THE INFLUENCE OF A
MATHEMATICAL MODEL IN PRODUCTION STRATEGY: CONCEPTUAL DEVELOPMENT AND EMPIRICAL
TEST
Dn. Paulo César Chagas Rodrigues
Instituto Federal de Educação, Ciência e
Tecnologia de São Paulo - Brazil
E-mail: paulo.rodrigues@ifsp.edu.br
Dr. Fernando Augusto Silva Marins
Universidade Estadual
Paulista (FEG/UNESP) - Brazil
E-mail: fmarins@feg.unesp.br
Dr. Fernando Bernardi de Souza
Universidade Estadual
Paulista (FEB/UNESP) - Brazil
E-mail: fbernardi@feb.unesp.br
Submission: 15/04/2012
Accept: 20/05/2012
Acquire and produce what is strictly
necessary are the goals of the organizations, since they aim companies more competitive
and thereby reducing production costs. The research method is applied in
nature, with a qualitative and quantitative approach, in which the objective of
the research will be: exploratory and descriptive, with technical procedures,
divided into: bibliographic, documentary, survey and concluding with a case
study. On this assumption the main objective of this research is to develop and
analyze a mathematical model that minimizes costs and maximizes the
postponement of stocks in a company in the pulp, paper and paper products.
Having been found only four papers, two articles and two theses that deal with
the issue of demand management, supply chain and inventory postponement. These
studies address the issue by modeling the productive time of the supply chain.
For production segments this research may enable development of management
practices demand and production strategy, allowing cost reductions and
productivity gains possible. With the development of the mathematical model
could ever analyze the behavior of demand and its influence on the productive
strategy, strategy formulation regarding the purchase of raw materials and
finished product storage in the last four years the company's results for the
proposed model.
Keywords: Production
strategy, demand management, mathematical modeling, S&OP.
1 INTRODUCTION
From
the event RIO +20 in 2012, organizations need to spend worrying not only to
reduce production costs, but to be optimizing the consumption of raw materials
in order to acquire what is strictly necessary to its production.
The
Brazilian business environment has undergone significant changes in recent
decades, these changes increase the competitive level of demand from customers,
global economic and political moments and the dynamic marketing increasingly
unstable, creating increasing challenges for companies in understanding
demands.
From
this context, management models and operations were suffering adjustments,
considered pioneers by companies as a way to adapt to the new political and
economic world, becoming a focus of academic study and application by other
companies.
Assuming
that companies need to be more competitive when compared to competitor and
market consumer, they are always looking for new ways to reduce production
costs and increase their competitiveness. The study of strategies for
production and mathematical modeling as a way to develop a model that allows to
reduce production costs and the storage of raw materials, process and in the
finished product.
Modeling is a method of quantitative
research of an applied nature, refers to research on models of causal
relationships between control variables and performance variables which can be
developed, analyzed or tested (BERTRAND; FRANSOO, 2002).
Lawrence
and Pasternack (2001) discuss the mathematical programming is the branch of
science that deals with the management of optimization problems, in which we
want to maximize a function (such as the return of profit, expected, or
efficiency) or minimize a function (such as cost, time or distance), usually in
a restricted environment.
According
to Goldberg (2000), the need to represent the context of reality simply brought
forth the concept of modeling, which is defined as a process of finding a
well-structured vision of reality, models are simplified representations of
reality, to certain situations and approaches.
Starting
from these premises the aim of this research is to develop a mathematical model
that will maximize the deferral of inventory, minimizing costs, for both this
modeling is based on the study of manufacturing strategies and mathematical
modeling.
Ko, Mehnen and Tiwari
(2010) add that a mathematical structure allows us to treat and represent
uncertainties in the self-perception of uncertainty, imprecision, partial
truth, and lack of information.
Allen and Schuster (2004)
add that mathematical programming is often the tool used for modeling
strategies in conditions of uncertainty. These models are static in nature,
assuming that the probabilities of occurrence are known beforehand and that all
decisions are taken at the same time.
2 DYNAMIC Stochastic Programming Model
According to García-Dastugue (2003),
the demand process, the fundamental equations of the model of
safety stock, and the optimization problem, are described below.
Graves and Willems (2000) and
García-Dastugue (2003) discuss that process in external demand is observed stages of demand that is
based on a stationary process with
a daily average (μ). Demand for
internship, are all
steps that are not considered stages
of demand are the sum of the demand of each customer
multiplied by the value of the corresponding arc.
