itjemast-v04(2):: international transaction journal of engineering, management, & applied...

IN THIS ISSUE Using DEMATEL Method to Analyze the Causal Relations on Technological Innovation Capability Evaluation Factors in Thai Technology-Based Firms The Effect of the 2,4-Dichlorophenoxy Acetic Acid, Benzyl Adenine and Paclobutrazol, on Vegetative Tissue-Derived Somatic Embryogenesis in Turmeric (Curcuma var. Chattip) Application of Bender’s Decomposition Solving a Feed–mix Problem among Supply and Demand Uncertainties Securing Bank Loans and Mortgages Using Real Estate Information Aided by Geospatial Technologies Theoretical Investigation of Hetero–Diels–Alder Functionalizations on SWCNT and Their Reaction Properties The Phenomenology of Lamban Tuha: The Local Wisdom of South Sumatra Traditional Architecture

International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies

Volume 4 Issue 2 (April 2013)

ISSN 2228-9860 eISSN 1906-9642

http://TuEngr.com

Cover Photo: Marina Bay Sands Building, in Singapore. Photo is taken by Mr. Saranyu SAVASRAT. Photo is used with permission.


http://TuEngr.com

International Editorial Board Editor-in-Chief Ahmad Sanusi Hassan, PhD Associate Professor Universiti Sains Malaysia, MALAYSIA

Executive Editor Boonsap Witchayangkoon, PhD Associate Professor Thammasat University, THAILAND

Noble Editorial Board: Professor Dr.Mikio SATOMURA (Shizuoka University, JAPAN) Professor Dr.Chuen-Sheng Cheng (Yuan Ze University, TAIWAN) Professor Dr.I Nyoman Pujawan (Sepuluh Nopember Institute of Technology, INDONESIA) Professor Dr.Neven Duić (University of Zagreb, CROATIA) Professor Dr.Lee, Yong-Chang (Incheon City College SOUTH KOREA) Professor Dr.Phadungsak Ratanadecho (Thammasat University, THAILAND) Professor Dr.Dewan M. Nuruzzaman (Dhaka University of Engineering & Technology, BANGLADESH) Professor Dr. Lutero Carmo de Lima (State University of Ceará, BRAZIL ) Associate Prof.Dr. Prapat Wangskarn (Dean of Faculty of Engineering, Thammasat University, THAILAND) Associate Prof.Dr.Uruya Weesakul (Past-Dean of Faculty of Engineering, Thammasat University, THAILAND )

