Facility Location

2009-2010 FALL SEMESTER
IE 441 FACILITIES PLANNING AND DESIGN COURSE

LECTURE NOTES 1 PART 1

1
WHERE TO PRODUCE: FACILITY LOCATION

INTRODUCTION

Our discussion of different aspects of a production system design has so far assumed as
given the location of the physical structure, whether we deal with new facilities or the
expansion of existing ones. The selection of location, however, is one of the most far-
reaching top management decisions for the following reasons:

1. It involves the long-term investment of large amounts of capital under conditions of
considerable uncertainty.
2. It determines a rather permanent framework of operating constraints (legal, labor,
community, etc.) that may be difficult and costly to change.
3. It has significant consequences on the competitive position or viability of an
organization by setting a minimum limit on the cost for production and distribution to
desired markets. This is especially true for service systems that must be near the
customers they serve, e.g. restaurants or movie theaters.

In addition, the final solution interacts strongly with other critical decisions considered in
systems design such as facilities layout. It is therefore important to consider thoroughly all the
economic, technological, social, and legal factors that will influence the choice of location.

THE LOCATION PROBLEM

Briefly, the location problem consists of selecting a site for new facilities that will
minimize the production and distribution cost of products and/or services to potential
customers. Such a problem may arise under different conditions, which prompt
management to consider alternative solutions. The final decision is usually based on the
evaluation of both objective and subjective factors for each alternative site.

Reasons for Considering Location Problems
For a new firm selecting a location is an inevitable decision in the phase of system design. Here
the options include the selection of a site for building new facilities or the rent or purchase
of existing ones. For an existing organization, the motivation to consider the location
problem may be attributed to economic, technological, social, or political factors.

2009-2010 SPRING SEMESTER

2
The most important reasons for the need to change or expand to a new location are the
following:

1. Significant changes in the level of demand
2. Significant changes in the geographical distribution of demand
3. Changes in the costs or quality requirements of critical production inputs (labor, raw
materials, energy, or other)
4. Significant increases in the real-estate value of existing or adjacent sites or in their
taxation
5. Need to change as a result of fire or flood or for reasons of prestige or improved public
relations

Even though the selection of a new location represents a serious planning effort, for well-
organized firms it is advisable to review location problems periodically in the light of
significant changes in environmental conditions.

Alternatives to New Location
Of the reasons listed above, the most common is an upward trend in the level of demand. Before
undertaking a detailed study for selecting a new location, however, it is advisable to examine
some alternatives for meeting the expected increase in demand, which include:
1. The increase of existing capacity by additional shifts or overtime, especially for capital-
intensive systems
2. The use of seasonal inventories to reduce the need for maintaining capacity for peak
demand
3. The use of subcontractors
4. The purchase of new equipment for the present location with a less expensive facilities
expansion

For limited or temporary changes in demand, the above alternatives, usually considered in
aggregate planning, represent more economical means of expanding capacity. Otherwise, the
location problem is a very real one and a feasibility study is usually undertaken by a team of
specialists or external consultants.

Significant Factors in Location Studies
As already suggested, the factors considered in location problems may relate to key
production inputs, to the process technology, or to the environment.

Production inputs: The incentive to move or expand to a new location may derive from the need

3
to secure a larger quantity or different quality inputs such as labor, raw materials, energy, or
other. These considerations are related to the markets of such inputs.

Raw materials: For many firms, especially in manufacturing, a dominant factor in plant
location is the need to be near the sources of raw materials. This is especially true when
processing results in significant weight reduction, e.g. firms engaged in mining iron ore,
copper, or marble and firms processing forest products. Proximity to the source of raw
materials is also important for industries that process or package perishable items (dairy
products, fresh fruit, or vegetables, etc.).

In general, an industry using an analytic process, in which raw material is broken down in
successive stages to produce different products, say a lumber mill, tend to locate near the
source of such an input (see Fig.la). On the other hand, for a synthetic process, which
combines a variety of materials, components, and parts in successive stages to assemble a
finished product, there is a tendency to locate near the market (see Fig. 1b).


4
Services in the private sector show a strong tendency to locate near the market of potential
customers. This would be true of a bank, a restaurant, a theater, or an auto-repair service. For
the public sector, however, the location of services is affected both by the geographical
distribution of the need for a service (schools, hospitals, fire or police stations) and by
budget limitations on the degree of desired decentralization (courts, internal revenue offices,
etc.).

