Using genetic algorithms and simulation as decision support in marketing strategies and long-term production planning

Proceedings of the 9th WSEAS International Conference on SIMULATION, MODELLING AND OPTIMIZATION
Using genetic algorithms and simulation as decision support in
marketing strategies and long-term production planning
FLORIN STOICA
Computer Science Department
Faculty of Sciences
University “Lucian Blaga” Sibiu
Str. Dr. Ion Ratiu 5-7, 550012, Sibiu
ROMANIA
florin.stoica@ulbsibiu.ro
LAURA FLORENTINA CACOVEAN
Computer Science Department
Faculty of Sciences
University “Lucian Blaga” Sibiu
Str. Dr. Ion Ratiu 5-7, 550012, Sibiu
ROMANIA
laura.cacovean@ulbsibiu.ro
Abstract: - This paper represents an approach of using simulation models and genetic algorithms for generating an
aggregative production plan to maximize the total profit of the firm. The described methodology provides a tool which
assists the company management in decision which of the suitable solutions will become the production plan. The
entire system is composed by a business information system - a database, a simulation model and a genetic algorithm.
The purpose of the integrated system is to help operative management personnel to take decisions with respect to long-term
production planning and marketing strategies.
Key-Words: - Production planning, Genetic algorithms, Co-mutation
1 Introduction
Most real life optimization and scheduling problems are
too complex to be solved completely. The complexity of
real life problems often exceeds the ability of classic
methods. In such cases decision-makers prepare and
execute a set of scenarios on the simulation model and
hope that at least one scenario will be good enough to be
used as a production plan.
A long time goal for scheduling optimization research
has been to find an approach that will lead to qualitative
solutions in a relatively short computational time. The
development of decision-making methodologies is
currently headed in the direction of simulation and
search algorithm integration. This leads to a new
approach, which successfully joins simulation and
optimization. The proposed approach supports man-machine
interaction in operational planning.
A group of widely known meta-heuristic search
algorithms are genetic algorithms (GA). Evolutionary
algorithms are a very effective tool that enables solving
complicated practical optimization problems. An
important characteristic of evolutionary algorithms is
their simplicity and versatility. Their main drawback is
a long calculation time, which, however, is not a serious
disadvantage nowadays with advanced computer
technology and does not limit their use for searching for
almost optimal solutions, if are not real-time restrictions
[2].
With computer imitation of simplified and idealized
evolution, an individual solution-chromosome represents
a possible solution to our problem. Chromosome fitness
is calculated with a fitness function. After being
evaluated with a fitness function, each chromosome in
population receives its fitness value.
The optimization is based on a genetic algorithm
which uses a new co-mutation operator called LR-Mijn,
capable of operating on a set of adjacent bits in one
single step.
In present, there is a major interest in design of
powerful mutation operators, in order to solve practical
problems which can not be efficiently resolved using
standard genetic operators. These new operators are
called co-mutation operators. In [7] was presented a co-mutation
operator called Mijn, capable of operating on a
set of adjacent bits in one single step. In [12] we
introduced and studied a new co-mutation operator
which we denoted by LR-Mijn and we proved that it
offers superior performances than Mijn operator.
The paper is organized as follows. In Section 2 we
make a brief presentation of the architecture of our
decision-support system, called GA-SIM. We also
present the basic idea of evolutionary method adopted
for optimization process. In section 3 is introduced the
co-mutation operator LR-Mijn. The evolutionary
algorithm based on the LR-Mijn operator, used within the
decision-support system is presented in section 4.
Section 5 contains the description of the simulation
model and its integration in the GA-SIM system.
Conclusions and further directions of study can be found
in section 6.
ISSN: 1790-2769 435 ISBN: 978-960-474-113-7