According to García-Dastugue (2003), a way to produce the upper limit of demand is assumed that demand is normally distributed and uses the average demand, standard
deviation of demand and factor of safety
to produce the upper limit according to Equation 1.
|
(1) |
Onde:
Upper limit of demand over
days of exposure in stage j
Days of Exposure
Safety factor
Standard deviation of demand
Graves and Willems (2000)
and Garcia-Dastugue (2003)
comment that the number of days of
exposure (t) depends on time of
processing stages, the input
service time is the maximum time it takes replenishment from
all providers and
service time quoted is the time promised
to the customer the
next layer, as stated in Equation
2.
|
(2) |
Onde:
Days of Exposure
Weather service entry
Stage processing time
Service time quoted to the next level of customer
The inventory model of
security can be defined as the expected
inventory level on a stage
at the end of each period is the safety stock planned
for the stage. At the end of each period, it is expected that the stock
throughout the cycle has been used. Equations
3:04 from the
expected inventory level at the end of stage time
t j (GRAVES;
WILLEMS, 2000; GARCÍA-DASTUGUE, 2003).
|
(3) |
|
(4) |
The authors complement discoursing
expected that the inventory at the end of period t of
stage j is equal to the upper limit of the average demand least during the exposure
time.
According Garcia-Dastugue (2003), the optimization problem is designed to find
optimal inventory levels, i.e. the level of said stock
to customers the next level, as
described in Equations 5, 6, 7, 8 and 9.
|
(5) |
|
(6) |
|
(7) |
|
(8) |
|
(9) |
Onde:
P Postponement
Upper limit demand over time exposure
Rate of annual insurance cost
Time service entry
Processing Stage Time
Quota-time service to the next level of customer
Average demand expected
Share of maximum service to customer’s next level
Rogers, Ribeiro and
Rogers (2004) report
that in some environments, especially
in manufacturing, the application does not come all at once as the
EOQ assumes, but
in fractions, and this premise, sets up what should be
the minimum volume being
produced, at which production
costs does not exceed the total
cost.
According to García-Dastugue (2003), the optimization problem is to minimize
the sum of the cost multiplied by the inventory level at all stages of the supply chain.
2.1 Definitions
In
this model, the objective function is to minimize the cost and maximize the
delay inventories of raw, in-process and final product. The constants that will
permit the implementation of the objective function are: Demand (Dj), the
proportion of each product being made on a monthly basis, the volume of paper
used in each of the products in grams (Y1), ink volume in milliliters (Y2) ,
volume of wire in meters (Y3), envelope in units (Y4), foil stickers on units
(Y5), Cover and cover units (Y6), processing time in minutes (Y7) and cost
(Y8).
2.2 Decision variables
The
decision variables used in this model are: number of books of 400 sheets (X1)
and 200 sheets (X2), because these two variables will display the quantity to
be produced of each product each month and will together with the constants Y1
, Y2, Y3, Y4, Y5, Y6, Y7 and Y8 allow minimize the production costs, while
maximizing the delay inventories.
The
variables X1 and X2 will be influenced by the estimated demand as per the
information of the last quadrennium, which will influence the decision about
what to buy and in what quantity.
From
this influence can be observed the behavior of manufacturing strategy as the
strategy of purchasing, production and storage by the proposed model.
2.3 Constraints
Equation 10 is the restriction on the sum
of the quantity to produce times
the volume of paper lined product during each of
the periods j = j
+1 which is
the total consumption of paper in
time, measures should be in weight units and
(kilograms) which should be less than or equal to the sum of the respective periods, which will inform
the total quantity of paper to be
purchased.
|
(10) |
The restriction referenced in
Equation 11 is the sum of the quantities produced during periods j = j
+1 which will
unit of measurement units, which
should be less than or equal to the sum
of the units to produce the respective periods.
|
(11) |
Equation 12 wherein
the restriction reference is made sum of the
periods j = j
+1, products X1 and X2, which is calculated on the amount of product
produced times the total amount of
ink used in liters in each of
periods which should be less than or equal to the sum of the periods in
which 12 were purchased inks.
|
(12) |
The constraint of Equation 13
is calculated the sum of the periods j = j
+1, products X1 and X2, which is calculated on the amount of product
produced times the total amount of
wire used for each of the periods
which must be less or equal to the sum of the periods in which 12
were purchased inks.
|
(13) |
In Equation 14 is treated
restriction which is the sum of the capacity in each
period j = j +1, which should be less than or equal to the sum of the demands of each of the 12 periods.
|
(14) |
Equation 15 discusses the
restriction which is the sum of the units produced in each period j = j
+1, which should be less than or equal to the sum of productive capacity
in each of 12 periods.