Human resources: For many firms with unique or very large labor requirements, proximity to
the appropriate labor market becomes a dominant factor in the selection of a location.
Accordingly, labor-intensive organizations like large assembly-line factories or large insurance
companies tend to locate in or near large metropolitan areas. Similarly, technical consulting
firms gravitate to large universities for securing the needed expertise more easily. In
addition to prevailing wages and salaries, location also determines the attractiveness of
a firm according to the time required to commute to work. In some cities, this may be so long
and tiresome that it may discourage potential employees.

The importance of labor may be reduced when it is possible to automate or mechanize one or
more stages of the production process. This represents a substitution of capital for labor, which
is possible through technological advances. Numerically controlled equipment in
manufacturing, printed circuits in electronics, and computers for data processing in large
office companies are just a few examples of this possibility. The same result is possible when
management can draw on a mobile or seasonal labor market, as in harvesting farms or
operating tourist resort areas.

At times the wish to reduce labor costs places unwarranted emphasis on alternatives for
location in a foreign country where wages are quite low. The cost of labor as an input,
however, must be examined together with labor productivity, because despite low wages in some
foreign locations, the accompanying low productivity results in higher overall production
costs.

Process technology: For some firms the technology used may restrict the number of
locations to sites that provide an abundant low-cost supply of some critical input, such as
water for pulp and paper mills or electrical energy for an aluminum plant or
electrochemical plating process. Occasionally, in addition to large amounts, of certain
inputs there is an added requirement for meeting strict quality specifications. For example,
water that may be suitable for human consumption may not be appropriate for certain
industrial uses such as steel manufacturing.

Environmental factors: Beyond the consideration of factors related to the production
process and its critical inputs, the location decision depends on several factors that define the

5
external environment:

1. The availability and reliability of supporting systems, including public utilities for
power and water, fire protection, easy transportation routes to suppliers and consumers, rapid
and reliable communication, etc. Examination of these factors is especially important
for the location of a new plant in a foreign country, whose economic infrastructure
may lag significantly behind that of industrialized countries. Foreign investors have
been discouraged from establishing new facilities in certain countries because long-
distance calls may take 3 to 5 days to place, delivery of supplies is unreliable due to poor
transportation, and other difficulties.

2. Social and cultural conditions may at times discourage the selection of a location that could
pass any economic- and technical-feasibility criterion. It is thus necessary to understand the
local population not only in terms of demographic variables (size. distribution, age.
migration shifts, etc.) but also in terms of their attitudes toward domestic or foreign new industry
and the quality, availability and reliability of potential employees. Certain traditions and
customs, especially abroad, may interfere, with known ways of doing business. In several
countries, for example, informal communication networks and personal relationships play a
more dominant role than formalized procedures and plans.

3. Legal and political considerations represent a wide variety of restrictions or
opportunities and must be studied very carefully before making a final choice. In some
locations there are very strict laws pertaining to pollution standards, zoning codes,
construction specifications, or import regulations. These restrictions may make it difficult to
operate with existing technologies at a profit. Also, certain states or communities provide a
wide variety of incentives to attract new employers. These may include reduced taxes, purchase
discounts for construction sites, and less elaborate licensing procedures for new installations.
One cannot overemphasize the importance of competent legal advice before making any final
decisions on locating new facilities.

Formulation of the Location Problem

In order to formulate the location problem more precisely, it is helpful to view the
operations system under study in relationship to its market and its sources of supply, i.e. its
economic environment. This is shown in Fig. 2, where some of the most important costs that
must be estimated for alternative location sites include:


6

1. The cost of procurement of needed production inputs C1 that is, the cost of raw
materials, labor, energy, etc. (mainly variable costs)
2. The cost of processing the inputs given the technology Cp, that is the overhead cost
(mainly fixed costs)
3. The distribution cost involved in shipping product(s) or making services available to
customers C
D
(mainly variable costs)

These costs are tangible; i.e. they can be estimated for different locations using standard
economic analysis. In addition, we must take into account certain intangible costs related to
the quality of the labor available in each location, the degree of cooperation and the
attitudes of the local government and community, possible relocation adjustments, and
others. Finally, there is an opportunity cost for each location resulting from failure to
select the best site possible if time and money impose no restrictions in the search for
alternatives.

