MIXED PRODUCTION AND SCHEDULING MODELS OF

PRODUCTIVE OPERATIONS OF LINEAR PROGRAMMING

PRODUCTION MIX
It is a type of models that appear, for example, in those situations where
which a company has available for a given period of quantities
limited biof resources (raw materials, machinery, money, skills
humans, etc.), which it uses for the production of n types of article, each
of which uses the resources in a certain combination. That is to say, they are known.
the values toijwhich indicate the number of units of resource i needed
to produce one unit of item j. The values c are also knownj
What are the unit utilities (or unit costs) of product j.

The problem then consists of determining the quantity Xj to be produced of

each of the articles that compete for the use of resources, such that
a way to achieve maximum production or maximum profit, or
a minimum cost or another special objective, within a specified period of time
production
Let's illustrate this type of problems with two simple examples:

EXAMPLE:
A small company manufactures items of two types from three materials.
cousins, called A, B, C. The type 1 article produces a profit of $400 per
unit, and for its production one pound of A, one pound of B is required.
three grams of C. The type 2 article produces a profit of $300 per unit, for
whose manufacturing requires one pound of A, two pounds of B, and two grams of
C.
The company has 150 pounds of A, 240 pounds of B, and 420 grams of C.
for the next production period (it can be an hour, a day, or another)
lapse).
The company wants to know how many units of each type of item it should
produce in the period with the aim of maximizing total utility per sale of the
articles. It is assumed that all produced articles are sold and that the
unitary utility remains constant, regardless of the amount sold.

Construction of the model:

Following the proposed methodology, once the situation is understood
It is described, we are going to organize the data in a table; which will make it more
easy to use for building the model.
TYPE OF CONSUMPTION OF RAW MATERIAL PRICE OF
ARTICLE A B C SALE
P1 1 1 3 400
P2 1 2 2 300
Available quantity 150 240 420

A sketch of the situation may be as shown in the following

graph, in which it is observed that the key elements of the problem are the
three raw materials and the two types of items, while the goal is
maximize utility. In this way, it becomes clear that the objective is
will be measured in pesos / period.
It is equally apparent that alternative activities are, as they
it says the statement, produce type 1 articles and produce type 2 articles.
Please note that the activities are not exclusive but can
to take place simultaneously at certain levels; from being carried out a
only of them, until both are executed in a certain combination. All of this
it depends on the relationship between your contribution to the goal and your consumption of
the raw materials.

We must use some variables to quantify the level or degree to which

we will carry out each activity.

Variables:
X1: number of type 1 items to be manufactured in the period.
X2: number of type 2 items to be produced in the period.

Objective function
Total utility = 400X1 + 300X2

Restrictions
1X1 + 1X2 ≤ 150
1X1 + 2X2 ≤ 240
3X1 + 2X2 ≤ 420
X1 ≥ 0
X2 ≥ 0

But with the purpose of illustrating how to verify compatibility of

units, which with the verification of the model's accuracy, as this is very
elemental, we are going to analyze a two-dimensional analysis of the functions.
In the objective function:

400 ($ / unit of P1) * X1 (units of P1 / period)

+ 300 ($ / unit of P2) * X2 (units of P2 / period) = ($ / period)

It is verified that they are the same units of the objective function.
In the restrictions, let us analyze only for the consumption of the material.
first A, since the others are similar.

1(pound of A/unit of P1) * X1 (units of P1/period)

+1 (pound of A/unit of P2) * X2 (units of P2/period) = (pounds of A/period)

That match the units of the right member of the first.

inequality.
(150 lb. per period)

Note that the non-negativity condition of the variables makes sense.

logical in this model, since a negative quantity cannot be manufactured
some of the types of the article.

We must also interpret the part of the statement referring to that the
the company assumes that it sells all the produced items and that the profit
remains constant.

The first part indicates that there are no limits on the number of
articles of each type to be produced, except those related to availability of
the productive resources. The second part allows us to consider the
proportionality between the number of articles and total profit from sales
each type of article.

In some market situations, there are limits on sales.

maximums or minimums of a certain type of article, which must
to be reflected in the model as a constraint on the value of the variable
corresponding to the activity.
For example, if we know that the maximum demand for item 1 is 30.
units, we must include in the problem the condition that they do not occur
more than 30 units of this item, since the additional production does not
would have a buyer, thus generating an inventory instead of contributing to
the total utility. Similarly, if we have a committed demand of 15
units of Article 2, which we wish to satisfy, we must add to the
model a constraint expressing that the number of units of item 2
it must be at least 15.

With the two previous considerations, the model must have these others.
two restrictions:
X1 ≤ 30
X2 ≥ 15

On the other hand, in some market conditions, the unit selling price
of the articles is "broken", that is, it decreases by some amount when
the number of items exceeds a certain value. For example, if X3 is the quantity
purchase of an item with a unit price of $10 when 0 ≤ X3 ≤ 100;
but it goes down to $7 when 101 ≤ X3 ≤ 250 and goes down again to $6
when X3 ≥ 251; in this case the condition is not met.
proportionality in the objective function since we do not have a coefficient
unique to multiply by the quantity purchased to obtain the cost of
purchase of the X3 units. The coefficient depends on the interval in which it is
find the value of X3. In some cases we can take averages of
the values of the relevant parameters and in others it can be raised the
model of the separable linear programming form, as we will learn
in a later example.

EXAMPLE:
A company produces items of three types, performing operations C,
F, T. the operation C machine costs $1500/hour of operation, the
the cost of operation F is $2400/hour and the cost of operation T is
$1200/hour.