2 The architecture of the decision-support
system
Companies need to be flexible to compete for the market
share and adapt to market demands by offering
competitive prices and quality. This calls for a wide
assortment of products or product types, small
production costs, etc. Simulation is a strong interactive
tool that helps decision-makers improves the efficiency
of enterprise actions. The ability of simulation to show a
real process on the computer with the consideration of
uncertainty is a big advantage when analyzing system
behavior in complex situations [15].
Our system is composed by a business information
system - a database, a simulation model and a genetic
algorithm, and is called in the following the GA-SIM
system.
A simulation model will be used for fitness function
computation of genetic algorithm results, as well as for
visual representation of qualitative evaluation of a
chosen production plan following genetic algorithm
optimization.
The value of the fitness function is computed by the
simulation model, with uses the respective chromosome
as input data. The value returned from the simulation
model is used to evaluate that chromosome, which
encodes a possible production plan.
By applying the genetic operations on the members of
a population of decisional rules at the moment t, it
results a new population of rules that shall be used at the
moment t+1. The population from the initial moment, t =
0 is generated randomly and the genetic operations are
applied iteratively until the moment T (at which the
condition for stopping the algorithm is accomplished).
The above iterative process may be interpreted
economically as it follows. The evolution process has as
objective the finding of the successful individuals. The
binary strings of these individuals (chromosomes) with
high fitness values (a high profit) will be the basis for
building a new generation (population). The strings with
lower fitness values, which represent decisions to
produce with a low profit, find few successors (or none)
in the following generation.
The purpose of the integrated system is to aid
operative management personnel in production planning
and marketing strategies (incorporated in simulation
model). The main advantage of the presented system is
to enhance man-machine interaction in production
planning, since the computer is able to produce several
acceptable schedules using the given data and a set of
criteria. The user then selects the most suitable schedule
and modifies it, if necessary.
After completion of the optimization process, the
most suitable production plans are simulated on the
visual model of the system. Using chosen parameters
and according to defined criteria, the decision-maker is
motivated to search for results which will have the most
advantageous influence on the whole production process.
The result, which is selected after simulation on the
visual simulation model, becomes the proposed
production plan
kth generation
Fitness function
Simulation model
New individual
evolution (co-mutation)
Result of fitness
eval. (profit)
Individual
fitness value
Selection
Last
generation
Optimal
production plan
Data
Business
Information
System
Fig. 1 The architecture of the decision-support system
Every chromosome codes o possible production plan.
The quality of a chromosome is represented by the
amount of profit which result from the simulation model
if it receives as input data the production plan coded in
that chromosome.
It is noticed that, through the modification of the
simulation model, it is obtained a more varied general
frame which leads to more diversified suggestions
concerning the decisions that refer to the quantities of
company products that shall be offered on the market.
3 The LR-Mijn operator
In this section we define the co-mutation operator called
LR-Mijn. Our LR-Mijn operator finds the longest
sequence of σp elements, situated in the left or in the
right of the position p. If the longest sequence is in the
left of p, the LR-Mijn behaves as Mijn, otherwise the LR-Mijn
will operate on the set of bits starting from p and
going to the right.
Let us consider a generic alphabet A = {a1, a2, …, as}
composed by s ≥2 different symbols. The set of all
sequences of length l over the alphabet A will be
ISSN: 1790-2769 436 ISBN: 978-960-474-113-7

