Most mechanical products are assembled from parts. After designing
the parts, they must be manufactured. Before FMS, parts for batch
production were created in machine shops by skilled machinists, who
would use general purpose machines such as lathes, drills and so on in
sequence to create the parts. To discuss the automation of such a
machine shop (creation of an FMS, which is also known as a
manufacturing cell), we must first consider the functions to be
accomplished at each machine:
a. Move the proper workpiece to the machine
b. Load workpiece onto the machine
c. Select proper tool
d. Establish and set machine speed
e. Control machine motion
f. Sequence different tools
g. Unload part
In mass production, the large volume makes having a special purpose
machine for each operation economical, but in batch production, it
simply is not economical to have special purpose machines for each
operation, because they would sit idle most of the time. The goal in
improving the efficiency of batch parts manufacture has been to create
computer controllers that will make general purpose machines flexible
enough to create multiple parts automatically. Since 1960, considerable
advances in FMS have been made from the numerically controlled machine
to the current flexible manufacturing cell, several general purpose
machines linked and controlled by a computer.
A schematic drawing of an FMS is shown below. Remember each of the
machines shown is a general purpose machine that must be programmed
through the steps a-g listed above.
The current country leading in this area is Japan. Fanuc, Ltd.
created a plant in the 80s that makes robots and CNC machine tools. The
plant is essentially an automated machine shop that produces parts for
these machines. Robots carry the parts from one group of machines to
another. Vehicles automatically store finished parts and retrieve raw
workpieces. There are 19 day shift workers and one night shift worker.
The use of FMS, instead of a general purpose machine shop staffed by
skilled machinists, can reduce the cost of manufacturing parts by a
factor of 5 to 10. With a general purpose machine shop, the machinists
spend a large portion of their time setting up the machines for the
next operation. With a FMS, this set up time is greatly reduced by the
computer which sets up the machines automatically. Because the set up
time is greatly reduced, machine time in creating parts increases from
3- 10% to 50% of the total time. Also, FMS requires from 10 to 30% of
the skilled labor that a general machine shop requires. Besides
reducing labor costs and increasing the output of the machines, a FMS
has two other advantages over a general purpose machine shop. With a
FMS, production can rapidly shift from one part to another. Thus, a FMS
can match supply to demand with very little inventory. In addition, the
use of FMS has led to much greater quality control.
A fundamental issue in the move to production for final demand, not only for a FMS but an entire factory, is how fast a machine can be converted from one job to another. After world war II Japanese manufacturers had powerful incentives to reduce the required land for manufacturing in order to compete internationally because Japan is about the size of California and only about 15% of the land is flat, thus in relationship to other countries land is very expensive in Japan. In the 50s Shigeo Shingo (1985) developed the SMED (single minute change of a die) system to reduce the required time and cost to shift a machine from the production of one product to another in less than 10 minutes. His SMED system consists of carefully observations to streamline and standardize the changeover process. Reducing the fixed cost of changeover makes smaller production runs economic in parts production and reduces the inventory costs of foregone return on investment and the space required to store the inventory. There are consulting firms selling the SMED system worldwide today.
The problems with FMS are the cost of setting up the stations and the
fact that technical expertise is required to set them up and run them.
If a component breaks, the FMS cell shuts down. Despite these
limitations, FMS has moved from the experimental to the rapid growth
stage. The demand by larger firms for higher quality control in order
to install just in time parts management has created incentives for
small firms to install FMS in order to achieve higher quality control.
An example of a small firm that installed an FMS cell is Frost, Inc.
with 1985 sales of $16M. For a $5.1M investment, sales per employee
have climbed from $86,000 to $130,000. Quality control has improved
from 1 reject in 4 to 1 reject in 20. Gross margins have increased to
35% in spite of price decreases of 21% since 1983. With its FMS, Frost
could shift from the production of one item to another in minutes
instead of 12 hours or more. Frost converted to automation at 1/3 the
cost of the expert plans. To capitalize on its experience with FMS,
Frost has set up a consulting company to advise other firms desiring to
follow suit.
The efficient route to automation is not to take several general
purpose machines and replace them with a manufacturing cell. This leads
to the term `islands of automation'. Installing manufacturing cells to
increase efficiency frequently requires reorganizing the entire
production process.
Some interesting sites to surf for FMS are:
The problem is that many plants produce multiple products on the
same assembly line. The right part must be at the right place at the
right time and any program to control this process must run in real
time. In the US, land is inexpensive so traditional US manufacturing
solved this difficult problem by having parts bins at each station so
that the workers could pick out the correct part. A simplified
schematic of this type of assembly line is shown in the diagram below:
This type of assembly line has many serious defects. First, much
inventory is tied up in the form of work in progress, that is the parts
in the bins. The firm does not obtain an economic return on these parts
until the product is sold. In countries where land is expensive this
form of assembly requires extra space for the parts bins. Second,
having parts in bins does not require much emphasis on quality control
since the worker can look through the bin to find a good part. In
traditional Detroit auto production, twenty five percent of the
assembly workforce fixed defects at the end of the assembly line.
After WWII the Japanese made many important innovations in
manufacturing. Since Japan is about the size of California and only
about 15 % is flat suitable for factories, and in Japan because land is
very expensive, manufacturers were encouraged to innovate in
manufacturing organization to save space. The most advanced plan, which
was developed by Toyota, is JIT, just-in-time, where the order for a
final product generates the orders for parts as they are needed (demand
pull). Toyota created this system using order cards without computers.
