[ Pobierz całość w formacie PDF ]

(distributor - warehouse - retail).
Re-work inventory is that material which is found defective. The material is held until
someone has time, or it can be scheduled, for correction and modification.
Inventory turnover
Inventory turnover, like asset turnover, is a useful tool to measure performance and
continual improvement. It is an accounting tool but offers some use to operations
managers as a conversion to JIT occurs.
Annual cost of material
Turnover
Value of current inventory
Many companies who thought that turns of 3-7 were good are now reaching 20, 30
and higher using JIT concepts.
Consider the example of a manufacturer who purchases $8,000,000 of materials a year
and maintains $2,000,000 in raw material inventory. The turnover is 4. Consider the
benefit to the company if, over time, you can increase the turn to 20. You will have
freed up $1,600,000 in available capital. Most companies use an internal rate of return
for capital at around 25%. This means that an additional $400,000 could be added to
the end of year profit. I emphasize  could. What the firm does with the added capital
can be debated for eternity. This kind of improvement can be done with no added
cost.
Transfer systems (methods, processes and equipment used to move materials from
one location to another) underwent significant change during the past 20 years. TQM
and JIT philosophies call for the transfer to be automatic and to assist with the release
of material to the next work station.
166
Lesson Six - Inventory, Scheduling and Robotics
Queuing
Queuing is the lining up of material at a station or the people waiting for service. An
advantage of JIT is that it focuses on queuing and bottlenecks. Bottlenecks exist any
time you do not have a balanced, well maintained line. Bottlenecks result in increased
queue length and therefore are a control mechanism for smooth system operation.
Queuing can be as obvious as people lined up at a grocery checkout line or bottles
moving thorough a production line. Queuing can also be less obvious, such as the
short delay in getting a long distance call switched. In this last example, you are not
aware of the other callers in the system.
Queuing theory carries with it a massive amount of mathematics which can be used to
define and work with queues. Queuing follows all three distributions: normal,
Poisson, and exponential. Normal is the typical bell-shape curve. Poisson is a skewed
distribution such as defects per 1000 items -- the heaviest concentration of occurrences
is closest to zero. Exponential is a distribution rising, or falling, sharply from the zero
point.
The intent of all the mathematics is to simulate the number of times that the product
or customer will not be served within an appropriate time after queuing. This is
important from a marketing stance when designing a service process. It is important
from a manufacturing stance in designing production flows.
This type of simulation before a plant is built can save millions of dollars in
production and improve customer satisfaction. Simulation can be performed after a
facility is built, but usually offers less gain. C. D. Lewis's book Scientific Inventory
Control is a good source of mathematical formulae to simulate amount in queue,
length of queue and mean wait time.
The new method of examining this problem is to build modular designs and modular
queuing flow to accommodate increasing and decreasing client flow within a
presumed range.
Here is one example in which a grocery store increased its sales by using customer
behavior in queue. The medium sized grocery store developed a marketing campaign
capitalizing on the client's dislike for lines. The advertising copy read that  three is a
crowd. If any check- out line reached three people, another check-out clerk was
immediately summoned for assistance to move clients through.
For the first week, clerks were announcing on the intercom that "Three's a Crowd" and
additional clerks would appear to move the queue through. Soon, if a clerk did not
notice the queue, the client's did and shoppers would take up the cry of  Three's a
Crowd." This is an excellent example of verbal KANBAN.
Kanban
Kanban means visual record. It is a means of requesting additional material or WIP or
to signal that more is coming to the next station. The simplest means of KANBAN is
the verbal record -- "Give me more"; "Can I help the next person in line, please?"
Colors on cards, lights, or the tray used to hold the item can also be used as a signal
that more material is needed. An example would be the workspace or the tray used to
store and transfer material. An electronics firm produces sensitive measuring
167
Total Quality Management
equipment. The components are transferred from work station to work station. A cart
carrying five units is rolled up to a work station. Transfer employees watch the work
and when a cart is empty, they know that it means another cart of five units must be
moved into place. This may be done by electronic means too. An example would be
the computer record of an item sold and a picking system releasing the next item for
delivery. Marketing enthusiasts will recognize the similarity of Kanban to a pull
strategy.
Scheduling
Items in queue are usually worked on in a "First In First Out" (FIFO) scenario.
There are other methods employed today.
In the "Latest schedule" scenario, any item arriving at a work station in which the [ Pobierz całość w formacie PDF ]