Anda di halaman 1dari 8

Developing decision support systems using the weighted product method for house

selection
Heru Supriyono, and Chintya Purnama Sari

Citation: AIP Conference Proceedings 1977, 020049 (2018); doi: 10.1063/1.5042905


View online: https://doi.org/10.1063/1.5042905
View Table of Contents: http://aip.scitation.org/toc/apc/1977/1
Published by the American Institute of Physics

Articles you may be interested in


Merging Pearson Correlation and TAN-ELR algorithm in recommender system
AIP Conference Proceedings 1977, 040028 (2018); 10.1063/1.5042998
Developing Decision Support Systems Using the Weighted
Product Method for House Selection
Heru Supriyono1, a) and Chintya Purnama Sari2
1
Department of Electrical Engineering, Universitas Muhammadiyah Surakarta, Indonesia
2
Department of Informatics, Universitas Muhammadiyah Surakarta, Indonesia
Jl. A. Yani Tromol Pos I Pabelan Kartasura Surakarta Jawa Tengah, Indonesia 57102
a)
Corresponding author: Heru.Supriyono@ums.ac.id

Abstract. Selecting a house may be a complicated task for non-experienced people due to the numerous criteria to
consider and frequently, one criterion is opposed other criteria. The difficulty causes people have to spend longer time for
making decision and consequently lead to an improper decision. However, the house selection problems cannot be solved
by using exact linear mathematical equation and require heuristic process that can be considered as a semi-structured
problem. The computer-based decision support systems can be used as an alternative for helping non-experienced users
to make decision as good as an expert. This paper presents an empirical work of a development of decision support
systems for house selection. The general objective of the research is to provide a web-based decision support systems
based on the weighted product method for house selection. There are eleven main criteria involved in this house selection
computation resulted from interview with respondents. The computation result showed that the weighted product method
is able to rank the given alternatives to aid people in choosing the best alternative.

INTRODUCTION
Choosing a house to live in is not a simple task for many people. There are many aspects or criteria affecting
people in selecting house, including price, land size, building size, number of bedroom, security services, access to
the main street, distance to the work, distance to the social facilities and so on. In addition, some of these criteria are
opposed the other criteria, for example between the price and the size of house. In general, the bigger the house, the
higher the price. However, people have a propensity to have big house but with the lowest price as possible. It is a
multi-criteria decision making or a multi attribute decision making task. Nevertheless, it cannot be solved by using
general linear mathematical formula due to the requirement for human expertise, the house selection problem can be
viewed as a semi-structured problem [1].
To reduce any complication, people usually seek advice from those who have experiences in buying a house or
consultant who has role as an expert. Such an advice is very useful, particularly related with the knowledge and
expertise to help solving the problem. However, according to [2], there are some potential limitations of working
with the expert for decision making such as the different level of expertise or the absence of experts, the lack of
capacity due to physical or mental workload, the missing of important details of a problem, inconsistent decision
making in the same situation, the limitation in storing and memorizing everything about work or incapacity in
recalling or understanding large data in very short time, the possible bias in decision making due to personal factors,
the evasion in responsibilities after decision making, fraudulence, age factor, and other factors that may decrease
the expertise level.
In order to overcome the drawbacks of working with human expert, a computer program of decision support
systems (DSS) which able to process data using expert’s knowledge can be used as a tool to help people with no
experience to make decision like an expert [1]. Researchers have made several attempts to develop DSS for solving
multi criteria decision making for broad area, among them for selecting scholarship recipient [3], selecting phones
for people who would like to buy new ones [4], providing design and selecting mechanical equipment of a craft [5],
advising high school students to choose major study at Higher Education [6], selecting major field of study for high
school students [7], selecting appropriate information system development methodology [8], choosing subjects for
university students [9], determining location of seminar [10], selecting suitable water pumps for agricultural

Human-Dedicated Sustainable Product and Process Design: Materials, Resources, and Energy
AIP Conf. Proc. 1977, 020049-1–020049-7; https://doi.org/10.1063/1.5042905
Published by AIP Publishing. 978-0-7354-1687-1/$30.00

020049-1
irrigation [11], recommending fashion style [12], analyzing industrial location [13], and selecting methods and tools
of design and environment [14]. These publications show that the performance of DSS in the decision making is
equal or almost equal to the expert and can help or advise non-experienced people to make decision like an expert so
that can increase effectiveness and efficiency.
There are some popular methods developed by researchers for solving multi criteria decision making problem,
one of them is the weighted product method (WPM) introduced by Bridgeman in 1922 [15] [16]. WPM is proven as
very reliable method for solving multi criteria decision making and has been tested for more than three criteria up to
one hundred criteria [15]. A number of researchers reported the successful application of WPM to solve multicriteria
decision making such as for selecting boarding house [17], choosing appropriate food [18], selecting suitable
learning platform [19], diagnosing duck disease [20] and other similar works. Based on these successful works, this
study attempted to implement WPM for solving house selection for people who face difficulties to make decision.
The method was computed and implemented in a web-based system. The main objectives of this study were:
x To develop house selection decision making model using WPM,
x To calculate and sort the recommendation values,
x To implement the decision support system into web-based environment.