Scientific and Technical Committee & Editorial Review Board on Engineering, Technologies and Applied Sciences: Associate Prof. Dr. Paulo Cesar Lima Segantine (University of São Paulo, BRASIL) Associate Prof. Dr. Kurt B. Wurm (New Mexico State University, USA ) Associate Prof. Dr. Truong Vu Bang Giang (Vietnam National University, Hanoi, VIETNAM ) Dr.H. Mustafa Palancıoğlu (Erciyes University, TURKEY) Associate Prof. Dr.Narin Watanakul (Thammasat University, THAILAND) Associate Prof. Commander Dr.Komsun Suwannarurk (Thammasat University, THAILAND ) Associate Prof.Dr.Peter Kuntu-Mensah (Texas A&M University-Corpus Christi, USA) Associate Prof.Dr. Anchalee Jala (Thammasat University, THAILAND ) Associate Prof. Dr. Masato SAITOH (Saitama University, JAPAN ) Assistant Prof.Dr. Zoe D. Ziaka (International Hellenic University, GREECE ) Associate Prof.Dr. Supornchai Utainarumol (King Mongkut's University of Technology North-Bangkok, THAILAND) Associate Prof.Dr.Chavalit Chaleeraktrakul (Thammasat University, THAILAND ) Associate Prof.Dr.Krittiya Lertpocasombut (Thammasat University, THAILAND ) Associate Prof.Dr. Bovornchok Poopat (King Mongkut's University of Technology Thonburi, THAILAND ) Associate Prof.Dr.Pudit Laksanacharoen (King Monkut's University of Technology North Bangkok, THAILAND) Associate Prof.Dr.K Pianthong (Ubon Ratchathani University, THAILAND) Associate Prof.Dr.Thanaporn Supriyasilp (Chiang Mai University, THAILAND) Associate Prof.Dr. Junji SHIKATA (Yokohama National University, JAPAN) Associate Prof. Dr.Aree Taylor (Thammasat University, THAILAND) Assistant Prof.Dr. Akeel Noori Abdul Hameed (University of Sharjah, UAE) Assistant Prof.Dr. Atch Sreshthaputra (Chulalongkorn University, THAILAND) Assistant Prof.Dr. Rohit Srivastava (Indian Institute of Technology Bombay, INDIA) Assistant Prof.Dr. Watanachai Smittakorn (Chulalongkorn University, THAILAND ) Assistant Prof.Dr. Kitjapat Phuvoravan (Kasetsart University, THAILAND) Assistant Prof.Dr. Khiensak Seangklieng (Thammasat University, THAILAND ) Assistant Prof.Dr. Chainarong Chaktranond (Thammasat University, THAILAND ) Assistant Prof.Dr.Kridayut Chompoming (Thammasat University, THAILAND ) Assistant Prof.Dr. Nopporn Leeprechanon (Thammasat University, THAILAND ) Assistant Prof.Dr. Suphattra KETSARAPONG (Sripatum University, THAILAND ) Assistant Prof.Dr. Sawat Pararach (Thammasat University, THAILAND ) Assistant Prof.Dr. Winai Raksuntorn (Thammasat University, THAILAND ) Assistant Prof.Dr. Watit Pakdee (Thammasat University, THAILAND ) Assistant Prof.Dr. Cattaleeya Pattamaprom (Thammasat University, THAILAND ) Assistant Prof.Dr.Puttipol Dumrongchai (Chiangmai University, THAILAND ) Madam Wan Mariah Wan Harun (Universiti Sains Malaysia, MALAYSIA ) Dr. David Kuria (Kimathi University College of Technology, KENYA ) Dr. Mazran bin Ismail (Universiti Sains Malaysia, MALAYSIA ) Dr.Isares Duchallaya (Thammasat University, THAILAND ) Dr.Bandit Suksawat (King Mongkut's University of Technology North-Bangkok, THAILAND ) Dr. Salahaddin Yasin Baper (Salahaddin University - Hawler, IRAQ ) Dr. Foong Swee Yeok (Universiti Sains Malaysia, MALAYSIA) Dr.Orawan Chunhachart (Kasetsart University Kamphaengsaen Campus, THAILAND ) Dr. Manop Kaewmoracharoen (Chiang Mai University, THAILAND)

2013 International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies.

http://tuengr.com/

i

:: International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies

Volume 4 Issue 2 (April, 2013) ISSN 2228-9860 http://TuEngr.com eISSN 1906-9642

FEATURE PEER-REVIEWED ARTICLES

Using DEMATEL Method to Analyze the Causal Relations on Technological Innovation Capability Evaluation Factors in Thai Technology-Based Firms 81

The Effect of the 2,4-Dichlorophenoxy Acetic Acid, Benzyl Adenine and Paclobutrazol, on Vegetative Tissue-Derived Somatic Embryogenesis in Turmeric (Curcuma var. Chattip) 105

Application of Bender’s Decomposition Solving a Feed–mix Problem among Supply and Demand Uncertainties 111

Securing Bank Loans and Mortgages Using Real Estate Information Aided by Geospatial Technologies 129

Theoretical Investigation of Hetero–Diels–Alder Functionalizations on SWCNT and Their Reaction Properties 145

The Phenomenology of Lamban Tuha: The Local Wisdom of South Sumatra Traditional Architecture 157

Contact & Office: Associate Professor Dr. Ahmad Sanusi Hassan (Editor-in-Chief), School of Housing, Building and Planning, UNIVERSITI SAINS MALAYSIA, 11800 Minden, Penang, MALAYSIA. Tel: +60-4-653-2835 Fax: +60-4-657 6523, [email protected] Associate Professor Dr. Boonsap Witchayangkoon (Executive Editor), Faculty of Engineering, THAMMASAT UNIVERSITY, Klong-Luang, Pathumtani, 12120, THAILAND. Tel: +66-2-5643005 Ext 3101. Fax: +66-2-5643022 [email protected] Postal Paid in THAILAND.


