Facility Location and Decision Tree

Facility Location
Facility location is the process of determining a suitable site for the operations of the company.
Factors affecting the location decisions are:

1. Market proximity
2. Availability of skilled labor
3. Site cost
4. Availability of amenities. (Ex. Housing, shops, community services, communication systems etc. )
5. Availability of transportation facilities.
6. Availability of services
7. Suitability of land & Services
8. Regional regulations
9. Safety requirements
10. Political, cultural & economic situations
11. Scope for extensions.
Methods of Location analysis:
1. Preference-Matrix Approach
2. Load-Distance Method
3. Centre of Gravity Method
4. Break-Even Analysis
Preference-Matrix Approach :
This approach considers the qualitative factors pertaining to each possible site into the location
evaluation. A qualitative factor is the one which cannot be evaluated in monetary terms such as quality
of life or community attitude.
In this approach the qualitative and quantitative factors are merged together by assigning a weight of
relative importance to each factor and by calculating the weighted score of each possible site using a
performance matrix. The site with the highest weighted score is the best site.
Load-Distance Method:
It is a mathematical model that is used to evaluate locations based on proximity factors.

This method assumes that several location factors relate directly to distance. Ex: Proximity to markets,
Proximity to material/suppliers/ resources, Average distance to target customers, proximity to other
company facilities etc.
The objective is to select a location that minimizes the total weighted loads moving into and out of the
facility.
The possible locations are identified by assigning grid coordinates on a map. The distance between two
locations on the map is calculated either by Euclidean distance measure or a Rectilinear distance measure.
a. Euclidean distance is the straight line distance or the shortest possible path. For ex: if (XA, YA) and (XB,
YB) are the coordinates of points A and B on a map then Euclidean distance between A and B is the length
of the hypotenuse of the right triangle and is given as:
𝑑𝐴𝐵 = √(𝑋𝐴 − 𝑋𝐵 )2 + (𝑌𝐴 − 𝑌𝐵 )2
b. Rectilinear distance measures distance between two points with a series of 90 degree turns. It is
calculated by adding absolute value of the difference in X-coordinates to the absolute value of the
difference in Y-coordinates. Essentially it is the sum of the lengths of the lines representing the base and
side of the right triangle and is given as:
𝐷𝐴𝐵 = |𝑋𝐴 − 𝑋𝐵 | + |𝑌𝐴 − 𝑌𝐵 |
To calculate load-distance or simply ‘LD’ score for any potential location the distance travelled calculated
either by Euclidean or Rectilinear distance measure is multiplied by the loads flowing to and from a facility
to that potential location. The location with the lowest total LD score is the best. Generally rectilinear
distance measure is used in location decisions.
Center of Gravity Method:
A good starting point for evaluating different locations in the

Load-Distance method is by calculating center of gravity of the
target area.
The center of gravity’s X-coordinate denoted as X* is found by multiplying each points X-coordinate say
Xi by its load Li, summing these points ∑(𝐿𝑖 ∗ 𝑋𝑖 ), and then dividing by sum of loads ∑(𝐿𝑖 ).
The coordinate denoted as Y* is also calculated in a similar fashion. The formulae are.
The location represented by center of gravity’s X* and Y* coordinates may not be the optimal one as far
as Euclidean and Rectilinear distance measures are concerned, But it is a good starting point for testing
different location in its vicinity.
By calculating the load-distance scores for different locations in the vicinity of center of gravity the optimal
location may be determined.
Decision tree
A decision tree is a decision support tool that uses a tree-like graph or model of decisions and
their possible consequences, includingchance event outcomes, resource costs, and utility. It is
one way to display an algorithm. Decision trees are commonly used in operations research,
specifically in decision analysis, to help identify a strategy most likely to reach a goal. Another use
of decision trees is as a descriptive means for calculating conditional probabilities.
In decision analysis, a "decision tree" — and the closely-relatedinfluence diagram — is used as
a visual and analytical decision support tool, where the expected values (or expected utility) of
competing alternatives are calculated.
A decision tree consists of 3 types of nodes:-
1. Decision nodes - commonly represented by squares

