The Philosophy of Opt

1
The Philosophy of Opt
2
Theory of Constraints
TOC (Theory of Constraints) It focuses on
constraints, as opposed to Lean Manufacturing,
which attempts to eliminate waste at all
operations. Goal To make more profit. In
pursuing this goal, three key financial
performance measures - throughput, inventory, and
operating expense determine the level of these
financial measures.
3
Financial Measures
Throughput is the rate at which the manufacturing
firm sells finished goods. Note that throughput
under TOC is not synonymous with production rate.
Inventory is defined by OPT to be the money
the firm has invested in purchasing things which
it intends to sell. These include raw
materials, components, and finished goods that
have been bought by the firm but not yet sold. In
a departure from standard accounting practice,
labor and overhead are not included in the
inventory. Operating Expense is the cost of
converting inventory into throughput. It includes
direct and indirect labor, electricity, etc.
4
Constraints

• Internal Resources
• A bottleneck machine
• Market Related
• Demand level
• Policy Related
• A policy that allows no work on
• weekends.

5
A Procedure for Dealing with Constraints

• Identify the primary constraint.
• 2. Find out how to exploit the constraint.
• 3. Subordinate everything else to the decision
• made in step 2.
• 4. Elevate the constraint so that higher level of
  
• performance can be achieved.
• 5. If the constraint is eliminated, go back to
  step
• 1.

6
Example
Two products A and B
7
Example

• In the above table, both products are assumed to
  
• have the same amount of direct labor cost,
  and
• therefore ignores in estimation of
  contribution
• margin.
• Since the direct labor hours are the same, then
  in
• traditional overhead allocation based on the
  direct
• labor hours, each product will receive the
  same
• amount of overhead, and thus also ignored.
• The contribution margin (not necessarily profit)
  is
• simply the difference between the price and
  the cost
• of materials.

8
Example

• in order to satisfy the total demand we will
  need
• 200(.2)110(.5) 95 hours of the bottleneck
• machine.
• However, we have only 60 hours of machine time
• available. Therefore we must make a decision
  as to
• how much of each product to produce to
  maximize
• the contribution to the profit of these two
  products.
• A naïve planner will look at the unit
  contribution
• margins of the two products and conclude that
  we
• must produce as many of B as possible as the
• bottleneck machine permits.

9
Example
The following table is a result of this logic.
10
Example (Concept of TOC)

• The bottleneck is already at hand.
• How to exploit this constraint. Looking at the
• contribution of each product to per bottleneck
  
• hour used we get for A
  60/0.2hrs 300/hr
• for B 75/.5hrs
  150/hr.
• Therefore the firm should produce as many of A as
  possible.

11
Example (Concept of TOC)

• Subordinate everything else to the decision in
• 2. That is the remainder of the capacity,
  after
• producing 200 units of A, should be allocated
• to product B. this translates into 40 units of
  B,
• as shown in table below.

12
Example (Concept of TOC)
Clearly, this schedule has increased the total
contributions by (15,000 - 9,750)/9,750
54. The Linear Programming can also help us to
solve the same problem. Let X be the amount of A
produced and Y be the amount of B produced. LP
formulation Maximize 60X 75Y Subject to
X ? 200
Y ? 110 0.2X0.5Y ? 60
X, Y ? 0 The solution to this simple LP is X
200 and Y 40 which is the same as the TOC
solution.
13
Example (Concept of TOC)

• The last step of TOC procedure of elevating
• the constraint can be pursued in a multitude
• of ways, some of which are listed below.
• Reducing variability at and surrounding the
• constraint
• Increasing the capacity of the constraint
• Eliminating all idle times at the constraint
• Shifting some of the constraints workload to
• other resources

14
TEN RULES OF OPT

• Utilization and activation of a resource are not
  
• synonymous.
• The level of Utilization of a non-bottleneck is
  not
• determined by its own potential, but by some
• other constraint in the system.
• An hour lost at the bottleneck is an hour lost
  for
• the total system.
• An hour saved on a non-bottleneck is a mirage.
• The bottleneck governs the throughput and
• inventory in the system.

15
TEN RULES OF OPT

• The transfer batch size should not necessarily
• equal the production batch size.
• The production batch size should not be the
• same from stage to stage in the process.
• Capacity and priority should be considered
• simultaneously.
• Balance flow, not capacity.
• The sum of local optima is not equal to the
• optimum of the whole.

16
Resource Sequences

• A bottleneck is prior to a non-bottleneck. The
• non-bottleneck should run only when part are
• available from the bottleneck. (little or no
  choice
• of other action)
• A non-bottleneck is prior to a bottleneck. The
• non-bottleneck should be paced to match with
• the rate of the bottleneck. If it runs slower,
  the
• bottleneck will starve, if it runs fast, the
  queues
• will build up in front of the bottleneck
  station.

17
Resource Sequences

• A non-bottleneck and a bottleneck are just prior
  
• to an assembly stage. The non-bottleneck
  should
• be paced at the rate of the bottleneck.
  Otherwise,
• inventory will build up at the assembly point.
• Both a non-bottleneck and a bottleneck supply
• independent market demand. Both should
• produce as close to the market demand for
  their
• product as possible.

18
DRUM-BUFFER-ROPE SCHEDULING
To operationalize the ten rules under different
sequence of resources, Goldratt and Cox (1986)
suggested the DBR scheduling. In this method,
first the bottleneck machine is determined by
using a rough-cut capacity planning method to
find out which resource will be utilized the
most, and therefore, will potentially be the
bottleneck. This resource becomes the drummer in
the DBR technique. It is meticulously schedule to
insure the delivery of the finished products on
time as much as possible. Once its schedule is
finalized, the other resources are subordinated
to this schedule. In accomplishing this, the
upstream stations are scheduled to feed the
bottleneck resource just-in-time. In another
words, all upstream non-bottleneck stations
operate at the bottlenecks pace.
19
DRUM-BUFFER-ROPE SCHEDULING
A buffer is set before the bottleneck to ensure
that it always can maintain the pace of the drum,
even if a prior non-bottleneck slows down for
some reason. Non-bottleneck operations are
scheduled to keep the buffer at the appropriate
level using a hypothetical rope. Jobs are
released only at the rate of the bottleneck, but
far enough, in advance, so that a time buffer is
maintained before the bottleneck. Inventory does
not increase unnecessarily, and the bottleneck is
never starved. Jobs, therefore, are pulled from
the first production stage by the bottleneck.
After the bottleneck, jobs are pushed to the end
of the line as quickly as possible.