*Corresponding author (P.Anuntavoranich). Tel: +66-2-657-6334. E-mail: [email protected], [email protected]. 2013. International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/081-103.pdf

81


http://TuEngr.com, http://Go.to/Research

Using DEMATEL Method to Analyze the Causal Relations on Technological Innovation Capability Evaluation Factors in Thai Technology-Based Firms Detcharat Sumrit a, and Pongpun Anuntavoranich a*

a Technopreneurship and Innovation Management Program, Graduate School, Chulalongkorn University, Bangkok, THAILAND. A R T I C L E I N F O

A B S T RA C T

Article history: Received 25 September 2012 Received in revised form 28 December 2012 Accepted 14 January 2013 Available online 18 January 2013 Keywords: Technology Innovation Capability; TIC evaluation factors; DEMATEL method; Cause and effect relationship.

This study analyzes the technology innovation capabilities (TICs) evaluation factors of enterprises by applying the Decision Making Trial and Evaluation Laboratory (DEMATEL) method. Based on the literature reviews, six main perspectives and sixteen criteria were extracted and then validated by six experts. A questionnaire was constructed and answered by eleven experts. Then the DEMATEL method was applied to analyze the importance of criteria and the casual relations among the criteria were constructed. The result showed that the innovation management capability perspective was the most important perspective and influenced the remaining perspectives. This work also presents the significant criteria for each perspective.

2013 INT TRANS J ENG MANAG SCI TECH.

1. Introduction Innovation’s importance has continuously increased and aligns with global business

growth. Bessant et al., (2005), and Huang (2011) clearly stated that Technological Innovation

Capabilities (TICs) play a crucial part in the initiation of firms’ competency and as the source of

sustainable competitive advantage. The enterprises, thus, are strongly required the periodical

monitoring their TICs and have to continuously strengthen their weak capabilities in order to


82 Detcharat Sumrit and Pongpun Anuntavoranich

facilitate the competitive advantage. This study mainly focuses on the technological-based

firms since they rely significantly on innovation development to pursue their business growth.

Although TICs were accepted as a main part of enhancing competitive advantage, TICs

assessment is rather complicated due to multi-dimensionality. The measuring indicators of

TICs are also diverse and difficult to assess by any single-dimension scale as they involve the

interaction among various resources (Chiesa et al., 1998, Hansen, 2001, Guan and Ma, 2003,

Burgelman et al., 2004). Guan et al., (2006) defined TICs measurement framework as

benchmark audition on the quantitative evaluation based on traditional DEA approach, which

relies on both technological capability and critical capabilities in the area of manufacturing,

marketing, organization, strategy planning, learning and resources allocation.

However, Wang et al., (2008) proposed the evaluation of high-tech firms’ TICs under both

quantitative assessment (by applying new fuzzy multi-criteria analytical approach) and

qualitative assessment (using five main aspects of capabilities i.e. R&D, innovation decision,

marketing, manufacturing and capital). Wang et al., (2008) viewed that the traditional

multi-criteria were not wholly suitable for TICs assessment. They also stated that the TICs

assessment was considered as subjective and ambiguous.

To clarify and reduce the subjective and ambiguous information, this study uses both

qualitative and quantitative methods. In this study, TICs’ critical evaluation perspectives and

criteria as well as the causal relations among them are presented. The result will aid the

managements in the determination of the degree of importance of critical factors/ criteria and

their influences on these factors.

Following this introduction, literature reviews of TICs and DEMATEL method were

illustrated in Section 2. Research methodology (including research framework, and the

procedure and results) was proposed in Section 3. Discussion and results were conducted in

Section 4. Finally, Section 5 drew the conclusion.


83

2. Literature Review

2.1 Technological innovation capability (TICs) TICs was defined as an enterprises’ ability to improve their technological innovativeness

in order to create new customer value through the introduction of new products and services,

the exploitation of new technologies and the exploration of new skill and competencies

(Perdomo-Ortiz et al., 2009, Wang et al., 2008, Huang, 2011). TICs assessments were also

included the aspects of multi-dimensionality, complexity, interactive innovation activities with

resource allocation to enhance competitive advantage (Wang et al., 2008, Chiesa et al., 1996).

Various researchers have developed the technological innovation framework, approaches

and components to evaluate a firm’s technological or innovation capabilities. For instance,

Baark et al., (2011) classified the assessment of a firm’s TICs into four approaches: (i) the asset

approach (Christensen, 1995), (ii) process approach (Chiesa et al., 1996; Burgelman et al.,

2004), (iii) output-based (Romijin and Albaladejo, 2002), and (iv) functional approach (Guan

and Ma, 2003; Yam. et al., 2004). Yam et al., 2004 developed an audit innovation capability

model by using functional approach, which consisted of seven components: learning capability,

R&D capability, resource allocation capability, manufacturing capability, marketing capability,

organizing capability and strategic planning capability. These studies of technological

innovation capability development are basically related to our research in view of providing an

overall framework for understanding the importance of such capability.