RESEARCH METHOD

The Criteria Extraction

In this empirical work, the first step of the research was to find the exact picture of the society in terms of the
factors in house selection from the available alternatives. The data was gathered by using interview from population
in Surakarta area including property marketing or salesperson or agents, property consultant or expert, respondents
who have experiences in choosing or buying house, and also respondents who have no experience in house
selection. Interview was conducted and questionnaire was distributed to the respondents. They had to list the factors
in house selection and the degree of importance (weight) in scale 5 (1 is the least important while 5 is the most
important factor). Respondents were required to list as many as possible criteria. Besides the weight, respondents
were also required to include each criterion into either benefit or cost category. It was a heuristic process, because
respondents used their knowledge and experience to mention the factors, weight and category. The top eleven
criteria mentioned by respondents including their weight and category are presented in Table 1.
TABLE 1. The Top Eleven Criteria
No. Criteria Weight Category
C1 Location: the distance of the house to the public facilities (in meter (m)) 5 cost
C2 Price (in Rp.) 5 cost
C3 Size of land (in m2) 2 benefit
C4 Size of house building in total (in m2) 2 benefit
C5 Number of bed room 1 benefit
C6 Number of bath room 1 benefit
C7 Security services: the length time of security guard availability in a day (in hour) 3 benefit
C8 Number of floor 1 benefit
C9 Environment (in scale of 5) 4 benefit
C10 Facility: type of available public facility such as mosque, open space, play 3 benefit
ground, and so on (in scale of 5)
C11 School: distance to school for children (in meter (m)) 3 benefit

These eleven criteria were then computed using WPM to obtain the recommendation value.

Computation Process

The computation steps of WPM method are as follows:


1. List the criteria used in the computation based on the weight and category.
2. Calculate the relative weight of j -th criteria to the total weight of all criteria using the following formula 1.
wj
wj N
(1)
¦w
j 1
j

020049-2
N
Where ¦w
j 1
j 1

3. Calculate the preference value of every i -th alternative using formula 2.


N

–x
wj
Si ij (2)
j 1

Where the value of w j is positive if the criteria is a benefit otherwise its value is negative.
4. Calculate relative preference value of each alternative to all alternatives using formula 3.
Si
Vi M
(3)
¦S
i 1
i

5. The higher V value means the better the alternative.

The relative weight of every criteria ( w j ) thus could be calculated as 0.17, 0.17, 0.07, 0.07, 0.03, 0.03, 0.1, 0.03,
0.13, 0.1, and 0.1 for C1 to C11 respectively. In this empirical study, the real data of available houses around
Surakarta City were used. The example of manual calculation for the seven available alternatives can be seen in
Table 2.
TABLE 2. Sample of manual calculation using empirical data
Altern C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11
atives -0.17 -0.17 0.07 0.07 0.03 0.03 0.1 0.03 0.13 0.1 0.1
Si Vi Rank
A1 300 575 127 90 4 2 24 1 4 3 3 0.540 0.147 2
A2 150 800 120 185 6 3 24 2 4 3 3 0.629 0.171 1
A3 500 462 115 41 2 1 24 1 4 3 3 0.463 0.126 7
A4 400 401 112 60 2 1 24 1 4 3 3 0.505 0.137 5
A5 300 490 100 50 2 1 24 1 4 3 3 0.502 0.136 6
A6 300 1000 105 150 3 2 24 3 4 3 3 0.515 0.140 4
A7 300 300 60 45 2 1 24 1 4 3 3 0.523 0.142 3

The manual calculation results are presented in Table 2, which show that WPM is able to distinguish
and rank the alternatives. Based on the given criteria and their weight, alternative A2 was found as the best
alternative.
The decision support systems (DSS) was implemented in web-based environment. The potential usage
of the DSS is its capacity to be used by people who seek advice without any requirement to actually meet
an expert. The user interface in DSS facilitates the user to interact with DSS using mouse click or keyboard
s and they can get the result of DSS output from the monitor display and printable version file. The user
interface will accept request from the user and forwards it into WPM model that will process the data
stored in data management sub-system and then produce a result. The schematic view of DSS is illustrated
in Fig. 1.

FIGURE 1. Schematic view of the developed DSS.

020049-3
The web-based DSS has two types of actor, i.e. administrator and user, in which both actors have to login to the
system in order to use the systems. Administrator has the rights to create, read, update, and delete the data of the
systems, including the alternatives and the criteria with their values. The user will be able to select the available
alternatives in the systems to compare, view computation process and obtain the rank of alternatives. These actions
can be seen in the use case diagram in Fig. 2.

a. b.
FIGURE 2. Use case diagram: (a) for administrator, and (b) for user.

The detail of each activity in the use case diagram in Fig. 2 is elaborated in flowchart, for example flowchart for
administrator in inputting the available alternative as demonstrated in Fig. 3.

FIGURE 3. Flowchart of Alternative Input.