denoted with Σ = Al.
In the following we shall denote with σ a generic
string, and σ = σl-1…σ0 ∈ Σ= Al, where σq ∈ A ∀ q ∈
{0, …, l-1}. Through σ(q,i) we denote that on position q
within the sequence σ there is the symbol ai of the
alphabet A.
σz
p, j denotes the presence of z symbols aj within the
sequence σ, starting from the position p and going left
right ,
n
and
p i
σ ( , ) specify the presence of symbol ai on
left ,
m
position p within the sequence σ, between right symbols
an on the right and left symbols am on the left. We
right ,
i
suppose that σ = σ(l-1)... σ(p+left+1,m)
p i
σ ( , ) σ(p-right-
left ,
i
1,n)... σ(0).
The Mijn operator is the mutation operator defined in
[7]:
Mijn : σ ∈ Σ, p ∈ {0, … , l-1} → σ’ ∈ Σ’ ⊂ Σ, where p
is randomly chosen
(i) σ = σl-1…σp+nσp+n-1,i σn-1
p, j σp-1…σ0 ⎯⎯⎯→ Mijn
σ’ = σl-1…σp+nσp+n-1,jσn-1
p,i σp-1…σ0, for n < l – p + 1
and
(ii) σ = σ -p
l σp-1…σ0 ⎯⎯⎯→ Mijn σ’ = σ -p
p, j
l σp-1…σ0,
p,k
for n = l – p + 1, with ak ≠ aj randomly chosen in A.
In [12], we introduced and study the properties of LR-Mijn
co-mutation operator
Definition 3.1 Formally, the LR-Mijn operator is defined
as follows:
(i) If p ≠ right and p ≠ l – left – 1,
LR-Mijn(σ)=
⎧
σ σ σ σ
⎪ ⎪ ⎪ ⎪ ⎪
σ σ
σ σ σ σ
⎨
⎪ ⎪ ⎪ ⎪ ⎪
⎩
right ,
i
l p left i p m
( 1)... ( 1, ) ( , )
= − + +
left m
p right n for left right
( 1, )... (0)
,
− − >
,
l p left m p n
( 1)... ( 1, ) ( , )
= − + +
p right i for left right
( 1, )... (0)
,
− − <
σ σ
or for left =
right with probability
, 0.5
left
rightt
right rleft
right n
left i
σ σ
(ii) If p = right and p ≠ l – left – 1,
right ,
i
σ = σ(l-1)... σ(p+left+1,m)
p i
σ ( , ) and
left ,
i
LR-Mijn(σ) = σ(l-1)... σ(p+left+1,m)
right ,
k
p k
σ ( , ) ,
left ,
i
where k ≠ i (randomly chosen).
(iii) If p ≠ right and p = l – left – 1,
right ,
i
σ =
p i
σ ( , ) σ(p-right-1,n)... σ(0) and
left ,
i
LR-Mijn(σ) =
right ,
i
p k
σ ( , ) σ(p-right-1,n)... σ(0), where k ≠ i
left ,
k
(randomly chosen).
(iv) If p = right and p = l – left –1, σ =
right ,
i
p i
σ ( , )
left ,
i
LR-Mijn(σ)=
⎧
σ σ
⎪ ⎪ ⎪
σ σ
⎨
⎪ ⎪ ⎪
σ σ
⎩
right i
p k for left right where k i
( , ) , ,
= > ≠
left k
right k
p k for left right where k i
( , ) , ,
= < ≠
=
, 0.5
,
left ,
i
,
,
or for left right with probability
left
rightt
right rleft
As an example, let us consider the binary case, the
string σ = 11110000 and the randomly chosen
application point p = 2. In this case, σ2 = 0, so we have to
find the longest sequence of 0 within string σ, starting
from position p. This sequence goes to the right, and
because we have reached the end of the string, and no
occurrence of 1 has been met, the new string obtained
after the application of LR-Mijn is 11110111.
The commutation operator LR-Mijn allows long
jumps, thus the search can reach very far points from
where the search currently is. We proved in [12] that the
LR-Mijn operator performs more long jumps than Mijn,
which leads to a better convergence of an evolutionary
algorithm based on the LR-Mijn in comparison with an
algorithm based on the Mijn operator.
In the following, we will consider that A is the binary
alphabet, A = {0, 1}.
4 The evolutionary algorithm based on
LR-Mijn operator
The basic scheme for our algorithm, called in the
following LR-MEA, is described as follows:
Procedure LR-MEA
begin
t = 0
Initialize randomly population P(t) with P elements;
Evaluate P (t) by using fitness function;
while not Terminated
for j = 1 to P-1 do
- select randomly one element among the
best T% from P(t);
ISSN: 1790-2769 437 ISBN: 978-960-474-113-7

- mutate it using LR-Mijn;
- evaluate the obtained offspring;
- insert it into P’(t).
end for
Choose the best element from P(t) and
Insert it into P’(t)
P(t+1) = P’(t)
t = t + 1
end while
end
5 The simulation model
The simulation model is implemented as an Excel
application. It is based on real data provided by the
Business Information System but also on forecasted data
(e.g. sales quantities and values for the following
months). The model take account of many dates and
variables: sales (values and quantities), raw material
costs, discounts, packaging costs, direct & indirect labor
costs, energy cost, depreciation from the rate of
exchange, warehousing and logistic costs, transport
costs, advertising & promotions, etc. The main
workbook contains few sheets, a VBA module, and
results are synthesized in the following pivot table:
Fig. 2 A pivot table from the simulation model
The role of the simulation model in the genetic
algorithm is presented in the Figure 3.
In fact, the simulation model represents the
implementation of the fitness function of the genetic
algorithm, needed in the procedure LR-MEA to evaluate
each member (chromosome) of the population.
The probability to select a certain chromosome
(possible production plan) in the next generation is
related with its performance in simulated conditions (the
profit provided by the simulation model). That profit
determines the fitness value of the respective possible
solution of our optimization problem.
Chromosome
(Possible production plan)
Profit
Fitness function
Simulation
Model
(Excel app)
Evaluation
module
(Java)
JExcelAPI
Fig. 3 Implementation of the fitness function through the
simulation model
Because the GA-SIM system is implemented in Java
as main language, was necessary a bridge between Java
code and the simulation model, implemented in Excel.
Our choice for this purpose was JExcelAPI
(http://jexcelapi.sourceforge.net/), a mature, open source
java API enabling developers to read, write, and modify
Excel spreadsheets dynamically [14].
6 Conclusions and further directions of
study
The purpose of the GA-SIM integrated system is to aid
operative management personnel in production planning
and marketing strategies (incorporated in simulation
model). The main advantage of the presented system is
to enhance man-machine interaction in production
planning, since the computer is able to produce several
acceptable schedules using the given data and a set of
criteria.
The GA-SIM system is fully implemented, and
currently is under evaluation in a big company from
Sibiu, Romania, which was interested in its acquisition.
As a further direction of study we want to compare
the results obtained by using different genetic operators
and to evaluate real codifications of variables, instead of
current binary one.
ISSN: 1790-2769 438 ISBN: 978-960-474-113-7