Based on the extensive literature review, overall TICs evaluation factors were concluded in

Table 1.


Table 1: Summary of the perspectives and criteria of TICs’ evaluations Perspectives/ Criteria Author

Innovation Management Capability (P1)

Strategic Management Capability (C1) Burgelman et al., (2004), O’Regan et al., (2006), Ceylan and Koc (2007), Dobni (2008), Yam et

al., (2004), Yam et al., (2011), Türker (2012).

Organization Capability (C2) Guan et al., (2006), O’Regan et al., (2006), Burgelman et al., (2004), Yam et al., (2004), Yam et

al., (2011), Ceylan and Koc (2007), Dobni (2008), Spyropoulou and Kyrgidou (2012), Türker

(2012).

Resource Allocation Capability (C3) Chiesa et al., (1996), Barney and Clark (2007), Burgelman et al., (2004), Guan et al., (2006),

Dobni (2008), Wang et al., (2008), Ceylan and Koc (2007), Yam et al., (2011), Spyropoulou

and Kyrgidou (2012), Voudouris et al., (2012).

Risk Management Capability (C4) Amabile et al., (1996), Isaksen et al., (1999), Forsman (2011), Yang (2012).

Collective Learning Capability (P2)

Learning Capability (C5) Guan et al., (2006), Chiva and Alegre (2007), Teece (2007), Alegre and Chiva (2008), Yam et

al., (2004), Yam et al., (2011), Camisón and Villar-López (2012).

Absorptive Capacity (C6) Ceylan and Koc (2007), Zahra and George (2002), Lane and Koka (2006), Camisón and Forés

(2010), Forsman (2011), Wonglimpiyarat (2010), Kim et al., (2011), Gebauer et al., (2012),

Lin et al., (2012).

Knowledge Management Capability (C7) Forsman (2011), Yang (2012).

Innovation Sourcing Capability(P3)

Network Linkage Capability (C8) Lin (2004), Chesbrough (2004), Tidd (2006), Kim and Song (2007), Spithoven et al., (2010),

Shan and Jolly (2010), Zeng et al., (2010), Huang (2011), Forsman (2011), Mu and Benedetto

(2011), Kim et al., (2011), Voudouris et al., (2012).

Technology Acquisition Capability (C9) Chiesa et al., (1996), Ceylan and Koc (2007), Lee et al., (2009).

Technology Development Capability(P4)

R&D Capability (C10) Guan et al., (2006), Wang et al., (2008), Yam et al., (2004), Yam et al., (2011), Zahra and

George (2002), Levitas and Mc Fadyen (2009), Kim et al., (2011), Forsman (2011), Lin et al.,

(2012).

Project Cross Functional Team Integration

Capability (C11)

Martins and Terblanche (2003), Lin (2004), Camisón and Forés (2010), Kim et al., (2011), Yam

et al., (2011).

Technology Change Management Capability

(C12)

Jansen et al., (2005), Garrison (2009), Forsman (2011).

Robustness Product & Process Design Capability (P5)

Product Structural Design and Engineering

Capability (C13)

Chiesa et al., (1996), Christensen (1995), Zhang et al., (2000), De Toni and Nassimbeni (2001),

Antony et al., (2002), Nassimbeni and Battain (2003), Lin (2004), Ho et al., (2011).

Process Design and Engineering Capability

(C14)

Chiesa et al., (1996), Zhang et al., (2000), De Toni and Nassimbeni (2001), Antony et al.,

(2002), Nassimbeni and Battain, (2003).

Technology Commercialization Capability (P6)

Manufacturing Capability (C15) Lin (2004), Yam et al.,(2004), Guan et al. (2006), Wang et al.,(2008), Yam et al., (2011), Kim

et al., (2011), Yang (2013).

Market Capability (C16) Lin (2004), Yam et al., (2004), Guan et al., (2006), Dobni (2008), Wang et al., (2008), Yam et

al., (2011), Forsman (2011), Mu and Benedetto (2011), Kim et al., (2011).