Our formulation of the location problem can now be made more complete and precise.
Management has the task of selecting among candidate locations the one that satisfies
existing technological, legal, and other constraints and minimizes the combined cost of the
relevant tangible, intangible, and opportunity costs. In practice, intangible costs can be

7
estimated only subjectively for each location, whereas opportunity costs are likely to be
ignored in most studies.

A PRACTICAL SYSTEMS APPROACH TO
LOCATION SELECTION

As we have seen, the problem of selecting a location is characterized by numerous factors with
complex interrelationships. Several of these factors can be evaluated only
qualitatively at best. Furthermore, the information needed is often incomplete due to the
inherent difficulty of predicting future conditions. Various sophisticated techniques
developed to solve parts of the total problem include linear programming, heuristic and
simulation models based on some dominant objective, such as minimizing distribution
costs.

Management, however, needs an approach that looks at the whole problem and allows for the
careful evaluation of both quantitative and qualitative factors. This can be attempted within
the framework of a systems analysis, which must cover the components shown in Figure 3.

Figure 3 Interacting factors in the selection of location.

In practice, it is often useful to study the location problem in two phases, as shown in Fig. 4.

8
First, there is a preliminary feasibility study, whose purpose is to determine whether the
environmental changes are important enough to warrant a more detailed analysis. Thus phase
one is mainly concerned with the study of trends in the level and geographical distribution of
aggregate demand to determine whether they justify the minimum economical addition to
capacity obtained by building new facilities.

Along with demand, the preliminary study focuses on other environmental factors such as the
availability of critical production inputs, their current and projected costs, and any
demographic changes that may affect the distribution of demand and/or the availability of
labor or other resources. If the results of phase one justify the need for a more detailed
analysis, management can proceed with phase two, which is conducted in three successive
stages, as explained below.

9

Figure 4. Procedure for selecting a location for new facilities


10
Stage 1: Analysis and Evaluation of Alternative Regions (or Countries)

In stage 1 of phase two we examine a number of general characteristics for different
geographic regions or countries as they relate to the scope of the firm's present and/or
proposed activities. The purpose here is to select the most suitable region and to determine the
capacity needed for the new facilities.

First, the marketing-research department or a special group for this task must prepare
estimates of the expected increase in demand for each region of the market, using a 5 to 10-yr
forecast. These estimates are then translated into requirements for new or additional
productive capacity. Given an estimate of future capacity requirements, we proceed to identify
alternative regions for which profits or other measure of effectiveness will be maximized for the
proposed capacity. For specific demand levels and prices, profits are determined from the
analysis of costs for production and distribution. Thus in stage 1 we must consider the
alternatives of meeting new capacity requirements from one centralized large installation or two
or more smaller-capacity facilities geographically dispersed to reduce transportation costs.

Figure. 5 Relationship of production and distribution cost curves and revenue curve for facilities of different
capacity.

Figure 5 shows the relevant trade-offs. To obtain lower production costs we must achieve
economies of scale made possible by larger-capacity plants. This pressure for capacity
concentration, however, results in larger distribution costs due to the larger distances for
product deliveries, compared with several small plants geographically dispersed. The
optimum configuration is selected from an economic analysis aimed at identifying the
minimum production-distribution cost alternative combined with other considerations for
greater flexibility and reliability usually in favor of decentralized arrangements.
The data needed to select the most suitable region(s) and the new capacity requirements include

11
the following:

A. Future increase in demand by region translated into productive-capacity requirements
B. Cost relationships for production and distribution
C. Identification of sources of needed production inputs:
1. Raw materials (quality, quantity, cost, reliability)
2. Labor market (available skills, wages, and supply levels)
3. Supporting systems, i.e., economic infrastructure, for supplying
a. Energy (sources, adequacy, cost)
b. Water (quality, quantity, and cost)
c. Transportation and communication networks (adequacy, reliability, cost)
4. Legal, social, and political factors
5. Environmental considerations (pollution, climate, quality of life)

According to Yasseen almost all elements contributing to cost can be affected up to 10
percent of the total for production and distribution by the selection of the geographic region.

Stage 2: Determination of Optimum Area for Location

After selecting the most suitable geographical region (or foreign country) and the
optimum-capacity plant, the analysis continues at the level of specific areas within the
region. If our criterion for selecting the best area is profit maximization, then with
production costs already determined in stage 1 the maximization of profit can now be
achieved by minimizing the cost of distribution.