2. Chance nodes - represented by circles
3. End nodes - represented by triangles
Drawn from left to right, a decision tree has only burst nodes (splitting paths) but no sink nodes
(converging paths). Therefore, used manually, they can grow very big and are then often hard to
draw fully by hand. Traditionally, decision trees have been created manually - as the aside
example shows - although increasingly, specialized software is employed.
Analysis can take into account the decision maker's (e.g., the company's) preference or utility
function, for example:
The basic interpretation in this situation is that the company prefers B's risk and payoffs under
realistic risk preference coefficients (greater than $400K—in that range of risk aversion, the
company would need to model a third strategy, "Neither A nor B").
Influence diagram
A decision tree can be represented more compactly as an influence diagram, focusing attention
on the issues and relationships between events.
The squares represent decisions, the ovals represent action, and the diamond represents results.
Uses in teaching
Decision trees, influence diagrams, utility functions, and other decision analysis tools and
methods are taught to undergraduate students in schools of business, health economics, and
public health, and are examples of operations research or management science methods.
Advantages
Amongst decision support tools, decision trees (and influence diagrams) have several
advantages:
Decision trees:
 Are simple to understand and interpret. People are able to understand decision tree
models after a brief explanation.
 Have value even with little hard data. Important insights can be generated based on experts
describing a situation (its alternatives, probabilities, and costs) and their preferences for
outcomes.
 Use a white box model. If a given result is provided by a model, the explanation for the result
is easily replicated by simple math.
 Can be combined with other decision techniques. The following example uses Net
Present Value calculations, PERT 3-point estimations (decision #1) and a linear distribution
of expected outcomes (decision #2):
Decision trees tutorial class solution
The decision tree for the problem is shown below.

This decision tree illustrates a reactive situation - we can
either carry out some action now (i.e. invest 750K) or we
can "wait and see" (don't invest). If we choose to wait and
see then we may react depending upon what our competitor
Y does.
Below we carry out step 1 of the decision tree solution
procedure which involves calculating the total profit for
each of the paths from the initial node to the terminal
nodes. We have also numbered all the nodes in the decision
tree from 1 to 21.
Note here that we have included in the decision tree
alternative 5 (no program). This is because it may not be
worthwhile undertaking the crash/normal programs of
minor improvements.
Step 1
 path to terminal node 9 - X invests 750K, Y improves,

sales increase 8% on sales of 500,000
Profit contribution (pc) change = +0.08 x 500 x 40
Total cost = 750
Total pc change = 850 (all figures in £K)
Note here that, instead of total profit which we had in the
M997 example, we focus here on the profit contribution
(pc) change based on the £40 per unit profit contribution.
 path to terminal node 10 - X invests 750K, Y doesn't
improve, sales increase 15%
Total cost = 750
Total pc change = 2250
Total cost = 750
Total cost = 750
 path to terminal node 13 - X doesn't invest, Y
improves, X has crash program costing 600K which is
successful and results in a sales increase of 7%
Total cost = 600
improves, X has crash program costing 600K which
fails and sales decrease 15%
Profit contribution (pc) change = -0.15 x 500 x 40
Total cost = 600
Total pc change = -3600
Total cost = 600
Total cost = 600
improves, X has normal program costing 400K
resulting in a sales increase of 5%
Total cost = 400
improves, X does nothing and sales decrease 15%
Total cost = 0
Total cost = 0
Total cost = 0
 path to terminal node 21 - X doesn't invest, Y doesn't
improve, sales unchanged
Profit contribution (pc) change = 0
Total cost = 0
Total pc change = 0
Hence we can form the table below indicating for each
branch, the total pc change involved in that branch from the
initial node to the terminal node.
Terminal node Total pc change (£K)
9 850
10 2250
11 1250
12 250
13 800
14 -3600
15 -2600
16 -1600
17 600
18 -3000
19 -2000
20 -1000
21 0
We can now carry out the second step of the decision tree
solution procedure where we work from the right-hand side
of the diagram back to the left-hand side.
Step 2
Consider chance node 7 (with branches to terminal nodes

14, 15 and 16 emanating from it). The expected monetary
value (EMV) for this chance node is given by
0.8 x (-3600) + 0.1 x (-2600) + 0.1 x (-1600) = -
3300
node 14 node 15 node 16
Consider chance node 5, the EMV for this chance node is

given by
0.9 x (800) + 0.1 x (-3300) = 390
node 13 chance node 7
Hence the EMV for the crash program decision is 390K.