|
(15) |
The restriction referenced in
Equation 16 states that the
sum of the quantities of product produced X1 must be
greater or equal to 0 in each period j = j
+1. This restriction is designed to force that exists in producing the product X1 and it is not negative.
|
(16) |
The restriction referenced in
Equation 17 states that the
sum of the quantities of product produced X2 should be
greater than or equal to 0 in
each period j = j
+1. This restriction is designed to force that exists in producing the product X2 and it is not negative.
|
(17) |
The restriction treated in
Equation 18 is the sum of units of the product produced
X1 must be greater than or equal to units of
each product to be produced, or is
the division of the units to produce
the period from the sum of the proportion of each product and result multiplied by
the proportion of the units to produce
X1.
|
(18) |
In Equation 19 discusses the
restriction which will be the sum of units of the
product produced X2 should be greater than or equal to units of each product to
be produced, or is the division of
the units to produce the period
from the sum of the proportion of each products and
the result multiplied by the proportion of the units to produce X2.
|
(19) |
In Equation 20 states that
the restriction is the sum of the products produced units X1 and X2 must be
less than or equal to units produced downstream of the
front and back in each period
j = j +1.
|
(20) |
In Equation 21 is the restriction on the sum
of the costs of storage must be greater
or equal to 0 in each
period j = j +1.
|
(21) |
Equation 22 is about the
restriction on the sum of the percentage of safety stock of products X1 and
X2 to be produced in each period which must be
greater than or equal to the sum of
the percentage overall safety stock of each
period.
|
(22) |
Onde:
P Postponement.
Storage cost.
Cost of production of product 1.
Cost of production of product 2.
Quantity producing product
1.
Quantity producing product
2.
Stock Product Safety 1.
Stock Product Safety 2.
Total weight of the leaves that
will make the product 1.
Total weight of the leaves that
will make the product 2.
Total liters of paint used
in product 1.
Total liters of paint used
in product 2.
Length in meters of
wire used to make
the product 1.
Length in meters of
wire used to make
the product 2.
Total units to be produced.
Production capacity in the period.
Actual demand in period.
Quantity to buy of Wire in the period in meters.
Ink Supply to buy in the period in liters.
Quantity to buy paper in the period in Kilograms.
Product
1.
Product
2.
Units producing Downstream Cover.
2.4 Objective
function
From
this model we intend to get a sense of the costs of production, watching what
and how much to buy and creating policies replenishment of inventories of raw
materials and finished product.
It
is intended to maximize the delay, minimizing the costs of production and
storage of stocks of raw material and finished product and maximize the
postponement of inventories of raw materials, in process and finished products,
this model will be tested with more than one type, which will be exemplified in
Equation 23.
|
(23) |
3 Illustrative example
Table 1 shows the items to be produced, which were defined as notebook
sheets 400 and 200,
the Constant paper (Y1) means the amount of paper being used measurement unit used was kg, paints (Y2) means amount
of ink to be used and the unit of
measure used was liters
wire (Y3) means
the quantity of wire to be used
and the unit of measure used was meters envelope (Y4),
adhesives (Y5), cover (Y6) signify amount was used
of these products and the unit of
measure used parts, time (Y7) means the production time of a unit of each product and unit of measure used was minute and finally cost
(Y8) means that the cost of producing a unit of each product and this expressed in Real.
Table 1: Items to produce
Table 2 deals with the inventory levels of product
safety Notebook 400 and 200 sheets, which are expressed in percentages during the 12 months since the volumes of safety
stocks may vary, depending on the volume to be produced in the month.
Table 2: Safety stock of items to be produced in%
Table 3 shows the proportion of units to be produced per month of each product notebook sheets 400 and
200, this ratio may vary as required monthly to
be produced from each template.
Table 3: Proportion of units to be produced
per month
Table 4 shows the
amount of products to be produced
Notebook 400 and 200
sheets, in the course of 12 months, this table there
are some constants that must be
nurtured by the management of the
company, in which we have storage
cost (Cj), demand
(dj), production capacity
(CPj), the units produce (UPj)
and units due to
the customer (UCj), and
based on this information in Tables 4,
5 and 6 will generate information on
the expected purchases of raw materials , this
volume already being planned margin of
safety stock.
Table 4: Proportion of units to be produced
per month
4 application in the
graphic industry
This model was
tested with two types of products and raw materials that compose them; it is not impediment to the inclusion of lines with new products in Table 1 which also will receive new columns to the raw materials that compose the new products, if necessary.