Consider the case of a new firm about to select the location for its new facilities. First, we
subdivide the market to be served by the new plant into different sections and estimate for
different areas the distances and projected requirements to cover future demand. For a given
location let

n = number of market sections to be covered by new plant from that location

D
j
= amount of monthly deliveries to section j (j = 1, 2, .... n)

r
j
= distance of section j from plant location

c
j
= unit shipping cost per unit distance to destination j

Then

12

This measure can be evaluated for each alternative location to choose the one, which
minimizes the total annual distribution cost.

For an existing firm, which operates one or more facilities, the situation is more complicated in
that demand for various market sections can be met from the old as well as the new
location(s). This type of problem has been solved by the transportation method of linear
programming. To illustrate it with a simple example, let us assume that the firm under study has
an old plant and is considering building a new one with capacity equal to 60 units/month in
location A, B, or C. The data needed are summarized below:

Note that:
c
ij
= cost of shipping one unit from plant I to market section j
Xij = number of units shipped monthly from plant i to j
D
j
= amount of monthly deliveries needed for market section j
Si = capacity available per month in plant i

By substituting the cost values for c
ij
that apply to the three areas A. B, and C and solving the
transportation problem repeatedly, we can determine which of the three locations results in the
minimum transportation cost per week.

Whether the area selection is made by straightforward economic analysis or a mathematical
model, it is important to conduct a sensitivity analysis to determine whether the area selected
maintains its relative advantage over other locations under different operating conditions or
changes in market characteristics.

13

Stage 3: Community and Site Selection

Within the area chosen in stage 2 it is desirable to consider alternative sites for the
construction of the new facilities. At this point, it is important to secure detailed economic
and demographic data, but most factors that must be analyzed in area and site selection relate to
technical, social, and legal considerations.

The factors that must be considered in the selection of a community and a site for a new
location can be classified as follows:
A. Projected requirements in production inputs
1. Human resources (skills, amounts, quality)
2. Raw materials, parts, semifinished components
3. Energy, water, and other services
4. Transportation and communication facilities
5. Physical space for planned facilities and future expansion

B. Objective factors that will affect the cost and profits of new installation
1. Projected levels of annual demand
2. Projected annual operating costs
a. Costs for purchase and transportation of raw materials
b. Costs for wages of required skills
c. Costs for requirements in energy (electricity, oil. etc.), water, telephone, etc.
d. Taxes on (sales, income, property, inventories, etc.).
e. Cost for construction of new facilities
4. Estimates of annual profits for successive years
5. Cost of purchasing construction site
C. Subjective factors that will influence the community and site selection
1. Existing laws that will affect the firm's activities
2. Labor-market, characteristics
3. Transportation networks
4. Supporting infrastructure systems (power, telephone, water, waste treatment. or
other)

5. Community characteristics
a. Population makeup, attitudes, traditions
b. Financial institutions
c. Cultural activities, schools, recreation
d. Quality of Life (noise, congestion, air pollution, etc)

14
e. Housing
d. Services

Evaluation of these factors is difficult because of the problems of (1) estimating objective
factors with accuracy, (2) assigning priorities to subjective factors that defy measurement,
and (3) coping with the uncertainty about the future impacts of present decisions when the
stakes are so high.

Logistics management can be defined as the management of the transportation and
distribution of goods (Russell and Taylor 1995). The term goods includes raw materials or
subassemblies obtained from suppliers as well as finished goods shipped from plants to
warehouses or customers. The logistics management includes all distribution and
transportation activities from suppliers through to customers. Similar to the definition of
material handling in Chapter 11, logistics management is the management of a series of
macro-level transportation and distribution activities with the main objective of delivering

the right amount of material at the right place at the right time at the right cost using the right
methods. The decisions typically encountered in logistics management concern facility location,
transportation, and goods handling and storage.

Logistics management problems can be classified into three categories:

1. Location problems

2. Allocation problems

3. Location-allocation problems

Location problems involve determining the location of one or more new facilities in one or
more of several potential sites. Obviously, the number of sites must at least equal the number of
new facilities being located. The cost of locating each new facility at each of the potential sites
is assumed to be known. It is the fixed cost of locating a new facility at a particular site plus the
operating and transportation cost of serving customers from this facility-site combination.

Allocation problems assume that the number and location of facilities are known a priori and
attempt to determine how each customer is to be served. In other words, given the demand for
goods at each customer center, the production or supply capacities at each facility, and the
cost of serving each customer from each facility, the allocation problem determines how much
each facility is to supply to each customer center.

Location-allocation problems involve determining not only how much each customer is to

15
receive from each facility but also the number of facilities along with their locations and
capacities.