85

2.2 DEMATEL Method DEMATEL method was originally developed between 1972 to 1979 by the Science and

Human Affairs Program of the Battelle Memorial Institute of Geneva, with the purpose of

studying the complex and intertwined problematic group. It has been widely accepted as one

of the best tools to solve the cause and effect relationship among the evaluation criteria (Chiu et

al., 2006, Liou et al., 2007, Tzeng et al., 2007, Wu and Lee, 2007, Lin and Tzeng, 2009). This

method is applied to analyze and form the relationship of cause and effect among evaluation

criteria (Yang et al., 2008) or to derive interrelationship among factors (Lin and Tzeng, 2009).

Based on Yu and Tseng (2006), Liou, et al., (2007), Tzeng, et al., (2007), Yang, et al., (2008),

Wu and Lee (2007), Shieh et al., (2010), the procedure of DEMATEL method is presented

below:

Figure 1: The process of the DEMATEL method.

Step 1: Gather experts’ opinion and calculate the average matrix Z

A group of m experts and n factors are used in this step. Each expert is asked to view the

degree of direct influence between two factors based on pair-wise comparison. The degree to

which the expert perceived factor i affects on factor j is denoted as xij. The integer score is

ranged from 0 (no influence), 1 (low influence), 2 (medium influence), 3 (high influence), and 4

(very high influence), respectively. For each expert, an n x n non-negative matrix is constructed

as Xk = , where k is the expert number of participating in evaluation process with 1≤ k ≤ m.

Thus, X1, X2, X3,.., Xm are the matrices from m experts.

Step 5:

Set the

threshold value

(α)

Step 6:

Build a cause and

effect relationship

diagram

Yes

No

Step 4:

Calculate the sums

of rows and

columns of matrix T

Step 2:

Calculate the normalized

initial direct-relation

matrix D

Step 3:

Derive the total

relation matrix T

The final cause

and effect

relationship

Step 1:

Gather experts’ opinion and

calculate the average matrix Z

Is a cause and effect

relationship diagram

acceptable?


To aggregate all judgments from m experts, the average matrix Z= [zij] is shown below.

zij = ∑ (1)

Step 2: Calculate the normalized initial direct- relation matrix D

The normalized initial direct-relation matrix D = [dij], where value of each element in

matrix D is ranged between [0, 1]. The calculation is shown below.

D = λ * Z, (2)

or

[dij]nxn = λ [zij]nxn (3)

where

λ = Min ∑ , ∑ (4)

Based on Markov chain theory, is the powers of matrix D, e.g. D2, D3,…, D∞

guarantees the convergent solutions to the matrix inversion as shown below.

lim = [0]nxn, (5)

Step 3: Derive the total relation matrix T

The total-influence matrix T is obtained by utilizing Eq. (7), in which, I is an n x n identity

matrix. The element of tij represents the indirect effects that factor i had on factor j, then the

matrix T reflects the total relationship between each pair of system factors.

T = lim (D + + …+ (6)

= ∑

where

∑ = D1 + + …+


87

= D (I + D1 + + …+ )

= D (I - D)-1(I - D)(I + D1 + + …+ )

= D (I - D)-1(I - Dm)

T = D (I - D)-1 (7)

Step 4: Calculate the sums of rows and columns of matrix T

In the total-influence matrix T, the sum of rows and the sum of columns are represented by

vectors r and c, respectively.

r = [ri]nx1 = ∑ nx1, (8)

c = ́1xn = ∑ ́1xn , (9)

where ́ is denoted as transposition matrix.

Let ri be the sum of ith row in matrix T. The value of ri indicates the total given both directly

and indirectly effects, that factor i has on the other factors.

Let cj be the sum of the jth column in matrix T. The value of cj shows the total received both

directly and indirectly effects, that all other factors have on factor j. If j = i, the value of (ri + ci)

represents the total effects both given and received by factor i. In contrast, the value of (ri-ci)

shows the net contribution by factor i on the system. Moreover, when (ri -ci) was positive, factor

i was a net cause. When (ri -ci) was negative, factor i was a net receiver (Tzeng et al., 2007; Liou

et al., 2007; Yang et al., 2008; Lee et al., 2009).

Step 5: Set a threshold value (α)

The threshold value ), was computed by the average of the elements in matrix T, as

computed by Eq. (11). This calculation aimed to eliminate some minor effects elements in

matrix T. (Yang et al., 2008).

= ∑ ∑

N (10)

where N is the total number of elements in the matrix T.