References:
[1] Jaber A. Q, Hidehiko Y., F., Ramli R., Machine
Learning in Production Systems Design Using
Genetic Algorithms, International Journal of
Computational Intelligence, No. 4, 2008, pp. 72-79.
[2] Kofjač D., Kljajić M., Application of Genetic
Algorithms and Visual Simulation in a Real-Case
Production Optimization, WSEAS TRANSACTIONS
on SYSTEMS and CONTROL, Issue 12, Volume 3,
December 2008, pp. 992-1001.
[3] Radhakrishnan P., Prasad V. M., Gopalan M.R.,
Optimizing Inventory Using Genetic Algorithm for
Efficient Supply Chain Management, Journal of
Computer Science 5 (3), 2009, pp. 233 - 241.
[4] Lo Chih-Yao, Advance of Dynamic Production-
Inventory Strategy for Multiple Policies Using
Genetic Algorithm, Information Technology Journal
7 (4), 2008, pp. 647-653.
[5] De Falco I., An introduction to Evolutionary
Algorithms and their application to the Aerofoil
Design Problem – Part I: the Algorithms, von
Karman Lecture Series on Fluid Dynamics,
Bruxelles, April 1997
[6] De Falco I., Del Balio R, Della Cioppa A.,
Tarantino E., A Comparative Analysis of
Evolutionary Algorithms for Function Optimisation,
Research Institute on Parallel Information Systems,
National Research Council of Italy, 1998
[7] De Falco I, A. Iazzetta, A. Della Cioppa, Tarantino
E., The Effectiveness of Co-mutation in Evolutionary
Algorithms: the Mijn operator, Research Institute on
Parallel Information Systems, National Research
Council of Italy, 2000
[8] De Falco I., Iazzetta A., Della Cioppa A., Tarantino
E., Mijn Mutation Operator for Aerofoil Design
Optimisation, Research Institute on Parallel
Information Systems, National Research Council of
Italy, 2001
[9] Chen J., Using Genetic Algorithms to Solve a
Production-Inventory Model, International Journal
of Business and Management, Vol. 2, No. 2, 2007,
pp. 38-41.
[10] Krishnakumar K., Goldberg D., Control system
optimization using genetic algorithm, Journal of
Guidance, Control, and Dynamics, no. 15(3), 1992,
pp. 735-740.
[11] Zhu X., Huang Y., Doyle J., Genetic algorithms
and simulated annealing for robustness analysis,
Proceedings of the American Control Conference,
Albuquerque, New Mexico, 1997, pp. 3756-3760.
[12] Stoica F., Simian D., Simian C., A new co-mutation
genetic operator, Proceedings of the 9th
WSEAS International Conference on Evolutionary
Computing, Sofia, Bulgaria, May 2008, pp. 76-81.
[13] Vapnik V., The Nature of Statistical Learning
Theory, Springer Verlag, 1995.
[14] Java Excel API, http://jexcelapi.sourceforge.net
[15] Kljajić M., Bernik I., Škraba A., Leskovar R.,
Integral simulation approach to decision assessment
in enterprises, Shaping future with simulation:
proceedings of the 4th International Eurosim 2001
Congress, Delft University of Technology, 2001.
ISSN: 1790-2769 439 ISBN: 978-960-474-113-7

Using genetic algorithms and simulation as decision support in marketing strategies and long-term production planning

More Related Content

What's hot (20)

Viewers also liked (18)

Similar to Using genetic algorithms and simulation as decision support in marketing strategies and long-term production planning (20)

More from infopapers (11)

Recently uploaded (20)

Using genetic algorithms and simulation as decision support in marketing strategies and long-term production planning