Facility location problems can be classified as single-facility and multifacility problems. As the
name implies, single-facility location problems deal with the optimal determination of the
location of a single facility, and multifacility location problems deal with the simultaneous
location determination for more than one facility. Generally single-facility location problems are
location problems, but multifacility problems can be location as well as location-
allocation problems (Hax and Candea 1984).

Another classification of location problems is based on whether the set of possible locations for
a facility is finite or infinite. If a facility can be located anywhere within the confines of a
geographic area, then the number of possible locations is infinite, and such a problem is
called a continuous space location problem. In contrast, discrete space location problems
have a finite feasible set of sites in which to locate a facility. The continuous space problem
assumes that the transportation costs are proportional to distance. Because facilities can be
located anywhere in a two-dimensional space, sometimes the optimal location provided by the
continuous space model may be infeasible. For example, a continuous space model may locate
a manufacturing facility on a lake! Therefore, for most real-world problems, the discrete
space models are more appropriate (Hax and Candea 1984).

TECHNIQUES FOR DISCRETE SPACE LOCATION PROBLEMS

Our focus is primarily on the single-facility location problem. We provide both discrete space and
continuous space models. The single facility for which we seek a location may be the only one
that will serve all the customers, or it may be an addition to a network of existing facilities that are
already serving customers.

Qualitative Analysis
The location scoring method is a very popular, subjective decision-making tool that is
relatively easy to use. It consists of these steps:
Step 1 List all the factors that are importantthat have an impact on the location decision.
Step 2 Assign an appropriate weight (typically between 0 and 1) to each factor based
on the relative importance of each.
Step 3 Assign a score (typically between 0 and 100) to each location with respect to
each factor identified in step 1.

16
Step 4 Compute the weighted score for each factor for each location by multiplying its
weight by the corresponding score.
Step 5 Compute the sum of the weighted scores for each location and choose a location
based on these scores.

Although step 5 calls for the location decision to be made solely on the basis of the weighted
scores, those scores were arrived at in a subjective manner, and hence a final location
decision must also take into account objective measures such as transportation costs, loads, and
operating costs. Quantitative methods that consider some of these measures are discussed in the
following section (Quantitative Analysis).

Example 1

A payroll processing company has recently won several major contracts in the midwest
region of the United States and central Canada and wants to open a new, large facility to
serve these areas. Because customer service is so important, the company wants to be as near its
"customers as possible. A preliminary investigation has shown that Minneapolis, Winnipeg,
and Springfield, Illinois are the three most desirable locations, and the payroll company has
to select one of these. A subsequent thorough investigation of each location with respect to
eight important factors generated the raw scores and weights listed in the following table.
Using the location scoring method, determine the best location for the new payroll processing
facility.


17

Solution Step 1, 2, and 3 have been completed. We now need to compute the weighted score
for each location-factor pair (step 4), add these weighted scores and determine the location
based on the scores (step 5). From the analysis below, it is clear that Minneapolis is the best
location based on the subjective information given.


18
Of course, as we mentioned before, objective measures must be considered, too, especially
because the weighted scores for Minneapolis and Winnipeg are close.

Quantitative Analysis

Several quantitative techniques are available to solve the discrete space, single-facility
location problem. Each is appropriate for a specific set of objectives and constraints. For
example, the so-called mini-max location model is appropriate for determining the location of an
emergency service facility, where the objective is to minimize the maximum distance
traveled between the facility and any customer. Similarly, if the objective is to minimize the
total distance traveled, the transportation model is appropriate. In this section we discuss the
transportation model in detail and briefly describe the mini-max location model.

The transportation model is well known and described in many introductory operations
research textbooks, including Winston (1994) and Hillier and Lieberman (1995). Therefore we
do not describe the model or the algorithm in detail but note that the model is typically used to
determine the optimal distribution of goods between a given set of "plants" and a known set
of "customers" so as to minimize the overall cost of transporting goods between the plants and
customers while satisfying customer demand and plant supply constraints. The reader may be
wondering: If the set of plants including their locations is given, where is the location problem?
To answer this question, consider the following problem: We have m plants in a distribution
network that serves n customers. Due to an increase in demand at one or more of these n
customers, it has become necessary to open an additional plant. The new plant could be located
at p possible sites. To evaluate which of the p sites will minimize distribution
(transportation) costs, we can set up p transportation models, each with n customers and
m + 1 plants, where the (m + 1)
th
plant corresponds to the new location being evaluated. Solving
the model will tell us not only the distribution of goods from the m +1 plants (including the
new one from the location being evaluated) but also the cost of the distribution. The location
that yields the least overall distribution cost is the one where the new facility should be
located. This is illustrated in Example 2.