Step 6: Build a cause and effect relationship diagram

The cause and effect diagram is constructed by mapping all coordinate sets of (ri +ci, ri -ci)

to visualize the complex interrelationship and provide information to judge which are the most

important factors and how influence affected factors (Shieh et al., 2010). The factors that tij is

greater than , are selected shown in cause and effect diagram (Yang et al., 2008).

3. Research Methodology

3.1 Research Framework of TICs This section established a structure for identifying the evaluation perspective and criteria

as well as their relationships of TICs factors. An overview of the proposed TICs evaluation

framework was illustrated in Figure 2. The details of each procedure and the results were

explained in next section.

Figure 2: The proposed procedure of TICs criteria assessment.

3.2 Procedure and the result This section is to describe the process of TICs evaluation perspectives and criteria,

according to Figure 2. Not only the determination of TICs evaluation perspectives and criteria

but also the measurement of the relationship among them was also performed. The process and

the result of each stage were presented in the following stages:

Content validation by experts

Extensive literature reviews

Applying DEMATEL method

Stage 1: Define the problem statement

Stage 2: Explore TICs measurement perspectives and

criteria

Stage 3: Develop interview questionnaire and validity

testing

Stage 4: Interview session with 11 industrial experts

Stage 5: Analyze cause and effect relationship among

and identify evaluation factors


89

3.2.1 Stage of defining the problem statements To encounter the fierce competition of the dynamic global environment and the upcoming,

TICs are considered as one of the significant factors of Thai technology-based firms to sustain

competitiveness. Hence, an evaluation of TICs turns to be a tool to aid managements to define

strengths and weaknesses in term of TICs. Appropriate factors of TICs then should be

identified. This study presents not only the appropriate factors but also the cause and effect

relationship among the perspectives and criteria.

3.2.2 Stage of exploring the TICs measurement perspectives and criteria from

literature reviews

The extensive literature review was conducted to identify multi-attributions and

multi-dimensionalities of the TICs evaluation factors. Based on the reviews, six perspectives

and sixteen evaluation criteria were derived as shown in Table 1.

3.2.3 Stage of developing a questionnaire After obtaining the sixteen criteria and six perspectives of TICs evaluation factors from

literatures, a questionnaire was designed. A group of qualified experts reviewed and tested the

designed questionnaire to assure the content validity of questionnaire. The group of qualified

experts was consisted of three professionals from academic institutions, two from industrial

sector and one from Thai Automotive Institution. After interviewing, the questionnaire was

revised based on the experts’ aspects.

3.2.4 Stage of interviewing session Eleven experts were asked to complete the questionnaire. The experts have at least 5 years

experiences and worked in management positions in well-known Thai technology-based firms

and some of the firms were awarded as Thailand’s Most Innovative Company in 2010. After

obtaining the completed questionnaires from the experts, DEMATEL analytical technique was

used to determine the causal relations and to identify the significant perspectives and criteria.

The results of analyses were shown in the next section.


3.2.5 Stage of analyzing the causal relation and identifying the evaluation

perspectives and criteria

Based on the six perspectives and sixteen criteria of TICs evaluation as stated above, this

study further employed the DEMATEL method to indicate the complex relationship and

identify the significant TICs evaluation perspectives and criteria. In this section, the

computation was divided into two parts for calculating on perspectives and criteria,

respectively. The procedure of the DEMATEL method and the results of each stage were also

presented as follows.

3.2.5.1 Applying DEMATEL method on the six perspectives

Xk showed the data gathered in terms of the six perspectives of expert k, where Xk = .

Step procedures of applying DEMATEL method were shown next.