020049-4
RESULTS AND DISCUSSION

The WPM for house selection was successfully implemented in a web-based environment by using personal
home page (PHP) programming language, xampp web server and MySQL database management systems. In order
to be able to use this developed DSS, firstly, the user has to sign up to get an account and log in to the system. The
example of sign up page for new user is presented in Figure 4. After log in, user can start to consult by choosing the
available house alternatives and press the process button on the user interface as displayed in Fig. 5. The alternative
data is displayed in the interface and then processed by WPM from the database. All the user interface in the
program are displayed in Bahasa Indonesia.

FIGURE 4. User interface for sign up of new user

FIGURE 5. House Selection interface page.

The DSS output is a ranking of alternatives based on the value of S vector and V vector. The output can be seen
in the monitor display as well as printable file in the portable distribution format (pdf) as demonstrated in Figure 6.
The final test for the application was a validation test by user (user acceptance test) [21]. In the user acceptance
test, users were required to try the DSS and then to fill the questionnaire by choosing the point for every given

020049-5
statement. There were six statements in the questionnaire: (1) system is very easy to use, (2) the template design of
the system is very good, (3) assessment of the people needs is appropriate, (4) the systems is helpful for people, (5)
the menu of the system is complete, (6) the system as a whole is very good. The results of the user acceptance test
showed that around 60% of respondents stated the system is very helpful and has complete features.

FIGURE 6. The computation result in the printable file in pdf format.

CONCLUSION
Based on the results, it can be concluded that the WPM has been successfully computed and implemented in a
web-based DSS to help people in selecting a house. Users are able to select the alternatives and compare them as
well as to get the computation result in the form of display in monitor or printable version file. This developed
system can be enhanced by implementing the method for other fields of application with various supporting
technology, e.g., for mobile devices.

REFERENCES

1. E. Turban, J. E. Aronson, J-P Liang JP and R. V. McCarthy, Decision Support Systems And Intelligent Systems,
7th Ed. (Prentice Hall of India, New Delhi, 2007).
2. A. Badiru and J. Y. Cheung, Fuzzy Engineering Expert Systems with Neural Network (John Wiley and Sons,
New York, 2002).
3. S. ‘Uyun, and I. A. Riadi, Telkomnika 9(1), 37-46 (2011).
4. R. N. P. Atmojo, Anindito, B. Pardamean, B. S. Abbas, A. D. Cahyani, and I. D. Manulang, American Journal
of Applied Sciences 11(4), 666-680 (2014).
5. T. Gonciarz, The International Journal on Marine Navigation and Safety of Sea Transportation, 8(2), 275-280
(2014).
6. Y. Bouaiachi, M. Khaldi, and A. Azmani, International Journal of Scientific & Engineering Research 5(11),
25-28 (2014).
7. D. Pratiwi, J. P. Lestari, and D. Agushinta, International Journal of Computer Trends and Technology (IJCTT)
10(3), 153-159 (2014).
8. A. N. H. Zaied, S. I. A. Aal, and M. M. Hassan, I.J. Intelligent Systems and Applications 2013(09), 19-26
(2013).

020049-6
9. S. Saraswathi, M. H. K. Reddy, S. U. Kumar, M. Suraj, and S. K. Shafi, IJRET: International Journal of
Research in Engineering and Technology 3(7), 314-319 (2014).
10. Muttaqin, and A. Harjoko, Journal of Asian Scientific Research 5(8), 431-438 (2015).
11. A. Y. Seflek, and K. Carman, Mathematical and Computational Applications 15(1), 108-116 (2010).
12. Y. A. Nada, and H. Meshref, International Journal of Computer Applications 105(4), 15-20 (2014).
13. A. Rikalovic and I. Cosic, Acta Polytechnica Hungarica 12(2), 33-51 (2015).
14. N. V. Hernandez, G. O. Kremer, L. C. Schmidt, and P. R. A. Herrera, Expert Systems with Applications
39(2012), 9543-9553 (2012).
15. E. Triantaphyllou, and S. H. Mann, Decision Support Systems 5(1989), 303-312 (1989).
16. G. O. Odu, and O. E. Charles-Owaba, IOSR Journal of Engineering (IOSRJEN) 3(10), 1-14 (2013) .
17. D. R. Anamisa, A. Rachmad, and R. Widiastutik, Journal of Theoretical and Applied Information Technology
92(1), 52-58 (2016).
18. Adriyendi, I.J. Information Engineering and Electronic Business 6(2015), 8-14 (2015).
19. I. M. Ardana, I. P. W. Ariawan, and D. G. H. Divayana, (IJARAI) International Journal of Advanced Research
in Artificial Intelligence 5(12), 15-18 (2016).
20. D. G. H. Divayana, (IJACSA) International Journal of Advanced Computer Science and Applications 5(8), 48-
54 (2014).
21. R. S. Pressman, Software Engineering A Practitioner’s Approach, 5 th ed. (McGraw-Hill, Boston, United States:
McGraw-Hill, 2001).

020049-7

Anda mungkin juga menyukai