c


c   


 
   
c
 
 
c
 





















  



 cc
c
cc

c
  c


c c
 cc


!""#

THEORY OF CONSTRAINTS IN OPERATIONS

INTRODUCTION

In today's competitive global economic marketplace, manufacturers are struggling to squeeze out
5% to 7% operational cost reductions. The reality is startling. If a plant is not consistently
improving performance, it is in danger of closing. However, if a company is able to find a way to
increase throughput with the same or less resources, it may mean the difference between
continuing operations in developed countries or moving them to lower cost regions around the
world.

Twenty years after Eliyahu Goldratt first introduced the Theory of Constraints in his book p

, the manufacturing world is again experiencing another paradigm shift in thinking. Through
the synthesis of the Theory of Constraints continuous improvement efforts and ultimately,
performance improvement are no longer measured over periods of years, but weeks.

When one looks at the Theory of Constraints, the underlying principle emphasizes the
importance of identifying and eliminating bottlenecks (constraints) in the manufacturing process-
not only to increase productivity, but as a tool for measuring and controlling the flow of
materials. The only problem is how to actually identify these constraints.

However, the fundamental approach to lean manufacturing is to maximize the amount of net
good parts per shift by striving for true one-piece flow. Targeting balanced flow, buffers of
excess inventory between each point in the process are removed, making it immediately apparent
which process is underperforming. With this knowledge, a manager can dispatch resources to
address the situation, known as "go and see."

While both ideas are productive, the key to unlocking true value and performance improvement
is in merging the two approaches. Maximize your "go and see" efforts by focusing resources on
the true constraint; the result is a seamless flow of production that generates the highest possible
return. This simple idea in theory was a nearly impossible task in reality, until now.

$%&'()*%'&

A constraint is anything that prevents the system from achieving more of its goal. There are
many ways that constraints can show up, but a core principle within TOC is that there are not
tens or hundreds of constraints. There is at least one and at most a few in any given system.
Constraints can be internal or external to the system. An internal constraint is in evidence when
the market demands more from the system than it can deliver. If this is the case, then the focus of
the organization should be on discovering that constraint and following the five focusing steps to
open it up (and potentially remove it). An external constraint exists when the system can produce
more than the market will bear. If this is the case, then the organization should focus on
mechanisms to create more demand for its products or services.

Types of (internal) constraints

â? Equipment: The way equipment is currently used limits the ability of the system to produce
more salable goods / services.
â? People: Lack of skilled people limits the system. Mental models held by people can cause
behaviour that becomes a constraint.
â? Policy: A written or unwritten policy prevents the system from making more`?

By taking a constraint-based approach to maximizing the flow of product through the plant, you
end up with less data, but more critical, decision-enabling information directly from the plant
floor. This is accomplished by limiting and prioritizing data collection and identifying the key
chronic constraints that must be corrected to meet a specified throughput target. This allows
plant personnel to focus their efforts on corrections that will lead to the greatest improvements in
plant performance. The results are astounding.

Recently, the idea of synthesizing lean and the Theory of Constraints was utilized by a major
Tier One automotive supplier with two goals: to provide better clarity and focus by identifying
true constraints, and to spend less time collecting data and more time solving problems.
The theory of constraints was developed by Goldratt and Fox in the 1980¶s and has since
evolved. Goldratt developed a scheduling approach known as optimized production technology
(OPT) that used TOC principles. This method was coined "synchronous manufacturing" in 1984
and became the theory of constraints in 1987.

TOC is usually explained using the Five Focusing Steps. The purpose of these steps is to center a
manager¶s attention on the constraining resources²those inhibiting profit growth.

There are a number of assumptions underlying the theory of constraints. The first is that the goal
is to make money now and in the future. Making money is the primary reason for a company to
exist²otherwise it wouldn¶t be in business. The second assumption states that throughput is
used as a way to measure money. Throughput is defined as revenue minus the variable cost of
materials and energy. It is important to understand that with this assumption direct labor is not a
variable cost, it is a commitment made to the workers for the planning horizon. It is also assumed
that overhead costs won¶t vary in the short-run. However, a short-run planning horizon doesn¶t
allow for costs that can change in the long run.

The third assumption is that there is always at least one constraint on each product that limits the
company¶s revenue. This constraint could be internal or external to the organization. The fourth
assumption builds on this constraint²there are three types of resources: scarce bottleneck
resources, non-bottleneck resources, and capacity constraint resources (CCR). A capacity
constraint resource is one that is not a bottleneck but possesses the ability to become a constraint
if not properly managed. The fifth assumption is that most manufacturing operations have only a
few CCRs, and thus it is easy to control them.

The theory of constraints also assumes two characteristics about the production process:

Ä? Dependent events exist that result in interactions between resources and products.

Ä? Within every manufacturing environment statistical fluctuations and random events

occur.

These assumptions create a need for scheduling and prioritizing the product flow. The OPT
method of scheduling implies a rather short time horizon.
The eight assumption states that the OPT system is implicitly stable²at any given time
bottlenecks are identified, and the order mix is stable with respect to given resources. For the
short time period, capacity is fixed and bottlenecks are inevitable.

c&&+,-'*$%&$.'/0/0$(1$.$%&'()*%'&

·Throughput is defined as revenue minus the variable cost of materials and energy.
·There is always at least one constraint on each product that limits the firm¶s revenue.
·There are three types of resources: scarce bottleneck resources, non-bottleneck resources, and
capacity constraint resources (CCR).
·Most manufacturing operations have only a few CCRs, and thus it is easy to control them.
·Dependent events exist that result in interactions between resources and products. Within
every manufacturing environment, statistical fluctuations and random events occur.
·The optimized production technology system is implicitly stable²at any given time
bottlenecks are identified, and the order mix is stable with respect to given resources.

/0*20$3+&*%4 '0-&$.'/0/0$(1$.$%&'()*%'&

1. Identify the system¶s constraints.

2. Decide how to exploit the system¶s constraints.
3. Subordinate everything else to the decision made in Step 2.
4. Elevate the system¶s constraints.
5. Go back to Step 1 if the constraint is broken in Step 4. Watch for inertia.





  
   
c
 c  



The improvement process focuses on three questions: ë  

ë  


 
 

These three questions must be answered in sequence to make the
improvement process effective.

TOC TP was developed as a set of logical tools that enables people to tackle nonphysical
constraints. TOC TP consists of a set of six logical tools including Current Reality Tree
(CRT), Evaporating Cloud (EC)1, Future Reality Tree (FRT), Negative Branch Reservation
(NBR), Prerequisite Tree (PT), and Transition Tree (TT). Each of these six can be used together,
or as stand-alone tools.

The first step in the improvement process is to determine what to change. Improvement requires
change, but change does not always lead to improvement. Sometimes, change makes things
worse. Change results in improvement only when it focuses on the right element to change. If we
view a system as a chain composed of many links, the strength of a chain is determined by the
weakest link. When the weakest link is strengthened, the change is an improvement. When a
non-weakest link is strengthened, the change is not an improvement.

Current Reality Tree (CRT) is used to identify the core problem by revealing causal relationships
among the undesirable symptoms that an organization exhibits. Building a current reality tree
begins with constructing a list of µUndesirable Effects¶ (UDEs) that are dysfunctional symptoms
or behaviors. Once UDEs are identified, the second step is to seek causal relationships between
these UDEs and possible causes. Which UDE is the cause of which other UDE? To identify,
refine and audit the causal relationships, a set of rules, called the Categories of Legitimate
Reservation (CLR) is used to find out if the logic presented makes sense
.
The next step in the improvement process is to determine what to change to. Once the core
problem is identified, the development of the solution may simply be the elimination of the core
problem. However, eliminating the core problem can also be hampered by the existence of two
opposing forces pulling the decision makers in opposite directions, resulting in a tension or
conflict. The Evaporating Cloud (EC) is a tool that helps the decision makers search for a
solution by challenging the assumptions underlying the conflict.

The construction of the EC starts with a desired objective (A), such as the opposite of the core
problem identified in the current reality tree (CRT). Next is a determination of requirements (B
& C) and prerequisites (D & D¶). Requirements are the necessary conditions to achieve the
objective, while prerequisites are the necessary conditions for requirements. The necessary
conditions are verified using the µIN ORDER TO«WE MUST HAVE«¶ logic. In order to have
the objective A, we must have the requirements B and C. In order to have the requirement B, we
must have the prerequisite D. But in order to have the requirement C, we also must have the
prerequisite D¶. As the two prerequisites D and D¶ are in conflict, the objective A appears to be
unobtainable. The resolution of the conflict requires the hidden assumptions of the necessary
conditions be surfaced and challenged. A solution that invalidates any of the assumptions is
called an µinjection.

EC helps determine the initial thrust or primary injection needed to create a future system that
produces the desired effects. However, this primary injection is just the first step. To build a
robust solution that actually will work; other injections need to be added to ensure that the
primary injection achieves the desired results while not creating new, undesirable problems.
The Future Reality Tree (FRT) looks similar to the Current Reality Tree. While CRT is used to
trace the undesirable effects to the root causes using IF-THEN causal relationships, FRT is the
tool used in a similar fashion to CRT to show how the stated changes solve the problem without
becoming the source of future problems

The last step in the TOC improvement process is the implementation of the solution. Its success
depends on the degree of understanding and support participants in the improvement process
might have about the implementation of the change. Participants might have doubts because they
may perceive some critical obstacles that prevent the change from being implemented. The
Prerequisite Tree (PT) is a tool used to identify these obstacles and to establish a series of
intermediate objectives to overcome them. The last step in planning is a detailed action plan,
embedded in the Transition Tree (TT).
The TOC TP tools are clearly useful in identifying root causes and the primary injection for
change. They are also useful as communication tools to facilitate cooperation, collaboration and
co-ownership of participants in the improvement process. As a team works together in
employing these tools for the purpose of organizational problem-solving and continuous
improvement, the team develops strong process skills and mutual trust in the process. Team
members learn a common vocabulary and reflective process for communicating about
organizational design, conflict management, action planning and organizational learning.

/0$(1$.3$%&'()*%'&*% -0()'*$%&

5)%''1-0&

There are four primary types of plants in the TOC lexicon. Draw the flow of material from the
bottom of a page to the top, and you get the four types. They specify the general flow of
materials through a system, and they provide some hints about where to look for typical
problems. The four types can be combined in many ways in larger facilities.

_? I-Plant: Material flows in a sequence, such as in an assembly line. The primary work is
done in a straight sequence of events (one-to-one). The constraint is the slowest
operation.
_? A-Plant: The general flow of material is many-to-one, such as in a plant where many sub-
assemblies converge for a final assembly. The primary problem in A-plants is in
synchronizing the converging lines so that each supplies the final assembly point at the
right time.
_? V-Plant: The general flow of material is one-to-many, such as a plant that takes one raw
material and can make many final products. Classic examples are meat rendering plants
or a steel manufacturer. The primary problem in V-plants is "robbing" where one
operation (A) immediately after a diverging point "steals" materials meant for the other
operation (B). Once the material has been processed by A, it cannot come back and be
run through B without significant rework.
 _? T-Plant: The general flow is that of an I-Plant (or has multiple lines), which then splits
into many assemblies (many-to-many). Most manufactured parts are used in multiple
assemblies and nearly all assemblies use multiple parts. Customized devices, such as
computers, are good examples. T-plants suffer from both synchronization problems of A-
plants (parts aren't all available for an assembly) and the robbing problems of V-plants
(one assembly steals parts that could have been used in another).

(+,+..0(
$-0
?
Within manufacturing operations and operations management the solution seeks to pull materials
through the system, rather than push them into the system. The primary methodology use is
Drum-Buffer-Rope (DBR) and a variation called Simplified Drum-Buffer-Rope (S-DB)

Drum-Buffer-Rope is a manufacturing execution methodology, named for its three components.

The 4  is the physical constraint of the plant: the work center or machine or operation that
limits the ability of the entire system to produce more. The rest of the plant follows the beat of
the drum. They make sure the drum has work and that anything the drum has processed does not
get wasted.

The 
protects the drum, so that it always has work flowing to it. Buffers in DBR have time
as their unit of measure, rather than quantity of material. This makes the priority system operate
strictly based on the time an order is expected to be at the drum. Traditional DBR usually calls
for buffers at several points in the system: the constraint, synchronization points and at shipping.
S-DBR has a buffer at shipping and manages the flow of work across the drum through a load
planning mechanism.

The 
is the work release mechanism for the plant. Orders are released to the shop floor at
one "buffer time" before they are due. In other words, if the buffer is 5 days, the order is released
5 days before it is due at the constraint. Putting work into the system earlier than this buffer time
is likely to generate too-high work-in-process and slow down the entire system.

There are several principles behind Drum-Buffer-Rope; some key ones are:

1. In any set of resources, some will be more heavily loaded than others (on average). These are
often termed the Capacity Constraints, or CCR (capacity constrained resource). In some plants
there is really only the one; in others, there might be several; sometimes, it¶s not easy to identify
them without training in Theory of Constraints.

2. The most capacity constrained resource will dictate the rate the whole plant can flow work
from materials through to finished goods.

3. There is no value in any resource that feeds the constraint, producing at a faster rate than the
constraint. Common measurements might encourage this but the only outcomes are increased
WIP, increased lead times, probably more expediting, probably damage to on-time performance,
and probably damage to productivity as people waste time trying to sort out priorities and sort
through the WIP to find what they are supposed to be working on next. Plus floor space gets
consumed, cash is tied up, there¶s more chance of damage, etc.

4. The rate of production of resources that are fed by the constraint is dictated by the rate of
output of the constraint.

This lays the foundation for Drum-Buffer-Rope.

/0


First, a schedule (a finite schedule) is developed for the work that has to pass through the
capacity constraint. The nature of this can be as simple as,

1. Make 100 of part X

2. Then, make the 25 of part Y.

3. Then, etc.

Often the schedule also includes timing.

This schedule is called the DRUM because it¶s essentially the drum beat to which the whole
plant will be marching. It sets the production pace for the whole plant. (The focus and leverage
benefits of this are tremendous, of course).

06'/0
$-0

The Rope concept is extremely simple; it is that the only parts and materials that should be
released into the ³gating´ processes in the plant should be those specifically needed to support
the Drum schedule. If this seems obvious « it isn¶t. In many plants, work is released into the
gating operations simply because people are running out of work and there¶s a threat of resources
being idled. And in others, work is released in quantities that are completely disconnected from
what¶s actually needed « there is often some ³batching,´ justified for various reasons. This
might still occur with drum0-buffer-rope but it is less common, and more rigorously controlled.

The name ³Rope´ can be explained by the visualization of ³lassoing´ just exactly those parts
needed for each job on the Drum. In a drum-buffer-rope implementation it is common to
introduce ³codes´ such that if someone issues more material or parts into the system than the
Drum calls for, then an explanation has to be provided or else there will be an inquiry. This is
because over-issuing materials is often a symptom of a policy constraint i.e. a policy or
procedure or measurement that encourages ³keeping busy.´

/0+..0(

The buffers in drum-buffer-rope can be viewed in two dimensions ± time, and stock.

When the constraint resource is close to being a bottleneck, every minute wasted is a waste of the
whole plant¶s production capability. To get the most out of the plant, the employees have to get
the most out of the constraint.

One thing that threatens this is, starving the constraint ± preventing it from being fully activated
because there aren¶t any parts or materials ready for the CCR (capacity constrained resource) to
work on, which can happen for any of several possible reasons. A logical solution to this is to
make sure that there¶s always a judiciously sized buffer of work sitting right in front of the
constraint so that it is NEVER starved of work. This is a stock buffer.

However, it is created as a result of managing TIME, not STOCK.

Let¶s say, for example, that when parts and materials are released into the gating operations of
the plant, they take (on average) 20 hours to flow through the different processes until they reach
the CCR. This would mean that if we released the work 20 hours before the Drum schedule
called for it to be at the CCR « often it would not get there in time, because of the many types
of delays and disruptions that can happen ² set-ups take longer than planned, or a machine
breaks down, or someone is absent, or else there¶s a quality issue to be resolved, or else there¶s
some data missing from a drawing, etc.

So, what we do is, release the parts and materials (just for purposes of illustration) 28 hours
ahead of when they are needed at the CCR.

This way, if everything flows smoothly the work will arrive at the CCR 8 hours earlier than
needed « creating an 8-hour stock buffer. The real advantage of this becomes obvious when you
consider what happens when work DOESN¶T flow smoothly « when there are some disruptions
and delays, provided those delays don¶t accumulate to exceed 8 hours, the work will still arrive
early at the CCR and the stock buffer will still exist, and the CCR can stay highly productive and
on schedule (i.e. the Drum is on-schedule).

There are basically three types of buffer:

)-)3*'1$%&'()*%'+..0(this buffer indicates that some parts are needed to arrive earlier at
the constraint area. In fact, the total processing times of these parts that need to arrive earlier is
equal to the time buffer.
c&&0,751+..0(this type of buffer is needed when a bottleneck part is assembled with a non-
bottleneck part. In this case, non-bottleneck parts accumulated in the front of assembly station
indicates the buffer.
/*--*%4+..0(this buffer protects the due dates from disruptions on the way from the
constraint buffer to the shipping dock.
To summarize:

1. Identify the CCRs in the system (this corresponds to the first of the 5 Theory of Constraints
Focusing Steps; ³Identify the constraints.´

2. By examining the orders in the system, and considering the finite capacity of the CCR
resources, schedule the work in detail through the CCR. This is the DRUM. (This corresponds to
the second of the 5 Theory of Constraints Focusing Steps, ³Decide how to exploit the
constraints.´)

3. Choose a Buffer Time to add to the set-up and run time associated with flowing work through
the resources to the CCR. This of course is the BUFFER. This is also in line with TOC Focusing
Step 2.

4. Only release materials and parts into the gating processes of the plant in sync with the needs of
the Drum and the timing of the Buffer « i.e. release only the needed quantity of only the needed
parts and materials, and do so only when the Buffer time offset from the Drum calls for the
release. This is the ROPE mechanism.

There are many variations and refinements on this basic technique but as described above, it¶s
the key to generating a huge boost in productivity and excellent on-time delivery performance
while shrinking lead times and work in progress inventories.

*55+&'()'*%4/0$(1$.3$%&'()*%'&8*'/06),-50

Teaching the theory of constraints can be aided by a simple simulation exercise with people in
any training class: Take small five boxes partially filled with 15 small candies each. The
instructor hands the boxes one at a time to the first participant who empties the contents and then
hands the box and candies to the next participant. The second person counts the number of
candies, puts them in the box and hands it to the third person who returns the box the instructor.
The instructor varies the rates at which the five boxes of candies are handed to the first person in
the "value chain" and times are recorded for each member's part in the process.

As the boxes of candies are handed to the first

person slowly, it is easy to observe that all
participants have ample time to carry out their
activities. At the slow pace, all resources are
being starved of work and the bottleneck is
external to the process, i.e., the instructor. Now
consider what happens if the instructor speeds
up the process and hands over the boxes at a much faster rate - then, all three participants start to
work faster, with the person in the middle always busy as the other two participants wait for
him/her. In essence the bottleneck has now changed to Person B, and a process constraint or
bottleneck has been identified.

Consider for the sake of argument that it takes the following average times for the participants to
carry out their activities:

0(&$%c Time = 2 seconds (Throughput = 30 boxes/minute)

0(&$% Time = 20 seconds (Throughput = 3 boxes/minute) 9--$''50%03:
0(&$% Time = 5 seconds (Throughput = 12 boxes/minute)

The efficiency of the process has slowed down to the slowest resource, the output of the whole
process is three boxes a minute; and Person A and Person C are not contributing to the overall
efficiency. If they were to slow down, they would not have an adverse impact on the process,
since they will always have idle time. Such resources with extra capacity are non-bottleneck
resources.

$%&*;0(*%4'/0..03'$.
%20%'$(1

Now consider the impact of inventory on the process. If only one box is sent through the value
chain, the total time taken by the box to be serviced is the sum of the activities of all three
participants, i.e., 2 + 20 + 5 = 27 seconds. This is the minimum theoretical time that a box of
candies can be processed in. From a customer perspective it would take 27 seconds for their box
of candies to be delivered. However, if there are other boxes within this value chain, then these
boxes will be serviced first. The number of boxes within the system is sometimes referred to as
work in process (WIP) and the relationship is expressed through Little's Law:

 

p 
 p

Where Throughput is the throughput of the process and is the throughput of the slowest resource
(Person B) and Flow Time is the average time that a typical flow unit spends within the process
boundaries.

Now suppose there are a total of four boxes ahead of a particular customer's box, thus Inventory
= 5 and the formula is:



 
 p

On overage the packet spends 1.67 minutes within the value chain before a customer gets his box
back. While the value-added work on the packet is just 27 seconds, it spends a total of 100
seconds within the system boundaries. A 27 percent efficiency ratio (value-add time/total flow
time).

While this simple example has been used to demonstrate the theory of constraints, the
implications are clear for an operations manager who deals with processes. Suppose the three
persons in the example represent three business units working together to create a horizontal
value chain. If Business Unit A becomes more efficient (e.g., reducing its cycle time from 5
seconds to 2 seconds), it may actually be detrimental to the overall process. Yet quite often,
Business Unit A's improvements will be rewarded since there is a tendency to measure results
vertically in silos. Additionally sometimes the work that an entity does is non-value-added. In the
candies example, Person C was transporting the boxes and did not provide any value to the
overall process. A good operations manager will think in depth about these issues.


/0$(1$.3$%&'()*%'&*% 0(2*30$-0()'*$%&

In order to determine the impact of actions on the organization, Goldratt [1990] defines the
following measurements: throughput (T), inventory investment (I), and operating expenses (OE).
Throughput is the rate at which the system generates money through sales.

Inventory is the money the system invests in purchasing items needed to generate throughput,
except for labor and overhead. Operating expenses is all the money the system spends generating
throughput. These measurements have been used by the manufacturing industries, in which T, I,
and OE are easy to quantify. Hence, a meaningful goal for an organization is to increase return
on investment (ROI), which is defined as follows:

ROI = (T-OE)/I

According to Goldratt, the goal of every for-profit organization is to make money. This goal can
be achieved by increasing ROI, which can be accomplished by increasing T, decreasing OE, or
decreasing I.
One major difference between the TOC approaches to managing an industrial organization and
the conventional approaches is in the relative priority given to these three measurements. While
most managers consider all three measurements important, the conventional approaches tend to
regard operating expenses (cost) as the most important.
The TOC sets different priorities, and suggests that throughput should be at the top of the list,
followed by inventory and operating expenses. To improve, an organization should first make an
effort to increase throughput, then decrease inventory and decrease operating expenses. A
measure of how well the organization's products or services sell in the competitive marketplace
is throughput. Since throughput is the only external component influencing ROI, it should be
elevated as the most important measure. Once an organization has focused its improvement
efforts on increasing throughput, the constraints that adversely affect throughput and profitability
can be identified and eliminated.


/0c--5*3)'*$%$. *%)%:&

Throughput is the rate at which a banking system generates revenue for services provided in a
way consistent with goal. Throughput in banks can be generated by investing in such markets as
customer lending, institutional lending, real estate, and investment firms.

Moreover, banks generate money by offering a variety of services such as wire transfers, foreign
exchange, and cashier's checks.

Operating expenses include all the money the bank spends in the process of generating
throughput. These expenses include all direct and indirect expenses except for the cost of
obtaining money in the market.

Inventory investment is the amount of money spent by a bank to raise capital necessary to
generate throughput. Inventory investment consists of the principal amount and the interest, if
any, paid on deposits. It should be noted that in the banking business, both the primary input --
inventory -- and output consist entirely of money. Banks use the money obtained from depositors
and invest it ventures varying in the degree of risk. In contrast with manufacturing, banks do not
need to convert physical inventory of products into money through sales.

Hence, the length of the process of generating throughput in banks is much shorter than in
manufacturing. It is intuitive that banks that want to grow should have no difficulty investing
unlimited amounts of money obtained from depositors in the market, whereas manufacturers
typically face finite demand for their output. Therefore, because of the nature of their markets,
banks have a stronger incentive to increase both their input and output compared with
manufacturers.

On the input side, banks typically base the rate they offer their customers on the amount of the
deposit and the length of time to maturity. They pay no interest for traditional checking accounts
that allow instantaneous withdrawals up to the balance in the accounts.
Longer term deposits pay higher interest compared with shorter-term deposits. Similarly, larger
deposits pay higher interest compared with smaller deposits. As compared with the
manufacturing sector, the banking industry has to pay a higher price for acquiring large deposits
(i.e., a quantity premium as opposed to a quantity discount). On the output side, banks have to
manage their investment risk. If they invest large amounts in a single project (as opposed to
small amounts in multiple projects), they expect high returns because of increased exposure.

Customers conduct a variety of transactions, including direct payroll deposits, money transfers,
and loans. They typically prefer to conduct their business at one bank. The customers who
provide the bank with deposits that contribute to inputs often use the same bank to meet their
needs for loans, which directly affect the bank's output. Hence, a strong link exists between
banks and their customers, on both the input and the output side. A similar supplier-
manufacturer-customer relationship does not exist in the manufacturing sector.

Throughput can be enhanced by increasing the number of borrowers and investment projects. It
can also be enhanced by increasing both the number and amount of transactions. Operating
expenses can be reduced by making operations more efficient (e.g., through investments in
technology).
Inventory investment can be reduced by reducing the cost of borrowing. As was established
earlier, since banks offer low interest on short-term and small deposits, it is beneficial for banks
to attract a large number of small and short-term depositors which, in turn, will increase the
number of transactions.