The cost of the material for one unit of item 1 is $50, for one unit
the price of item 2 is $80 and for one unit of item 3 it is $140.
The sale prices for the items are respectively $400, $420 and
$500 each.
The processing times required by a unit of each type of material,
they are given in the following table:

Minutes of operation per unit:

TYPE OF ARTICLE C F T
A1 2.5 2.0 4.0
A2 2.5 1.0 2.5
A3 2.0 0.5 2.0
The company needs to know how many units of each type of item
It must be produced in one hour to achieve maximum profit.

Construction of the model:

Initially, we can create some tables with the problem data, as follows:
MACHINE C F T
Operating cost ($/minute) 2 4 20
5 0

ARTICLE 1 2 3
Cost of the material 50 80 140
($/unit)
Selling price ($/unit) 400 420 500

The process scheme may be as follows:

The elements are the raw materials, the three process operations, and the
three articles. Determining the elements of the problem is very helpful for
the formulation of the objective function and the functions of the three constraints.
The objective will be to maximize the utility resulting from production in one hour.
This utility will be the difference between sales income and expenses for
raw material and operations on the machines.
Let's calculate the net profits like this:

Construction of the exercise:

Maximize:
Z = 127.5X1 + 187.5X2 + 250X3

Restrictions:
2.5X1 + 2.5X2 + 2.0X3 ≤ 60
2.0X1 + 1.0X2 + 0.5X3 ≤ 60
4.0X1 + 2.5X2 + 2.0X3 ≤ 60
X1. X2, X3 ≥ 0

The restrictions refer to the fact that a machine cannot be used during
an hour for a total time greater than the hour. That is, the time that a
machine dedicated to the production of item 1, plus the one dedicated to
Article 2 plus the one dedicated to Article 3 cannot exceed one hour of
capacity, as that is the period of time that was used as a reference.
PRODUCTION SCHEDULING
This is one of the areas of Linear Programming richest in applications.
A production scheduling problem can be seen as a
production mix problem for multiple periods into the future. It
they want to determine the production levels that will allow the company
obtain the minimum cost (or the maximum profit), complying with the
requirements of the limitations in labor, machinery, materials,
storage space, demand requirements, etc. The problems of
Production scheduling has a recurring nature, that is to say that it
they present one period after another, only with some variations in certain
data, such as in claims, or in the availabilities of
some resources. For this reason, P.L. models are used
extensively in this field, since once a model was solved
for a certain period, it is enough to repeat its solution for the data of
new period, to obtain recommendations about the optimal program
of production.
Let's illustrate this type of applications with the following simple example:

EXAMPLE:
A manufacturer must fulfill the following commitments, in the first
quarter

The monthly production capacity of your plant is 20,000 units.

The unit production cost varies each month, as follows: January $20, February $9 and
March $12. The company estimates the storage cost at $3.
each unit held in the warehouse on the last day of the month. The capacity of
The warehouse it has is for 22,000 units.
The company has 50 units in inventory and wants to have 70 at the end.
the problem to solve consists of determining the production schedule
monthly that minimizes total costs in the quarter.
It is supposed that the production takes place throughout the month and the dispatch is...
it takes place on the last day of the month.

Construction of the model:

The graphic that describes the problem can be:
We wish to determine the production program to achieve the minimum.
cost in the quarter. To do this, we define the variables as follows:

Variables:
Pinumber of articles produced in month i (Jan, Feb, Mar).
IFiunits in the final inventory of month i.

Minimize:
Z = 10(XE) + 9(XF) + 12(XM) production cost
+ 3(IFE + IFF + IFM) storage cost

Restrictions:
Production capacities per month
-January: PE ≤ 20,000
-February: PF ≤ 20,000
-March: PM ≤ 20,000

Committed shipments every month

-January: PE= 10,000 + IFE
-February: IFE + PF = 30,000 + IFF
March: IFF + PM = 20,000 + 70

Warehouse capacity
PE ≤ 22,000
-February: IFE + PF ≤ 20,000
March: IFF + PM ≤ 22,000

Problems of this type can also be modeled in another way as

it is suggested by the following graph:

Here the variables are defined as:

Sean Xijnumber of articles produced in month i destined for the
sales of the month j (i= E, F, M and j= E, F, M).
In this way, the final inventory of each month is made up of the
quantities produced that month intended for the following months.
Minimize:
Z = 10(X11 + X12 + X13) + 9(X22 + X23) + 13(X33 + X34) + 3(X12 + X23 +
X34) + 6(X13)
Note how the values (X12 + X13 and X23 + X13) are equivalent to the inventories.
the end of the months of January and February.

Restrictions:
Production capacities per month
January: X11 + X12 + X13 + X14 ≤ 20,000
February: X22 + X23 + X24 ≤ 20,000
March: X33 + X34 ≤ 20,000

Committed shipments each month

January: X11 + 50 = 10,000
February: X12 + X22 = 30,000
March: X13 + X23 + X33 = 20,000

In this formulation, we do not include the possibility of having inventory.

initial nor with final inventory, which is left as an exercise for the student.
For example, what would be the production schedule if there are 5,000?
initial inventory units and 10,000 of final inventory are desired.
The production programming problems they can also
to be solved in a relatively simple way by arranging the data in a
table, in which the costs of each alternative action are noted and
proceed to obtain a solution taking advantage of the characteristic structure of the
process. The end of the chapter will resolve this same problem through the
mentioned table.
SOURCE:[Link]
Operations Research Modeling. Applications of Programming
Linear.
Rafael Freites

UNIVERSITY OF VALLEY

FACULTY: ADMINISTRATIVE SCIENCES

PROGRAM: BUSINESS ADMINISTRATION
V

Operations Research
TEACHER: MARTHA PATRICIA ROA

STUDENTS: LEIDY VIVIANA MAYORAL,

MARIA ANDREA TABARES
Fabian Valencia Acosta