Table 2 also will be adding lines with new products because if it is necessary to define a margin for safety stock for the new product. As Table 3 is to be produced the proportion of each product per period, should also be included in the new product line.
Table 3 should receive more than one column must reference the new product, this column will be the variable
that will show the quantity to be produced from the new product compared to others.
Shares in Tables 1, 2 and 3 are necessary because they serve as a parameter to define the
values that will power the
speakers' QTY X1, X2 QTY, QTY X3 ... ".
After these actions, should be reviewed in the objective
function for the new product as well as the creation of constraints that will enable the implementation of the model and present the
results in variables.
This model can be run in Excel since it included the tool that was used for the tests, the software plug-in Risk Solver Platform company Frontline Systems, which proved easy interaction, applicability
and generation of reports, but there is other software that
allows the execution of the model in Excel, for example, the software company Oracle Crystal Ball.
5 Final Thoughts
Because it is a constant concern for
the company to remain competitive. Able to participate in the development of a
mathematical model that reduces storage costs and production and at the same
time observing the level of service demand as a way to increase profitability.
According
to the Board of Industrial and manages the company's PCP when inserting the
data model generated from previous years and the graphics of the results
obtained, could ever observe the behavior department supplies relative to
demand for the product spiraled notebook of 400 and 200 sheets.
Data
were entered in a monthly model for the last four years, which allowed us to
observe the behavior of demand and its effect on supply chain management.
This comparative analysis of the behavior of demand and supply chain over the
past four years, with the result that the model was allowed to observe the
times the storage costs were above and below the model showed.
This
analysis allowed us to observe the possible periods which increased the cost of
storage, with an output higher than necessary and consequently a volume storage
of raw material and finished product high.
Also
allowed to observe the periods in which the volumes of stocks of finished
products and raw materials were below the necessary and which consequently
generated some kind of cost for loss of market.
According
to board the mathematical model can support decision making as to what and how
much to produce and when to start and end the production, they may add to the
model the four product lines and observing the proportion to be produced for
each product line months the month.
The
model will also examine the quantities of raw materials purchased in the course
of a year, observing the times of high prices, production and consumption of
the finished products. This will support purchasing decisions and delivery.
Allowing suppliers to organize and participate on the volumes to be delivered
and their respective dates.
Also
according to the board model was easy to understand and applicability also
allowed to observe more clearly the monthly performance as the cost of
production and storage of the product spiral notebook.
Another
factor that the company found interesting was the development of the
mathematical model on the platform and Excel with the ability to use other
solutions that complement Excel, for example: Risk Solver Platform and Crystal
Ball.
REFERENCES
ALLEN S.
J.; SCHUSTER, E. W. (2004) Controlling the risk for a agricultural harvest. Manufacturing
& Service Opertions Management, v. 6, n. 3, p 225-236.
BERTRAND, J. W. M.; FRANSOO, J. C. (2002) Modeling and
Simulation: Operations management research methodologies using quantitative
modeling. International Journal of
Operations & Production Management, v. 22, n. 02, p. 241-264.
GARCÍA-DASTUGUE,
S. J. (2003) Dynamic time-based postponement: Conceptual development and
empirical test. Ph.D, Ohio:
Ohio State University.
GOLDBERG,
D. E. (2000) Genetic algorithms in search, Optimization and Machine
Leaning. Addison Wesley
Limited.
GRAVES,
S. C.; WILLEMS, S. P. (2000) Optimizing strategic safety stock placement in
supply chains. Manufacturing and
Service Operations Management, v. 2, n. 1, p. 68-83.
GRONDIN,
F. (1998) Improvements of the dynamic programming algorithm for tree bucking. Wood and Fiber Science, v. 30, n. 1, p. 91-104.
KO, M.; TIWARI, A.; MEHNEN, J. (2010) A review of soft
computing applications in supply chain management. Applied Soft Computing. V. 10, p. 661-674.
LAWRENCE, J. A.; PASTERNACK, B. A. (2001) Applied management science: modeling, spreadsheet
analysis, and communication for decision making. 2 ed. New Jersey: Wiley.
ROGERS, P.; RIBEIRO, K. C. S.; ROGERS, D.
(2004) Avaliando o risco na gestão financeira de estoques. In: SIMPOSIO DE ADMINISTRAÇÃO DA PRODUÇÃO, LOGÍSTICA E OPREAÇÕES
INTERNACIONAIS, 7., São Paulo. Anais… São Paulo: FGV, 1 CD-ROM.