Example 2

Seers Inc. has two manufacturing plants at Albany and Little Rock that supply CANMORE
brand refrigerators to four distribution centers in Boston, Philadelphia, Galveston, and
Raleigh. Due to an increase in the demand for this brand of refrigerators that is expected to last
for several years, Seers Inc. has decided to build another plant in Atlanta or Pittsburgh. The
unit transportation costs, expected demand at the four distribution centers and the
maximum capacity at the Albany and Little Rock plants are given in the following table.
Determine which of the two locations, Atlanta or Pittsburgh, is suitable for the new plant.

19
Seers Inc. wishes to utilize all of the capacity available at its Albany and Little Rock
locations.

Solution We must set up two transportation models: first assuming that the new plant will be
located in Atlanta and second in Pittsburgh. These two models are displayed below. Notice
that the maximum capacity of the new plant required at either location is 330 because the
capacity at Albany and Little Rock is to be fully utilized. Solving the model using the well-
known transportation algorithm, yields total costs of $7980 and $9510 when the new plant is
located in Atlanta and Pittsburgh, respectively. In both cases the distribution pattern (not
shown) remains the same, with Boston receiving all of its shipments from Albany, Galveston
from Little Rock, Raleigh from the new plant, and Philadelphia receiving 50 units from
Albany and 50 from the new plant. Because the Atlanta location also minimizes the cost, the
decision is to construct the new plant in Atlanta.


20
We mentioned earlier that the location of an emergency facility such as a fire station, police
station, or hospital can be treated as a mini-max location problem if the objective is to
minimize the maximum distance traveled or the "cost" incurred by any customer. We now
show how such a problem can be solved trivially by inspecting a cost or distance matrix.
Consider the problem of locating a fire station in a county. Assume that we have m possible
sites and that the county can be divided into n locations based on population. The distance
between the centroid of each of the n locations and each potential site for the fire station is
shown in this distance matrix:

To find the optimal location we simply need to scan each row for the maximum distance and
choose the site corresponding to the row that has the smallest maximum distance. This
location ensures that the maximum distance that needs to be traveled by any customer is
minimized.

Hybrid Analysis

A disadvantage of the qualitative method discussed in Qualitative Analysis section is that a
location decision is made based entirely on a subjective evaluation. Although the quantitative
method overcomes this disadvantage, it does not allow us to incorporate unquantifiable
factors that have a major impact on the location decision. For example, the quantitative
techniques can easily consider transportation and operational costs, but intangible factors,
such as the attitude of a community toward businesses, potential labor unrest, and reliability of
auxiliary service providers, though important in choosing a location, are difficult to
capture. We need a method that incorporates subjective as well as quantifiable cost and other
factors.

In this section we discuss a multi-attribute, single-facility location model based on the ones
presented by Brown and Gibson (1972) and Buffa and Sarin (1987). This model classifies the

21
objective and subjective factors important to the specific location problem being addressed as:
Critical;
Objective;
Subjective.
The meaning of the latter two factors is obvious, but the meaning of critical factors needs
some discussion. In every location decision, usually at least one factor determines whether or not
a location will be considered for further evaluation. For example, if water is used extensively
in a manufacturing process (e.g., a brewery), then a site that does not have an adequate water
supply now or in the future is automatically removed from consideration. This is an example
of a critical factor. Some factors can be objective and critical or subjective and critical. For
example, the adequacy of skilled labor may be a critical factor as well as a subjective factor.

Let us assume that we have m candidate locations and p critical, q objective, and r subjective
factors. We can determine the overall critical factor measure (CFM), objective factor
measure (OFM) and subjective factor measure (SFM) for each location i with these
equations:


22
where is the weight assigned to the objective factor measure. Notice that even if one critical
factor is not satisfied by a location i, then CFMi and hence LMi are equal to zero. The OFMi
values are calculated so that the location with the maximum
j
OF
ij
gets an OFM
i
value of zero
and the one with the smallest
j
OF
ij
value gets an OFM
i
value of one. Equation (4) assumes
that the objective factors are cost-based. If any of these factors are profit-based, then a negative
sign has to be placed in front of each such objective factor and Equation (4) can still be used.
This works because maximizing a linear profit function z is the same as minimizing -z.
(The following example illustrates how profit-based as well as cost-based objective factors can
be incorporated into the same model.)