Consider chance node 6 (with branches to terminal nodes
18, 19 and 20 emanating from it). The EMV for this chance
node is given by
0.8 x (-3000) + 0.1 x (-2000) + 0.1 x (-1000) = -
2700
Then for the decision node relating to whether to do a
crash/normal/no program we have the three alternatives:
 crash program EMV = 390 (calculated above)
 normal program EMV = 600 (terminal node 17)
 no program EMV = -2700 (calculated above)
It is clear that, in £ terms, the normal program alternative
is the most attractive alternative and so we can discard the
other two alternatives, giving the revised decision tree
shown below.
We can now continue the process. The EMV for chance
node 4 is given by
0.6 x (2250) + 0.3 x (1250) + 0.1 x (250) = 1750
The EMV for chance node 2 is therefore given by

0.5 x (850) + 0.5 x (1750) = 1300
node 9 chance node 4
Hence the EMV for the decision "X invests 750K" is

1300K.
The EMV for chance node 3 is given by
0.5 x (600) + 0.5 x (0) = 300
program decision node node 21
Hence at the initial decision node relating to whether to

invest 750K or not we have the two alternatives
 invest 750K EMV = 1300
 don't invest 750K EMV = 300
Therefore our recommendation is that X should invest
£750K now with an expected monetary value of £1300K.
Note however that should we follow this recommendation
the actual monetary outcome will be dependent upon
chance events but will be one of [850, 2250, 1250, 250]
corresponding to terminal nodes 9, 10, 11 and 12
respectively. Therefore, based on our figures, we are
always going to show a profit of at least 250K on our
New PrecisionTree® 5.7

PrecisionTree performs decision analysis in Microsoft Excel using decision
trees and influence diagrams. Decision trees let you visually map out
complex, multi-layered decisions in a sequential, organized manner. This
helps you identify all possible alternatives and choose the best
option.
New PrecisionTree 5.7 Is fully compatible with 64-bit Excel 2010.

64-bit technology enables Excel and PrecisionTree to access much more
computer memory than ever before. This allows for vastly larger models and
greater computational power.
PrecisionTree 5.7 has also been fully translated

into Spanish, German, French, Portuguese, Japanese, and Chinese.
» What’s New in PrecisionTree 5.7
What’s New in PrecisionTree

Getting Started in PrecisionTree

Facility Location and Decision Tree

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Facility Location and Decision Tree

Загружено:

Авторское право:

Доступные форматы

Facility Location

Factors affecting the location decisions are:

Methods of Location analysis:

It is a mathematical model that is used to evaluate locations based on proximity factors.

𝑑𝐴𝐵 = √(𝑋𝐴 − 𝑋𝐵 )2 + (𝑌𝐴 − 𝑌𝐵 )2

𝐷𝐴𝐵 = |𝑋𝐴 − 𝑋𝐵 | + |𝑌𝐴 − 𝑌𝐵 |

Center of Gravity Method:

A good starting point for evaluating different locations in the

A decision tree consists of 3 types of nodes:-

1. Decision nodes - commonly represented by squares

The decision tree for the problem is shown below.

 path to terminal node 9 - X invests 750K, Y improves,

Consider chance node 7 (with branches to terminal nodes

Consider chance node 5, the EMV for this chance node is

Hence the EMV for the crash program decision is 390K.

The EMV for chance node 2 is therefore given by

Hence the EMV for the decision "X invests 750K" is

Hence at the initial decision node relating to whether to

New PrecisionTree® 5.7

New PrecisionTree 5.7 Is fully compatible with 64-bit Excel 2010.

PrecisionTree 5.7 has also been fully translated

» What’s New in PrecisionTree 5.7

What’s New in PrecisionTree

Вам также может понравиться