flow, creating demand for the capacity)
environment
information.
II. DISCUSSION
A. Material Purchases
In most manufacturing situations, volume
creates the ability to gain price reduction
as most material manufacturing is geared
to higher production volumes and rates.
In low volume manufacturing, production
lot sizes for motors may be in the hundreds,
tens or single units. Volume purchasing
for materials here may be a detriment as
inventory grows.
Twenty-plus years ago, inventory was an
asset (and still is shown as such on financial
statements), but since then we've learned
that inventory is a liability for cash flow,
operational flexibility and obsolescence
reasons. Inventory carrying costs have been
estimated from 15-55% depending on certain
risk factors. Unless price break for a higher
volume purchase warrants a discount greater
than that, it's better to buy in smaller
volumes though a higher price. Often, suppliers
can provide smaller volumes at a low price
because they've adapted to short setup times
(changeover times). Often that's the single
biggest opportunity for volume pricing,
amortizing the setup costs over a larger
number of pieces.
B. Capital Equipment Investment
In the authors' experience, there continues
to be a drive for customization in the motor
industry. Less satisfied with off-the-shelf
motors to put into the application, more
engineers are asking for motors designed
to meet their specific packaging and performance
requirements. This leads to many unique
designs in low volume manufacturing operations.
In order to build unique lamination configuration,
winding parameter and motor length, tooling
is created for automated equipment that
is specific to that motor design. Tooling
can run tens of thousands of dollars, which
effectively adds tens of dollars to the
unit cost of the motor.
Utilization of the equipment should be
high to keep return on the capital investment
high. Under-utilized equipment is a material
flow and space detriment, becoming more
of an obstacle than an aid to throughput
(revenue generation).
Simple, part specific tooling is often
inexpensive and sufficient to meet product
quality requirements.
Not only does the tooling need to work
with each design, but also it should be
created to accommodate quick setups and
changeovers. Otherwise, operational costs
increase. To minimize this, the tooling
challenge is to make it flexible enough
to handle several configurations.
In many operations, with low volume, there
are also small production lot sizes. Each
workstation should be able to have materials
readily at hand, and change from one job
to another smoothly and efficiently. Many
custom motor operations have a lot size
less than 10 pieces; some are approaching
the Lean Manufacturing (Toyota Production
System) ideal of 1 piece per lot., or 1
piece per stock move. In some operations,
dedicated assembly cells are not feasible
because of the product mix.
Short manufacturing cycle time and order
fulfillment lead-times are gained when this
is done well. Improved delivery and customer
service keeps the competitive advantage.
C. Process Design
Manufacturing processes have to be robust
enough to handle material piece-to-piece
and lot-to-lot variations. In low volume
circumstances, the processes also have to
tolerate a variety of materials. For example,
varnish impregnation equipment has to have
lines that can be quickly purged and reservoirs
that can be swapped out. Manual operations
may also work with smaller containers or
hand application.
Process control is difficult in many manual
operations. There is a reliance on test
and inspection in those cases. Short-run
Statistical Process Control (SPC) and mistake-proofing
(poka-yoke) techniques can help. Low-cost
jigs and fixtures should be utilized as
much as possible to reduce the potential
for errors. Inattention to details or oversight
can occur frequently. Some studies have
shown that dynamic (real-time) significant
changes or events can be overlooked. These
studies are a warning to those endeavors
that rely heavily on operator involvement.
There are some design configurations that
require process technology that is specific
to that motor. In this case, the customer
is looking for technology and not the delivery/service
flexibility nor necessarily the lowest price.
D. Training
Training programs need to focus on the
key aspects of the manufacturing processes
that apply to all parts. Specific instructions
for particular parts need to be part of
the documentation (Bills-of-Material and
routings-see Information below).
Basic training needs to include terminology,
material handling, process concepts, and
testing protocols. Operators should be taught
good decision-making and problem-solving
skills.
E. Management
Decision-making and problem-solving at
the operator level is critical in low volume
operations. Like a battlefield, the operational
conditions change quickly and frequently:
what parts need to be worked on; what projects
need to be squeezed in; what problems are
occurring now; what can't be worked on.
Schedules change due to expedites and rework.
If operators and their teams need to wait
for management or an engineering function
to make a decision or analyze and take corrective
action, operations will slow down. Wrong
parts will be finished instead of the ones
the customers needed today. Customer service
performance deteriorates.
Therefore, management needs to delegate
decision-making and problem-solving. Management
changes to coaching, mentoring and passing
along as much customer information as is
tolerable to the operators. It is difficult
to over-communicate.
F. Environment
Besides normal environmental factors of
the physical parameters (temperature, breathable
air, lighting, ergonomically sound work
stations and tools, clean/chemical free),
the work areas organization and layout is
key. Materials and tools can be found quickly
and close by. Work-in-process materials
flow naturally from station to station in
the shortest distance possible. As discussed
in the capital investment section, dedicated
cells may not be possible. However, sequential
and parallel operations for completing the
motor can be situated in the factory in
a sequential and parallel manner.
G. Information
Excellent Bills-of-Material (BOM) and routing/work
instructions are mandatory. The information
needs to be quickly accessed. Many operations
still rely on "travelers". This
can be difficult in other operations with
many parallel operations. Some have gone
to shop floor documentation systems accessed
by computer at centrally located PC's or
wireless personal computing 'tablets'. Because
of the variety in low volume operations,
it is important to instill the discipline
of reviewing the documentation for each
job, even if the operator build the same
type of units the day before or the morning
of the same day. Relying on "old"
documentation or memory will only cause
problems.
Having computerized shop floor control
system means there's an information technology
support service. This can add overhead cost.
However, the tradeoff of not having it and
having perhaps a lower price means that
the quality may suffer, delivery lead-times
may be slower unless more 'bureaucratic
systems are created. As one author's axiom
goes, poor systems generate more systems.
Schedule information needs to be flexible
and as up-to-date as possible to allow the
operators to make sound decisions about
their next activity. Customer application
information is also good to provide. It
gives operators an additional heightened
sense for the quality requirements that
are written or generally known, especially
in aerospace or military applications. It
may also increase creative choices for cost
reduction or problem-solving.
III. CONCLUSIONS
Low volume manufacturing has many challenges
and tradeoffs on customer value offerings.
Depending on the company's strategy, as
to which competencies to foster, the decisions
on how to set up low volume manufacturing
will create the customer value offering
that emphasizes singly or in combination
quality, delivery, technology, service and/or
price. The level of investment and flexibility
for low volume manufacturing will greatly
pre-determine the offering and the effectiveness
of the company to execute its contracts
accordingly.
Jerry Kauffman holds a BS in Mechanical
Engineering from the Montana State University.
He worked for 3M prior to coming to Windings.
He's currently the VP of Engineering and
Operations.
Norm Kopp holds a BS in Math/Physics from
Milton College and an MBA from Cardinal
Stritch College. He has worked for several
motor manufacturers throughout his career.
He is currently the Director of Specialty
Motors Business for Windings, Inc.
Scott Ward has a BA in Chemistry from Carleton
College (the lone liberal arts major of
this group). After working for several companies
and one consulting stint, he had led the
Operations group at Windings and is currently
VP of QA and Strategic Value, including
customer management and supply chain efforts.
|