After LMi is determined for each candidate location, the next step is to select the one with the
greatest LM
i
value. Because the a weight is subjectively assigned by the user, it may be a
good idea for the user to evaluate the LMi values for various appropriate a weights, analyze the
trade-off between objective and subjective measures, and choose a location based on this
analysis.

Example 3

Mole-Sun Brewing company is evaluating six candidate locationsMontreal, Plattsburgh,
Ottawa, Albany, Rochester, and Kingstonfor a new brewery. The two critical, three objective,
and four subjective factors that management wishes to incorporate in its decision making are
summarized in Table 13.7. The weights of the subjective factors are also provided in the table.
Determine the best location if the subjective factors are to be weighted 50% more than the
objective factors.

Solution Clearly, = 0.4 so that the weight of the subjective factors (1 - = 0.6) is 50%
more than that of the objective factors. To determine OFMi we first determine the j OFij
values as shown in the following tables and then ascertain the maximum and minimum of
these to be -35 and -95, respectively. From these two values, it is easy to calculate OFM
i
values using Equation (2). For example, for the Montreal location,

-35 - (-67)
OFM = = 0.53
-35 (-95)

Similarly the SFM, values are determined for each row in the following table, using the
information available in the last four columns and Equation (3). Once again, for the Montreal
location,


23
SFM = (0.3) (0.4) + (0.4) (0.8) + (0.25) (0.2) + (0.05) (0.8) = 0.53

Using the CFMi, OFMi, and SFMi values, we get the location measure LMi. The values are
shown in tables below. Based on an value of 0.4, the Plattsburgh location seems favorable.
However, as the weight of the objective factors, , increases to more than 0.6, the Montreal
location becomes attractive.

The Brown-Gibson Approach for Site Selection

Suppose that after stage 2 we have identified three alternative sites for a. new plant location.
We wish to develop for each site a measure of preference that combines both objective and
subjective factors. For this purpose we might use the method for evaluating alternative
product designs. A versatile approach especially designed for the location-selection problem is
a model developed by Brown and Gibson; its key features are displayed in Fig. 6. In
applying the Brown-Gibson approach, outlined in Fig. 6, we go through the following
sequence of steps:

24
1. Eliminate any site that does not meet certain basic requirements. This may be a
technical requirement, e.g., the availability of abundant cheap electricity for an
aluminum plant, or a budget constraint for the purchase of the site.
2. Compute an objective-factor measure of performance OF
i
for each site. Usually we
estimate all relevant costs to compute the total annual cost for each site Ci. Next
we determine the objective-factor measure OFi by multiplying Ci, by the sum of the
reciprocal site costs (1/C
i
) and taking the reciprocal.
3. Determine key subjective factors and estimate their subjective-factor measure
SF, for each site by
a. Deriving a factor rating wj for each subjective factor (j = 1, 2,..., n)
using a forced- choice pairwise comparison procedure. Accordingly, one
factor is selected over another, or they are rated equal.
b. Ranking each site for each subjective factor separately R
ij
(0 R
ij
1, R
ij
=1).
c. Combining for each site the factor rating and site ranking, SFi = w1Ri1 + w2Ri2+ . .
. + wnRin
4. Combine for each site the objective-factor measure OF
i
and subjective-factor measure
SF
i
by assigning weights k and 1 - k respectively, to obtain a location preference
measure (LPM), i.e.,

LPMi = k(OFi) + (l-k)(SFi) 0 < k < 1
If the site selection is going to be based entirely on costs, k = 1.0 and the subjective-
factors are ignored. If objective factors ore going to count 4 times as much as the
subjective ones, then k = .8 (.8/.2 =4), etc. The preference measure for a given site
can be tested under different assumptions concerning costs and the importance of
various subjective factors.

5. Select the site with the maximum LPM.


25

This procedure will be illustrated with an example. Note, however, that it is quite general in
nature and can be used not only for site selection but also for many complex decision
problems where it is necessary to combine subjective and objective, factors into an overall
measure of preference for each alternative. The Brown-Gibson method would be equally
suitable for the evaluation of alternative product designs, processes, or layouts of facilities, to
name a few potential applications.

Example 5

Miranda Pacific, Inc., a pulp and paper company, has just completed a feasibility study
(Phase one) for the construction of a new plant to cover future increases in demand, as
evidenced by a steady upward trend during recent years. The study has concluded that the
expected increase in demand justifies the minimum economic plant size needed for a paper
mill, since capacity changes for this industry can be made only in large increments.

In stage 1 of phase two the study of market changes and pulp and paper manufacturing
characteristics for different geographic regions has shown that the most suitable region for a
new plant is the Pacific Northwest (Idaho, Oregon, Washington) because of its proximity to the
raw material (wood chips converted into paper pulp and then various types of paper) and good
transportation networks (waterways and rail and highway systems).


26

In stages 2 and 3 for the selection of the best area and sites, a detailed analysis of
requirements has indicated that four sites along the Columbia River would be satisfactory in
terms of the large quantities of water required for pulp and paper processing. Let us call these
sites A. B. C, and D near different towns along the river. The application of the Brown-
Gibson approach for this case is illustrated below.

Step 1 Site D is eliminated early because real estate in that area is prohibitively expensive and the
sum needed to purchase the required number of acres exceeds the amount allowed in the capital
budget.
Step 2 In the remaining sites all annual costs for production inputs (raw materials. energy,
etc.) are the same except for labor, distribution, and taxes:

The objective rating factor for each site is obtained by substitution in the formula

Thus, the objective factors for alternative sites are

Step 3 With climate, recreation opportunities, city services, and labor being the same in all
areas, the key subjective factors in the selection of a site have been identified as (1) housing,
(2) education, and (3) community attitudes.

These must now be evaluated using a forced-choice procedure, which compares each factor
with the others by considering one pair at a time. Repetitive application of the forced-choice
procedure yields two results: (1) we obtain a subjective factor importance index w
j
, which
simply measures the relative weight given to each .subjective factor, and (2) for each
subjective factor separately we obtain a ranking of each site Rij.

27
Suppose that the special group assigned to select a site responded to paired comparisons as
follows:
1. Education (1) versus housing (2): both judged equally important
2. Education (1) versus community attitudes (3): choose community attitudes as more
important
3. Housing (2) versus community attitudes (3): choose community attitudes
This information can be summarized in a table in which we can compute the subjective-
factor importance index wj:

In the column for each pairwise comparison possible we assign 1 to the factor preferred and 0 to
the other, while for the case of equivalence both factors are assigned a value of 1. It is
important in using such a procedure to check preferences for consistency. Thus, if factor 1 is
preferred to 2 and 2 is preferred to 3, 1 must be preferred to 3; otherwise the responses are
inconsistent.

Next, for each subjective factor separately we repeat the same pairwise comparisons with sites to
determine their relative ranking R
ij
. For the sites considered by Miranda Pacific, Inc., the results
of this procedure are shown in the following table.


28

To determine the subjective factor measure for a site SF
i
we must multiply the site rating for a
given factor Rij by the relative importance index wj of that factor and sum over all the
subjective factors included in the analysis

Step 4 Having completed the evaluation of both objective and subjective location factors, we
can now proceed to combine the results and determine an overall location-preference
measure LPM, for each site. This synthesis requires a crucial decision, the weight that will be
assigned to each category of factors. If the weight given to the objective factors is k (0<k<1),
the subjective factors receive a weight 1 - k. Thus, for a given site we have

For our example, management considers the weight of objective factors 4 times as
important as that of subjective ones; that is, k = 4(1 - k). This is true when k = .8. The

29
location-preference-measure values for the sites considered will be

Step 5 According to the Brown-Gibson approach, Miranda, Inc., must choose site C, since this
receives the highest value for a location measure. We note that even though it is the least
attractive alternative based on objective factors, its superiority in subjective-factor rating puts
it at the top, despite a high weight value for cost factors. The sensitivity of the preference-
location measures for each site with respect to the weight assigned to objective and
subjective factors can be shown with the following diagram. For our example, only for a
value of k near 1.0, for which subjective factors are essentially ignored, does the choice
switch to site A (A potential weakness of the Brown-Gibson approach is in the use of
pairwise comparisons. These may oversimplify a choice by failing to reveal the strength of a
staled preference).


Facility Location

Uploaded by

Copyright:

Available Formats

Facility Location

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Facility Location

Uploaded by

Copyright:

Available Formats

2009-2010 FALL SEMESTER

IE 441 FACILITIES PLANNING AND DESIGN COURSE

You might also like