The ideal of JIT is that there should be no inventory; consequently,
every part must be perfect when it reaches the assembly line. At Toyota
parts are ordered from suppliers only as they are needed. Obviously to
make this work the suppliers must be located adjacent to the Toyota
factory. While some Japanese firms have been able to goad their
suppliers into this level of quality control and obtain instantaneous
coordination with suppliers, few firms outside of Japan have been able
to successfully implement JIT.
For example, in US auto assembly the parts firms are scattered over
several states. The manufacturer must keep a supply of parts on hand in
case transportation is interrupted by, for example, a major blizzard.
Thus, if the parts suppliers are distributed over a wide area it is not
optimal to try to reduce the inventory of parts on hand to one.
The US approach to improve the flow of parts and resources to obtain
substantial savings through reduction of inventory and wastage is the
creation of software inventory control packages. The US contribution is
called manufacturing resource planning, MRP, which schedules
the flow of parts as part of the forecasted production schedule. A more
advanced form of MRP is manufacturing resources planning, MRP II. MRP
II also considers the cash flow required to order the parts and pay
expenses in the forecasted manufacturing plan. It is important to note
that both MRP and MRP II are future oriented plans. US software
approach to lean inventory has advanced to Enterprise
Resource Planning, integration of all corporation information for
analysis and planning, and now Supply
Chain Management, reducing inventory at all levels from input, work
in progress, to output at the factory and in the distribution chain.
These information systems have expanded from just manufacturing
scheduling, to linking manufacturing to all office processes, to
linking the firm to the acquisition of inputs, and the delivery of
outputs.
A common feature of both MRP and JIT is the emphasis on reducing
inventory and making factories more efficient. JIT places more emphasis
on efficiency and quality.
Another aspect of automation in parts manufacturing is how parts are
delivered to their assembly point in a factory. In many older
manufacturing plants, parts are handled as many as 10 to 15 times from
the time they enter the factory until they reach their assembly
station. Obviously, the more a part is handled the greater the chance
for a fiasco. One advance in manufacturing automation is to automate
the delivery of parts to their final destination.
Ideally, parts from outside suppliers are handled twice. Once when they
arrive at the factory and once when they are assembled into the
product. One of the first firms in the US to do this was Apple in the
production of the first Macintosh around 1982. Jobs and the chief
engineer, Irwin, spent two years (probably part time) studying Japanese
production methods. The original Macintosh factory incorporated three
basic concepts:
a. Just in time parts delivery
b. Linear production system
c. Good environment for workers.
The first Mac had about 500 parts. To supply the parts to the linear
assembly line, Apple installed three automatic parts delivery systems
and one manual system:
a. Totes or plastic bins: These stored electronic parts
b. Overhead rail: This delivered bulkier items.
c. Automatically guided vehicles (AGV): Delivered other items
d. Humans delivered screws once a month or so because automating this
delivery would be too expensive.
The trend in automatic delivery of parts is in making advances in AGV
systems and automatic warehouses. To eliminate the possibility of a
Murphy's Law type foul up, parts should be handled twice: when they
arrive and when they are installed on the product. When they arrive,
they would be placed in an automatic warehouse until needed. The
program controlling the assembly process would send parts automatically
to the various assembly sites as needed.
In order to affect Kanban the quality of the parts must be extremely
high because the one part must be correct when delivered. As it has
been pointed out, about 25% of the US auto assembly workforce was
engaged in repairing defects when the autos rolled off the assembly
line before the recent move to better quality control. The old style
quality control, at which the Germans were the masters, was to have
teams of inspectors at various stations and test the products. Because
QC was thought to require additional workers, QC was considered an
added cost. Hence, as you might expect, higher quality products would
cost more.
The new approach to quality control pioneered by the Japanese totally
upended the cost quality relationships. In the 1930s the concepts of
statistical quality control, SQC, were invented in the US. After WW II,
Deming and Juran, SQC experts, could not convince US manufacturers to
adopt SQC. Deming and Juran then went to Japan where they were treated
like heroes as the Japanese manufacturers rapidly adopted SQC, which
enables manufacturing engineers to identify problems in the production
process without inspectors. S. Taguchi introduced the idea of
designing products so that performance is not affected by minor
defects, Ealey (1994).
The Japanese also achieved better quality control by carefully considering the human factors in manufacturing. Shigeo Shingo (1986) created the Poka Yoke system of quality control by systematic observation of the manufacturing process incorporate steps such that quality control become an integral part of the production process. To make continual improvements in quality a permanent part of the factory, Japanese managers have organized workers into quality circles that meet weekly. In these meetings workers propose improvements that engineers and managers review and then implement the best ideas. By eliminating the need for repair workers and inspectors, the new approach saves money and creates satisfied customers. These innovations lead to the Toyota or Japanese system of manufactures. Toyota products have been more reliable than their US counterparts and Toyota has greatly increased its share of the US auto market.
Since the 1980s US and European firms have imitated Japanese quality control concepts. For example, the quality of US automobiles has greatly improved and is slowly closing on the constantly improving Japanese quality standards. Currently, the Europeans have created an international quality standard called ISO9000. Many US firms are becoming certified as having met these standards.
Some interesting sites to surf for parts coordination and quality are: