Chapter 4

Manufacturing Strategy and

the Supply Chain

4.1 MANUFACTURING STRATEGY

If you don’t know where you're headed, you can't tell if you're moving in the right di-
rection. The company's strategic plan provides a view of where the company is headed
and a road map showing the path it intends to take to get there. The strategic plan de-
scribes the desired positioning of the company in the marketplace and the manner in which
it will strive to satisfy customers. Included are target measures for factors such as sales,
market share, inventory levels, quality, delivery time, and profitability. The strategic man-
ufacturing plan identifies product families that will be produced, production technologies
that will be used to produce those products, and general policies that will be pursued for
purchasing, production, and distribution.
Formed at the upper levels of the firm by senior management, the company's strate-
gic plan defines the frame into which lower-level production system decisions must fit.
The components of the plan defining corporate strategy for marketing, manufacturing, and
distribution must fit together. One company may choose to concentrate on marketing with
all manufacturing and distribution being subcontracted out to independent providers. An-
other company may concentrate on manufacturing and contract with independent sales-
people and advertising agencies for marketing. Whatever the choice, the central corporate
strategy should guide all lower-level strategic and tactical planning decisions. After set-
ting the manufacturing strategy, the company must ensure that subsequent decisions re-
inforce that strategy. The location of production facilities, the design of the distribution
network, and the inventory policies adopted must match the choice of geographical mar-
kets and speed of delivery. Employment, training, and benefits policies must match the
decision regarding capital or labor-intensive production and the importance of a knowl-
edgeable workforce. It would, for instance, be appropriate to build comfortable facilities
in a scenic setting and offer opportunities for company-financed graduate education if the
company decided that its competitive advantage relied on the creative ingenuity of its en:
gineering staff. In general, this reliance on strategic decisions as a guiding framework
should move down to the lowest level of planning, such as the salary and training sched- -
ule of an individual employee or the inventory policy for a specific raw material. In the
remainder of this section, we detail the components of manufacturing strategy and how
they fit together.
 4.1 Manufacturing Strategy — 85

Dimensions of Manufacturing Strategy

Manufacturing strategy has several dimensions, including the range of product lines pro-
duced, the geographical and economic markets served, the technology base used, the core
competencies emphasized, the degree of vertical integration pursued, and competitive po-
sitioning in the marketplace. Market opportunities, and unfilled niches in particular, offer
an important guide to forming these plans. The location of the company's products on the
spectrum, from product-based to technology-based, provides another clue. Product-based
companies normally operate in highly competitive consumer markets and may be referred
to as market-pulled companies. Exceptional service, low cost, and an increasing variety
of choices and features over time become necessities for competitive success. The prod-
uct is defined by the customer, and the producer must find the best technology for pro-
ducing, delivering, and supporting the product. On the other hand, the rechnology-push
companies are leaders in scientific research and offer unique products. As technological
innovations develop, the company may develop new types of products for which to apply
those technologies and be forced to convince the public of the value of those items. For
instance, a decade ago, few people realized that a full life would require a pager and cell
phone. Yet, instantaneous communication—to any one, at any time, from anywhere—will
soon be an expectation. The development of products is impelled by technology. Auto-
mated location and direction systems, made possible by global positioning systems, will
soon be standard for automobiles and hikers. Post-its, composite tennis racquets, and early
microwave ovens are examples of technology-push products. Once the technology is de-
veloped, we often see a proliferation of product versions built on the same platform. For
instance, note the variety of watches with the same quartz crystal movement, or automo-
biles with the same engine, brake system, or chassis design. Research in technology-driven
companies tends to focus on basic science, whereas companies in the consumer goods
market concentrate on industrial design activities, such as product aesthetics (look, feel,
and ease of use).

4.1.1.1 Core Competencies

One aspect of manufacturing strategy relates to defining the core competencies of the
company. Core competencies are best described as a “mission” or goal that will provide
an advantage in the marketplace. We may be tempted to view a specialized technology
protected by patents or an effective marketing network of sales representatives as a core
competency. It is best, however, to define core competencies along more general lines that
will help direct strategic planning decisions to maintain this competency over time. One
approach uses the measures of competitiveness, which are discussed in Chapter 1: price,
features, service, delivery, and quality. Motorola became widely recognized in the 1980s
for its commitment to quality in its Six Sigma program. Feeling the need to compete with
the perception of high-quality Japanese electronics, the Six Sigma program strived to re-
duce the percent of defective products produced to no more than 3 per million. A com-
pany that claims rapid delivery as a core competency would need to plan an efficient dis-
tribution system. The production planning and control system would need timely data on
inventory levels and demand rates, as well as the ability to change production plans on
short notice to incorporate updates in the data.
Tn addition, concepts such as a highly skilled workforce can constitute the core com-
petency. Often, core competencies can be identified in a company's advertisement. For
example, a telecommunications company will guarantee a low rate or clear signal.
86 Chapter4 - Manufacturing Strategy and the Supply Chain

McDonald's can consider assured availability of a consistent quality product as a core

competency. Defining the core competency as being the technology leader will promote
continual research and development. On the other hand, calling a current state-of-the-art
manufacturing process a core competency may lead the company to rely too heavily on
that technology instead.of continuing to develop new and improved approaches to ac-
complish the same customer-valued objective. Soon, competitors will imitate, and then
surpass, that technology, leaving the company without a market. Kuglin [1998] empha-
sizes that a core competency must:
* Facilitate access to markets;
* Be perceived by customers as adding significant value to products; and
* Be difficult to imitate, thus providing a barrier to competitors.

It is important to distinguish between the enterprise’s current competencies and the

core competencies necessary for market success. Surveys of suppliers and customers can
be useful in defining the important market competencies.

4. 2 Customer Markets and Distribution

Another aspect of strategy relates to the specific markets to serve. Markets in this sense
are defined by the geographical location and the expectations of the customer. The workd
has become a global marketplace. Later in this chapter, we discuss the construction a
global supply-production-distribution networks to efficiently serve geographical market
We need to be concerned with product differences required for different cultures and
cal regulations on importing and recycling products, as well as the logistical details of
tually delivering the finished product to the customer. Such details can affect where
ucts should be produced and which countries actually offer profitable opportunities %
all factors are considered. Customer expectations determine design, delivery, and ses
requirements for competitiveness. Markets can be segmented, based on the level of 3
uct features expected. This is sometimes referred to as quality-of-product design.
e
$40,000 luxury sedans and $10,000 entry-level vehicles have markets around the
but the customers are different individuals with different requirements. The more
sive vehicle would be expected to have more power, a quieter ride, and impeccabe
§
ability. In addition, combined geographical cultural factors must be addressed. The
luxury sedan in northern Europe may not need as clean an engine or as effici
conditioning system as the $10,000 vehicle destined for the arid Southwesterr:
States. Market surveys and interviews with focus groups of potential custome=x
in &
helpful in identifying the functional and aesthetic expectations of customers
bination of a geographical and economic submarket.

4.1.1.3 Vertical Integration

supply
Many companies resort to vertical integration to ensure a reliable
owning two or more of the produc
Vertical integration refers to a company
dritli
from basic raw materials through finished goods delivery. When Texaco
and then transport s the gasoline throu:
ships the crude to its own refinery,
to a neighborhood gas station owned by the company, thatis considered vertica
sells i
Likewise, when Coca-Cola manufactures the syrup, then bottles and
too, is vertical integrati on. The compaz
pany-owned vending machine, that,
raw materials , and can also assume 5
the quality, price, and availability of
 4.1 Manufacturing Strategy 87

maintaining a market for the finished product. The vertically integrated company is less
susceptible to price wars or competition that may squeeze out the profits at one level of
the product hierarchy. Vertical integration also relates to the company's supply chain man-
agement strategy, the manner in which the company partners with suppliers and customers,
and manages the flow of items and information from basic materials to delivered prod-
ucts. This topic will be examined in more detail later in this chapter.

4.1.14 The Level of Flexibility

The ease and extent to which a facility can adapt to change is referred to as flexibility.
Flexibility comes in many varieties. Both external (to the firm) and internal variability
create the need for flexibility. External factors include changing levels of product demand
or changes in the proportional mix of products demanded. This includes changes in fea-
tures, quality, and prices expected by customers to stay competitive. Internal factors in-
clude machine breakdowns and normal engineering design changes to improve product
design and manufacturability. These occurrences generate the need for alternative ways
to manufacture the product. Although many types of flexibility have been discussed in the
literature (see for instance, Browne et al. [1984] and Sethi and Sethi [1990]), no stan-
dardized definition or measures have yet been adopted by industry. We can, however, di-
vide the types into strategic and operational categories. Strategic flexibility offers the po-
tential to adjust to unpredictable and dynamically changing market opportunities.
Operational flexibility allows parts to be re-routed in the face of minor demand changes,
i.e., machine breakdowns or temporary scheduling bottlenecks. Strategic flexibility en-
compasses volume flexibility, expansion flexibility, and product flexibility. Volume flexi-
bility measures the ability of the production system to operate profitably at multiple lev-
els of total demand. A high-volume flexibility system is one with a low break-even point.'
Expansion flexibility measures the possibility of adding capacity to meet more significant
increases in the level of total demand. Product flexibility relates to the variety of product
types that can be produced by the system. Can new products be made without major dis-
ruption to operations and finances? This includes the ability to satisfy customer demands
for new versions of existing products and generations of new products. Operational flex-
ibility includes part-mix flexibility, routing flexibility, operation flexibility, machine flexi-
bility, and material handling flexibility. The ability of the system to produce products in
different proportions without major tooling changeovers or reconfiguration of equipment,
measures part-mix flexibility. Flexible machines that can be used to make a variety of
similar parts or be overhauled to produce new product families, increase part-mix flexi-
bility. The presence of duplicate machines (sometimes referred to as parallel processors)
enhances routing flexibility that is measured by the number of routes each product can
take through the production facility. This is related to operation flexibility that describes
the number of different process plans and processing sequences that can be used to man-
ufacture a part. Options for using a sequence of special purpose machines in lieu of one
machining center, or purchasing product modules instead of producing internally, impact
operation flexibility. Machine flexibility supports operation flexibility. Although operation
and routing flexibility focus on the product, machine flexibility focuses on the process. If
a machine can be adjusted to perform quickly more than one operation, it has machine
flexibility. Material handling flexibility refers to the system's ability to transport varying

'Break-even point refers to the level of demand at which the company can cover all costs. We assume the
company will be profitable if demand exceeds the break-even point.
88 Chapter4 Manufacturing Strategy and the Supply Chain

item and unit loads, the frequencies with which loads can be moved, and the ability to
vary pick-up and delivery points. As well as allowing product mix and routing changes,
a flexible material handling system allows frequent modification of the facility layout to
adjust to changing market demands while maintaining efficiency.
The firm can acquire flexibility through several strategies. Purchasing excess capac-
ity permits more rapid adaptation to demand changes. Excess capacity can be acquired
by employing more workers or installing more equipment. Other strategies may be more
cost effective, however. Cross-training workers and purchasing machines with flexible op-
eration capabilities allow the company to quickly adjust to changing product mixes and
machine breakdowns. In general, people tend to provide the most flexibility, but the flex-
ibility of intelligent, automated systems is constantly improving. For instance, free-
roaming automated guided vehicles and bar-coded transport containers and parts allow for
flexible material handling. Information can also be a strategic factor in flexibility. Accu-
rate market forecasts allow re-engineering of products in an effective time frame and the
chance to develop efficient schedules for resource utilization. Direct reports of customer
purchases allow production plans to be adjusted to meet market conditions. Knowledge
of the current status of equipment allows rapid replanning of order releases and process
routes to avoid congestion.

4.1.2 Supporting Decisions

The company makes a variety of planning decisions. Each is an opportunity to support its
manufacturing strategy. In this section, we describe several important decision types.

4.1.2.1 Make versus Buy Decision

Deciding which parts to make and which to buy (purchase from vendors) is often an im-
portant decision for the firm. This decision relates to the choice of core competencies and
the customer's values. Most companies are unwilling to relinquish control of key parts to
vendors. Commodities, parts that are available from many vendors at competitive prices,
and normally with quick delivery, are seldom produced by the manufacturer of the final
product. It is often cheaper to purchase these items, and there is ordinarily little risk in
doing so. On the other hand, parts that directly relate to the core competencies of the com-
pany are usually produced internally.
A simple economic trade-off model can sometimes help clarify the make versus buy
decision. Suppose a part can be purchased from a vendor at a cost of $c per unit in-
cluding relevant ordering and receiving costs. If we decide to make the item internally,
we must invest $C in a new plant and equipment, but we can make the units at a variable
production cost of $c, per unit. The corresponding cost curves, as a function of total pro-
duction quantity X, are shown in Figure 4.1. Algebraically, we see that we should produce
the item if:
CHeX<c-X (4.1)

or, we produce if the expected production quantity satisfies?:

Cc (4.2)
X> —
C1
— e

2We assume ¢, > c,, otherwise there is no economic justification for internal production. Other strategic fac-
tors may still favor intemal production however.
 4.1 Manufacturing Strategy 87

maintaining a market for the finished product. The vertically integrated company is less
susceptible to price wars or competition that may squeeze out the profits at one level of
the product hierarchy. Vertical integration also relates to the company's supply chain man-
agement strategy, the manner in which the company partners with suppliers and customers,
and manages the flow of items and information from basic materials to delivered prod-
ucts. This topic will be examined in more detail later in this chapter.

4.1.1.4 The Level of Flexibility

The ease and extent to which a facility can adapt to change is referred to as flexibility.
Flexibility comes in many varieties. Both external (to the firm) and internal variability
create the need for flexibility. External factors include changing levels of product demand
or changes in the proportional mix of products demanded. This includes changes in fea-
tures, quality, and prices expected by customers to stay competitive. Internal factors in-
clude machine breakdowns and normal engineering design changes to improve product
design and manufacturability. These occurrences generate the need for alternative ways
to manufacture the product. Although many types of flexibility have been discussed in the
literature (see for instance, Browne et al. [1984] and Sethi and Sethi [1990]), no stan-
dardized definition or measures have yet been adopted by industry. We can, however, di-
vide the types into strategic and operational categories. Strategic flexibility offers the po-
tential to adjust to unpredictable and dynamically changing market opportunities.
Operational flexibility allows parts to be re-routed in the face of minor demand changes,
ie., machine breakdowns or temporary scheduling bottlenecks. Strategic flexibility en-
compasses volume flexibility, expansion flexibility, and product flexibility. Volume flexi-
bility measures the ability of the production system to operate profitably at multiple le
els of total demand. A high-volume flexibility system is one with a low break-even point.'
Expansion flexibility measures the possibility of adding capacity to meet more significant
increases in the level of total demand. Product flexibility relates to the varicty of product
types that can be produced by the system. Can new products be made without major dis-
ruption to operations and finances? This includes the ability to satisfy customer demands
for new versions of existing products and generations of new products. Operational flex-
ibility includes part-mix flexibility, routing flexibility, operation flexibility, machine flexi-
bility, and material handling flexibility. The ability of the system to produce products in
different proportions without major tooling changeovers or reconfiguration of equipment,
measures part-mix flexibility. Flexible machines that can be used to make a variety of
similar parts or be overhauled to produce new product families, increase part-mix flexi-
bility. The presence of duplicate machines (sometimes referred to as parallel processors)
enhances routing flexibility that is measured by the number of routes each product can
take through the production facility. This is related to operation flexibility that describes
the number of different process plans and processing sequences that can be used to man-
ufacture a part. Options for using a sequence of special purpose machines in lieu of one
machining center, or purchasing product modules instead of producing internally, impact
operation flexibility. Machine flexibility supports operation flexibility. Although operation
and routing flexibility focus on the product, machine flexibility focuses on the process. If
a machine can be adjusted to perform quickly more than one operation, it has machine
flexibility. Material handling flexibility refers to the system's ability to transport varying

'Break-even point refers to the level of demand at which the company can cover all costs. We assume the
company will be profitable if demand exceeds the break-even point.
88 Chapter4 Manufacturing Strategy and the Supply Chain

and the ability to

item and unit loads, the frequencies with which loads can be moved,
product mix and routing changes,
vary pick-up and delivery points. As well as allowing
tion of the facility layout to
a flexible material handling system allows frequent modifica
efficienc y.
adjust to changing market demands while maintaining
ng excess capac-
The firm can acquire flexibility through several strategies. Purchasi
capacity can be acquired
ity permits more rapid adaptation to demand changes. Excess
Other strategie s may be more
by employing more workers or installing more equipment.
ng machines with flexible op-
cost effective, however. Cross-training workers and purchasi
changing product mixes and
eration capabilities allow the company to quickly adjust to
flexibilit y, but the flex-
machine breakdowns. In general, people tend to provide the most
g. For instance, free-
ibility of intelligent, automated systems is constantly improvin
container s and parts allow for
roaming automated guided vehicles and bar-coded transport
factor in flexibilit y. Accu-
flexible material handling. Information can also be a strategic
an effective time frame and the
rate market forecasts allow re-engineering of products in
n. Direct reports of customer
chance to develop efficient schedules for resource utilizatio
market conditions. Knowledge
purchases allow production plans to be adjusted to meet
of order releases and process
of the current status of equipment allows rapid replanning
routes to avoid congestion.

4.1.2 Supporting Decisions

an opportunity to support its —
The company makes a variety of planning decisions. Each is -
several important decision types.
manufacturing strategy. In this section, we describe

4.1.2.1 Make versus Buy Decision

se from vendors) is often an im-
Deciding which parts to make and which to buy (purcha
of core competencies and
portant decision for the firm. This decision relates to the choice
to relinqu ish control of key parts .
the customer's values. Most companies are unwilling
many vendors at competitive prices.
vendors. Commodities, parts that are available from
ed by the manufa cturer of the fina:
and normally with quick delivery, are seldom produc
and there is ordinar ily little risk 1f
product. It is often cheaper to purchase these items,
relate to the core compet encies of the ¢
doing so. On the other hand, parts that directly
pany are usually produced internally.
help clarify the make versus
A simple economic trade-off model can sometimes
a vendor at a cost of $c¡ per unit “-
decision. Suppose a part can be purchased from
Tf we decide to make the item internal *
cluding relevant ordering and receiving costs.
nt, but we can make the units at a vari:
we must invest $C in a new plant and equipme
g cost curves, as a function of total
production cost of $c, per unit. The correspondin
ically, we see that we should proci:
duction quantity X, are shown in Figure 4.1. Algebra
the item if:
C+e,-X<a-X
satisfies”:
or, we produce if the expected production quantity

for internal production. Other strates

2We assume ¢, > ¢y, otherwise there is no economic justification
tors may still favor internal production however.
 4.1 Manufacturing Strategy 89

Break-even Point
Production Quantity
Figure 4.1 The Make versus Buy Tradeoff

EXAMPLE41 A new bracket has been designed to correct a stability problem identified with a communication
device when used in an environment subject to high vibration. The company has the machine ca-
pacity to produce the new bracket but estimates a present worth cost of $12,500 to obtain the tool-
ing needed to produce the bracket. An additional worker would have to be hired to support internal
production. The hiring and training cost is estimated at $1,000. Combining raw material and labor
costs with estimated standard hours for production, the estimated internal variable production cost
is $1.12 per unit. A vendor has quoted a price of $1.55 per unit. Forecasted demand is 10,000 units
per year for the next two years. The company believes the vendor can meet the delivery and qual-
ity standard. Should the bracket be made internally or purchased?

SGLUTION The breakeven point is found from Equation (4.2) as

X= 12,500
ss + -1j
1,000 —-- 31395

Since lifetime demand is expected to be only 20,000 units, we should purchase the bracket.

of course, this model is rather simplistic. A more complete model might include un-
certainty in demand, price breaks if we buy in large quantities, supplier limitations, mul-
tiple technology and stair step capacity costs, nonlinear variable production costs, and
risk-based decision criteria. We also need to worry about control issues. Can we really
count on the vendor to deliver the requested number of units on time? Will product qual-
ity be dependable? Will the price remain stable? Could we decrease our production costs
over time through process improvements? Can we obtain the necessary capital to buy the
equipment needed to make this product? Will in-house production strain equipment and
human resources? These and other questions should be addressed as part of the make ver-
sus buy decision.

4.1.2.2 Make-to-Stock versus Make-to-Order

Another basic decision concerns whether we should anticipate demand for a product
and store it in inventory, or wait and produce it only after it has been ordered by a cus-
90 Chapter4 Manufacturing Strategy and the Supply Chain

tomer. Make-to-Stock items are kept in inventory so that they can be shipped to cus- s
tomers as soon as the customer order arrives. This quick delivery can be a competitive
advantage and allows us to plan production ahead of time. On the other hand, we must
incur inventory costs for carrying products while we wait for demand to occur, and we
run the risk of product obsolescence when we err in forecasting demand. In general,
high-volume products are stocked. Raw materials may also be stocked to save on or-
dering costs and to reduce production lead time. Some key spare parts may also be
stockpiled to reduce repair times for plant equipment. Items not stockpiled are produced
only when ordered (or ordered only when needed for planned production). This would
not be practical for items with high setup costs and frequent small customer orders, but
is often the best approach when customers are willing to wait for product delivery, and
when orders occur with low frequency. These are termed Make-to-Order items. Although
strategic factors, such as the delivery lead time demanded by customers, often domi-
nate the justification for whether an item should be stocked, we can model the economic
aspects of this decision.

Assumptions:
1. Unit variable costs and setup/ordering costs are constant and equal for the stock-
ing and nonstocking cases.
. The demand rate is constant and deterministic.
D

. A backorder cost of 7 is incurred each time we need to order under no stocking.

p

This includes goodwill loss due to slower delivery of the final product to the cus-
tomer.
. The cost of monitoring inventory levels under the stocking option costs C, per
>

time.
Let O, be the average order quantity under the nonstocking option, D the annual
demand rate for the item, and A the setup/ordering cost. If we do not stock the item, we

will have orders per time period. The cost of ordering per period, including back-
s
(A + 7) -D
orders, will be . If we do stock the item, we know from the EOQ model of
Qns
Chapter 2 that the cost of setup/ordering, plus inventory holding cost, will be V 2ADh per
time where 7 is the holding cost per unit, per time. Additionally, we have the inventory
monitoring cost, C,. Thus, our decision is to stock the item if:

\2ADh
+ ¢, < 4t m-D
ns
Assuming the make-to-order cost exceeds C,, we make-to-stock if:

((A+w)<D‘C)2
N —Qu 4.
2:A:D 3
Otherwise, we make-to-order.

A raw material normally used in batches of 30 units has annual demand of 150 units. The ordering
cost is $145. The annualized cost of adding a part number to the automated inventory tracking sys-
tem is estimated at $400. The cost per order of delayed product delivery is estimated at $100. The
cost to hold a part for one year is $2.00. Should we stock the item?
 4.1 Manufacturing Strategy 91

SOLUTION The condition of Equation (4.3) becomes

[(A +Qw) . D Cl 2 [(2453):)150) _ 400]2 ,

< 24D T aaa --

Since the actual holding cost of $2.00 is well under the threshold and the term in brackets is posi-
sed tive, we should stock the item. Using the EOQ, we place orders for:
ald
2= 24D
/ 2(145)(150) =
7 .
out W7~ 150 uniss

4.1.2.3 Selection of Technologies and Equipment

A choice of processing technologies exists for most products. Spot-welding can be per-
formed by manual labor or robots; parts can be moved by people pushing carts, drivers
on forklifts, or automated guided vehicles. The level of capital investment versus manual
labor used forms a strategic decision. Even within a level of capital investment, choices
exist between intelligent, flexible technology, and special purpose processors. Do we pur-
chase a computer-controlled machining center that can be programmed and tooled to
produce many machining operations or design a machine to efficiently punch one partic-
ular sized hole in one particular material? Economics and the degree of certainty con-
cerning future demand certainly play a key role here. But corporate image and recruiting
er are also important. The most talented employees may be attracted to companies that pur-
sue new technological innovations. Customers may prefer the company with the high-tech
image and showpiece factory.
Let's assume we have set the scale of the facility, and we now must select the tech-
nology to be used to produce the selected production volume. By technology, we may
mean a particular machine, a process requiring several related machines that produce a
feature or product, or an entire manufacturing facility. This will depend on the applica-
tion being modeled. Let D; be the selected production level of product hI=1,...n. I
products require multiple operations, we will expand the list j to include each major op-
eration of each product. Let i, i = 1, ..., m be the technologies available for performing
one or more of the operations required. Technologies may be specific types of machines
or different strategies. Let C; be the fixed cost per period, per unit of technology i that is
acquired. Fixed cost refers to the amortized cost of purchasing and installing the machine,
and renting the space it occupies. The variable cost of using a unit of technology i to pro-
duce product j, is v;. Variable costs include factors such as having a worker load and un-
load and, if necessary, operate the machine, and the cost of utilities, and other support
functions while the machine is running. We will denote the productivity factors by a;;, the
number of units of product (operation) j that can be produced per unit of technology i.
Our decision variables are:

Decision Variables:
X;; the number of units of technology ¡ used to produce j; and
) the integer number of units of technology i acquired.
A mathematical statement of the problem of selecting the optimal set of technologies for
this plan is:
92 Chapter4 Manufacturing Strategy and the Supply Chain

Minimize 3, C¡* y; + 3 3, Yy " Xy

=1 i=1j=1
@4
subject to:
Ea
S ay y=D Vi @3
i
> =y Vi “6
=
i integer; x; = 0 (4.7)
The model will select the technologies to be acquired and how they will be used to pro-
duce the assorted products. The objective function (4.4) minimizes the sum of fixed ac-
quisition plus variable processing costs. Constraints (4.5) ensure that the amount of tech-
nologies allocated to each product is sufficient to meet demand. Each a;; - x;; tells us the
number of units of product j made using technology i. This is summed over all technolo-
gies to determine total production of j. For each technology type i, Equation (4.6) requires
us to purchase the number of units of technology i that are used to produce all of the prod-
ucts. Finally, Equation (4.7) forces us to purchase an integral number of technology units
and register the basic non-negativity restrictions on production.
The mathematical program (4.4) to (4.7) can be difficult to solve for problems with
many products (operations) and technological choices. In most cases, the presence of in-
teger variables, such as the y;'s, makes an optimization problem difficult. A variety of
heuristic approaches exist for finding good solutions. For instance, we could ignore the
integer restrictions on the y;, solve the resulting linear problem, and then round up the y;
to the next largest integer. By relaxing the integrality restrictions, constraints (4.6) can be
shown to satisfy E x; = y; for all i. This allows us to write the objective function in terms
J
of just x;; terms and the solution will be to set:
C; + yi) - D,
X= i = arg min; (—¡)——L. (4.8)
a'./
Ay = otherwise

To explain (4.8), first note that if we use technology ¡ for making all of product j, we will
D, -
need x; = — units of i. The heuristic assumes we can buy only as much of technology
ij
i as necessary, including a fractional amount if desired. It then selects the cheapest tech-
nology for each product considering both variable and apportioned fixed costs. The reader
should consult a general optimization text for suggestions on finding optimal solutions.

A facility is being constructed to produce two product families for sale in Eastern Europe. The
choice of technologies is between a special purpose process for each family or purchasing some
flexible capability that can produce either product family. Period demand is 100 units for family
one and 200 units for family two. Additional data are as follows:
i Ci Vit v ay — aD
1 (Flexible) — $150 $100 $100 60 60
2 (Family 1) - $100 $100 infinite 75 0
3 (Family 2) $100 — infinite $100 0 75

Select a good technological solution for manufacturing these product families.

 4.1 Manufacturing Strategy 93

SOLUTION The two choices for product family 1 are the flexible process with cost:
4)
(Cy + Di — (150+ 100)(100)
= $416.7
ay 60 s

or the dedicated process with cost:

(G + v)D: _ (200)100) $266.7.

Ay 75

The dedicated process is better and we set:

D, 100
* ==
ay 75 =1.
The two choices for product family 2 are the flexible process with cost:

(C + v12)
: Dr
= $833.3
a12
or the dedicated process with cost $533.3. The dedicated process is better and we set x3, = 2.7. In
actuality, we must buy integral amounts of capacity, however. Therefore, our real cost includes
2 units of the dedicated process for family 1 and 3 units of the dedicated process for family 2. The
fixed cost is, therefore, 2(100) + 3(100) = 500. The variable production cost for family 1 will be

P 75 $133.3

and
vi2 Dy — 100(200)
= $266.7
ay 75 *
for family 2. Thus, actual total cost will be

3) $500 + $133.3 + $266.7 = $900,

as we sety; = 2 and y, = 3.
We now have a good equipment acquisition plan, assuming the given data are correct. If, in
reality, there is some uncertainty in the actual demands, it may be preferable to acquire some flex-
ible capacity. Then, once we determine actual demand, we can allocate the flexible machines to the
products as needed to meet demand. Another option we should consider is purchasing 1 unit of the
dedicated technology for product family one, 2 units of dedicated capacity for product family two,
and then purchasing flexible machines to manufacture the remaining demand for each family.

4.1.2.4 Justification of New Systems and Technologies

The competition does not stand still. If you want to be competitive, you must improve over
time. Strategic planning and investment decisions must reflect this reality. Unfortunately,
traditional discounted cash flow (DCF) techniques, such as net present value and rate of
return analyses, are inappropriate for evaluating complex strategic investment decisions.
The DCF approaches compare changes against a do-nothing alternative that assumes a static
world. The traditional approaches overlook factors that cannot be quantified by dollars and
have difficulty accounting for risk. It is very difficult to put a dollar value on improved
customer appreciation, resulting from faster response to inquiries as we install a new or-
der tracking system, or improved employee morale after being offered the opportunity to
learn and use the newest process technology. Nevertheless, we should strive to include these
intangibles into the justification of new information systems and processing technologies.
94 Chapier4 - Manufacturing Strategy and the Supply Chain

Kaplan [1986] notes that DCF approaches lead to nonoptimal facilities with obsolete process
technology. Fortunately, alternative approaches exist. Uncertainty can be modeled by treat-
ing cash flows as random variables and estimating the distribution of net present value. The
value of flexibility can then be evaluated by considering the distribution of possible cus-
tomer demand, market price curves, discounting sales revenue, and operating expenses un-
der each possible scenario. Intangibles such as the impact of a new system on customer
appreciation and employee morale can be considered by defining qualitative benefits for
each intangible factor under each alternative and computing a net present qualitative flow
value. For instance, suppose we are using the interest rate “7” in our net present worth com-
putations, and the planning horizon is 7 periods. Define Quy as the qualitative flow value
in period + for intangible attribute K, if investment alternative j is adopted. Intangible at-
tributes include issues, such as the types of flexibility and compatibility with core compe-
tencies, and additional factors such as public relations, employee morale, and product man-
ufacturability, discussed carlier in this chapter. Financial risk could also be treated as an
intangible, if desired. Qualitative flow values can be thought of as intangible or surrogate
revenue. Higher values imply a better score. The discounted net present qualitative flow
(NPQF) value of investment alternative j with respect to attribute % is then:
T

NPQFy =3, Qu;- (1 + 1), (4.9)

=
Once the net present qualitative value is computed, multiobjective decision techniques can
be used to compare alternatives based on monetary and intangible benefits. For compar-
ison purposes, qualitative flows can be measured between +1 (ideal solution) and —1
(worst solution), and each attribute can be assigned a relative weight.

A company that prides itself on its quick delivery is considering which of two packaging systems
to use—manual or automated. The automated system will cost $300,000 to purchase plus $50,000
per year to operate. The key attributes are cost, market position, and manufacturing flexibility. Once
installed, the automated system is expected to provide a public relations benefit when trying to at-
tract new customers because lower error rates and quicker delivery times can be promised. The man-
ual system may be a liability when customer expectations change and our competitors adapt in the
next several years. The proposed system will automatically identify parts and merge stored product
into orders for shipping. Some training is required to perform the tasks manually. Workers have re-
stricted capability and must be retrained as product lines change. The automated system is expected
1o be more flexible to product mix changes. Cash flows and qualitative rankings of the alternatives
for the next three years are given in the following table. After some discussion, the company de-
cided that each attribute can be considered of equal importance when measured between + 1. A
$0 cost system wóuld be a + 1, and a three-year present worth cost of$1 million would be the max-
imum possible investment and be rated as a —1.

Year Attribute Manual Automated

1 Cost $125,000 $350,000

2 Cost $130,000 $50,000
3 Cost $145,000 $50,000
1 Public Relations 0.0 -3
2 Public Relations —0.20 0.5
3 Public Relations —05 04
1 Flexibility 03 04
2 Flexibility 02 0.5
3 Flexibility 02 05
 41 Manufacturing Strategy — 93
SOLUTION The two choices for product family 1 are the flexibl
e process with cost:

(€ + vi)D; _ (150+ 100)(100)

ay = $416.7
60 $
or the dedicated process with cost:

(G + D2 _ (200)(100)
1 = $266.7.
d 75
The dedicated process is better and we set:

Di 100
M == T =113
The two choices for product family 2 are the flexible process
with cost:

= $833.3
a
or the dedicated process with cost $533.3. The dedica
actuality, we must buy integral amounts of
ted proces
s is better and we set X3 = 2.7. n
Capacity, however. Therefore, our real cost
2 units of the dedicated process for family 1 and 3 units of the dedicated includes
fixed cost is, therefore, 2(100) + 3(100) process for family 2. The
= 500. The variable production cost for
family 1 will be
v2 Dy _ 100(100) -
any 75 $133.3
3.
and

y32 Dy - 100(200) = 82667

axn 75
for family 2. Thus, actual total cost will
be

$500 + $133.3 + $266.7 = $900,

as we set y, = 2 andy, =3,
We now have a good equipment acquisitio
will n plan, assuming the given data are correc
reality, there is some uncertainty in the actual t. H, in
demands, it may be preferable to acquire
some flex-
ogy
sch-
ider
»ns.

4.1.2.4 Justification of New Systems and Techn

The ologies
»me The competition does not stand still. IF
you want to be competitive, you must
nily time. Strategic planning and investment improve over
decisions must reflect this reality. Unfor
traditional discounted cash flow (DCF tunately,
) techniques, such as net present value
return analyses, are inappropriate for and rate of
evaluating complex strategic investment
The DCF approaches compare changes decisions.
against a do-nothing alternative that assum
world. The traditional approaches overl es a static
ook factors that cannot be quantified
have difficulty accounting for risk. It by dollars and
is very difficult to put a dollar value
customer appreciation, resulting from on improved
faster response to inquiries as we instal
der tracking system, or improved empl l a new or-
oyee morale after being offered the
learn and use the newest process techn oppor tunity to
ology. Nevertheless, we should strive
intangible s into the justification of new informatio to include these
n systems and processing technologies.
94 Chapter4 Manufacturing Strategy and the Supply Chain

Kaplan [1986] notes that DCF approaches lead to nonoptimal facilities with obsolete process
technology. Fortunately, alternative approaches exist. Uncertainty can be modeled by treat-
ing cash flows as random variables and estimating the distribution of net present value, The
value of flexibility can then be evaluated by considering the distribution of possible cus-
tomer demand, market price curves, discounting sales revenue, and operating expenses un-
der each possible scenario. Intangibles such as the impact of a new system on customer
appreciation and employee morale can be considered by defining qualitative benefits for
each intangible factor under each alternative and computing a net present qualitative flow
value. For instance, suppose we are using the interest rate “r”” in our net present worth com-
putations, and the planning horizon is T periods. Define Qy; as the qualitative flow value
in period ¢ for intangible attribute K, if investment alternative j is adopted. Intangible at-
tributes include issues, such as the types of flexibility and compatibility with core compe-
tencies, and additional factors such as public relations, employee morale, and product man-
ufacturability, discussed earlier in this chapter. Financial risk could also be treated as an
intangible, if desired. Qualitative flow values can be thought of as intangible or surrogate
revenue. Higher values imply a better score. The discounted net present qualitative flow
(NPQF) value of investment alternative j with respect to attribute K is then:
T
NPOFyy = Y, Qu- A+ 77 (4.9)
á
Once the net present qualitative value is computed, multiobjective decision techniques can
be used to compare alternatives based on monetary and intangible benefits. For compar-
ison purposes, qualitative flows can be measured between +1 (ideal solution) and —|
(worst solution), and each attribute can be assigned a relative weight.

A company that prides itself on its quick delivery is considering which of two packaging systems
to use—manual or automated. The automated system will cost $300,000 to purchase plus $50,000
per year to operate. The key attributes are cost, market position, and manufacturing flexibility. Once
installed, the automated system is expected to provide a public relations benefit when trying to at-
tract new customers because lower error rates and quicker delivery times can be promised. The man
ual system may be a liability when customer expectations change and our competitors adapt in the
next several years. The proposed system will automatically identify parts and merge stored product
into orders for shipping. Some training is required to perform the tasks manually. Workers have re-
stricted capability and must be retrained as product lines change. The automated system is expected
to be more flexible to product mix changes. Cash flows and qualitative rankings of the alternatives
for the next three years are given in the following table. After some discussion, the company de-
cided that each attribute can be considered of equal importance when measured between = 1- 5.
$0 cost system would be a + 1, and a three-year present worth cost of $1 million would be the max-
imum possible investment and be rated as a —1.

Year Attribute Manual Automated

1 Cost $125,000 $350,000

2 Cost $130,000 $50,000
3 Cost $145,000 $50.000
1 Public Relations 0.0 -3
2 Public Relations —0.20 0.5
3 Public Relations —0.5 0.4
1 Flexibility 0.3 04
2 Flexibility 0.2 0.5
3 Flexibility 0.2 0.5
 4.1 Manufacturing Strategy — 95

The company uses a minimum attractive rate of return of 15% to evaluate investment opportunities.
Which system should be selected based on total net present value? (Assume returns occur at the
end of the year.)
SOLUTION We begin by using Equation (4.9) to compute a net present value for each alternative on each at-
tribute. For the cost attribute for the manual system we obtain:
NPV ot manuat = — 125,000)(1.15)-* + (—130,000)(1.15) + (—145,000)(1.15)3) = —$214,014.
Similarly,
NPVt automared = (—350,000)(1.15)™" + (=50,000)(1.1572 + 1.15%) = —$427,973.
For the other attributes,

NPOF public relations, manual — 20(1.15)? — 0.5(1.15)3* = —0.428.

Computing in a similar fashion yields,

NPQF public relations, automated = 0.382, NPOF pesibitir, manual = 0.361,
and
NPQF pesiitiy. auomared = 0.750.
The net present values are next scaled to the [—1, 1] interval. For public relations and flexibility,
the maximum net present qualitative factor value would be:

INPOF| = 1(1.15™" + 11572 + 1.157%) = 1.587.

Normalizing the qualitative flows by dividing by 1.587, and then taking a sum of the three attrib-
utes we obtain:

NPQF public relaions. manual = —0.428/1.587 = —0.270,

NPQF public relarions. automarea = 0-382/1.587 = 0.241,
NPO
i manual = 0.361/1.587 = 0.227,
and
NPOF i automarea = 0.750/1.587 = 0.473.
To normalize the cash flows, we must divide by 1,000,000(1.587) = 1,587,000. For the cost of the
manual system, this yields
Mpyremaica — —214014 —
cost.manual = Y 587 000 — —0.135.

Taking the weighted sum across attributes, we obtain:

—0.270 + 0.227 + (—0.135) _

NPVnanual = 3 —0.059;

and
0.241 + 0.473 + (—0.270)
NPVautomared = =0.148.
3

Although the automated system costs more, both initially and over the three-year horizon, when the
intangibles are added in, the automated system has a positive present worth compared with the neg-
ative present worth of the manual system.

Other approaches exist for including qualitative factors and choosing between alter-
natives in multicriteria decision problems. The Analytical Hierarchy Process, or AHP,
96 Chapter4 - Manufacturing Strategy and the Supply Chain

(Saaty [1980], Zahedi [1986], and Golden et al. [1989]) has been widely studied and used
for making such choices. In AHP, analysts judge the relative desirability of alternatives
through pairwise comparisons. Key criteria are first divided into subcriteria to create a
problem hierarchy. For instance, the criterion “Performance” might be subdivided into
“accuracy,” “repeatability,” and “reliability.” At each level pairwise comparisons are made
to assess the relative importance of each subcriterion. ltems on the same level should be
of the same magnitude of importance, thus relative ratings are between 1 and 9. Finally,
the decision alternatives are compared on each of the subcriteria. Weighting the relative
evaluations of the decision alternatives by the relative weights found for the various sub-
criteria, a composite score is summed for each alternative. The AHP approach has the ad-
vantage of permitting consistency checks on pairwise comparisons to ensure rational com-
parisons and can automatically adjudicate between minor inconsistencies in comparisons.
Sensitivity analysis can also be performed.

4.2 SUPPLY CHAIN MANAGEMENT CONCEPTS

From ore to door, materials have to be transported. Beginning with the extraction of ba-
sic elements and grains from the earth, materials, parts, and products must be packed,
loaded, transported, unloaded, and unpacked at every stage of the journey as we convert
raw materials into fabricated parts, then finished products, and, finally, delivered customer
orders. The planning and execution of these activities encompass the function of logis-
tics. The logistics function views the entire supply chain from raw materials to delivered
products. Its task centers on coordinating the flow of materials and associated informa-
tion.
Logistics activities typically account for 25% of the cost of manufacturing. Effective
logistics requires recognition of the geographical location and functional capability of all
potential material suppliers, production facilities, and customer markets, and the trans-
portation options connecting those resources. The combination of material sources, pro-
duction facilities, distribution locations, and transport options providing the most com-
petitive option with respect to product features, cost, quality, and service to customers is
selected. The set of resources selected may include partners external to the firm as well
as internal departments. Given the time span and geographical distance involved in the
logistics network, information management becomes critical. The information required to
execute the logistics function must be shared across time, space, and partners. In this sec-
tion we will examine how logistics has evolved in the modern global marketplace.

4.2.1 Global Logistics

The latter half of the twentieth century has been marked by an accelerating trend toward
globalization. Disparate wage rates, shifting raw material sources, expanding overseas
markets, government policies, and worldwide technical expertise have enticed firms to
take advantage of international opportunities by creating global production and distribu-
tion networks. Opportunities exist for acquiring component parts and new technologies
worldwide to improve production operations and to reduce transit costs. With markets ex-
panding worldwide, it may no longer be desirable to manufacture and ship a product from
a single factory. International markets have grown rapidly in recent years compared with
the U.S. market. Consider, for instance, traditional U.S. companies, such as Coca Cola
and Procter and Gamble. As of 1997, two-thirds of Coca-Cola's and half of Procter
& Gamble's revenues came from overseas operations. Competition now requires that cus-
tomized products be made in small quantities to satisfy local tastes, laws, and infrastruc-