0 3 3 3 3

4

0 2 3 3 2

4

0 3 2 2 3

4

0 2 2 3 3 3

2 0 3 3 2 2 2 0 3 3 3 3 2 0 2 3 3 0 1 0 3 3 3 3

X1= 1 1 0 2 1 1 X2= 1 2 0 2 2 0 X3= 1 3 0 1 2 1 X4= 1 2 0 0 2 2

1 3 3 0 1 3 2 4 1 0 2 2 2 3 2 0 2 1 1 3 2 0 2 2

2 1 2 3 0 3 1 2 2 1 0 3 1 2 2 2 0 4 2 2 1 3 0 3

1 2 2 1 2 0 2 1 2 3 2 0 1 2 1 1 2 0 2 1 2 2 2 0

0 2 2 0 2

3

0 2 2 3 2

4

0 2 3 2 2

3

0 1 2 3 2 3

2 0 2 3 2 2 2 0 2 3 3 3

2 0 3 3 2 0 2 0 3 3 2 2

X5= 1 2 0 2 2 1 X6= 1 2 0 2 2 1 X7= 2 3 0 2 3 2 X8= 1 2 0 2 2 2

2 3 1 0 3 2 2 3 2 0 0 1

1 4 2 0 2 1 2 4 1 0 1 2

2 1 2 2 0 3 2 2 2 3 0 2

2 1 2 2 0 3 1 2 2 1 0 3

2 1 2 2 2 0 1 2 2 2 1 0

2 1 1 1 2 0 3 1 2 1 1 0

0 1 2 3 2

4

0 2 3 2 2

3

0 2 1 1 2

3

3 0 3 3 3 2 2 0 2 3 3 0

1 0 2 3 2 2 X9=

2 3 0 2 2 2 X10=

1 2 0 2 2 2 X11=

1 2 0 2 2 1

2 4 2 0 3 2 2 3 2 0 2 3

2 3 1 0 2 2

2 2 2 0 0 3 2 2 1 1 0 3

2 1 2 2 0 2

2 1 1 1 2 0 2 1 1 2 2 0

1 1 2 2 3 0


91

(1) The computation of the average matrix Z was constructed by using Eq. (1).

0 2 2.2727 2.2727 2.2727 3.4545

1.90909 0 2.54545 3 2.5454 1.7272

Z = 1.18181 2.1818 0 1.7272 2 1.3636

1.72727 3.3636 1.72727 0 1.8181 1.9090

1.72727 1.6363 1.81818 1.8181 0 2.9090

1.72727 1.2727 1.63636 1.6363 1.9090 0

(2) The normalized initial direct-relation matrix D was calculated by using Eq. (2) to

Eq.(5).

0.0000 0.1760 0.2000 0.2000 0.2000 0.3040

0.1680 0.0000 0.2240 0.2640 0.2240 0.1520

D = 0.1040 0.1920 0.0000 0.1520 0.1760 0.1200

0.1520 0.2960 0.1520 0.0000 0.1600 0.1680

0.1520 0.1440 0.1600 0.1600 0.0000 0.2560

0.1520 0.1120 0.1440 0.1440 0.1680 0.0000

(3) The total relation matrix T was calculated by using Eq. (6) to Eq. (7) as shown below.

1.1983 1.6022 1.5638 1.6194 1.6335 1.7897

1.3147 1.4312 1.5522 1.6366 1.6189 1.6449

T = 0.9837 1.2421 1.0379 1.2130 1.2383 1.2532

1.2195 1.5575 1.4040 1.3262 1.4714 1.5434

1.1290 1.3342 1.3001 1.3464 1.2203 1.4966

0.9883 1.1431 1.1256 1.1653 1.1928 1.1104

(4) The sums of rows and columns of matrix T were calculated by using Eq. (8) to Eq. (9)

as shown in Table 2.

Table 2: The sums of given and received among six perspectives. P1 P2 P3 P4 P5 P6 ri cj (ri+ cj) (ri- cj)

P1 1.1983 1.6022* 1.5638* 1.6194* 1.6335* 1.7897* 9.4068 6.8335 16.2403 2.5733

P2 1.3147 1.4312* 1.5522* 1.6366* 1.6189* 1.6449* 9.1985 8.3103 17.5087 0.8882

P3 0.9837 1.2421 1.0379 1.2130 1.2383 1.2532 6.9683 7.9836 14.9519 -1.0153

P4 1.2195 1.5575* 1.4040* 1.3262 1.4714* 1.5434* 8.5219 8.3069 16.8288 0.2151

P5 1.1290 1.3342 1.3001 1.3464 1.2203 1.4966* 7.8266 8.3752 16.2018 -0.5486

P6 0.9883 1.1431 1.1256 1.1653 1.1928 1.1104 6.7255 8.8382 15.5637 -2.1127


(5) The set up of the threshold value (α)

The threshold value was derived from the average of elements in matrix T, which was

calculated by using Eq. (10).

= .

= 1.351

(6) The construction of the cause and effect relationship diagram

The values of tij in Table 2, which were greater than α (1.351), were shown as tij*, which

presented the interaction between perspectives, e.g. the value of t12 (1.6022) > α (1.351), the

arrow in the cause and effect diagram was drawn from P1 to P2. The cause and effect diagram of

six perspectives was constructed as Figure 3.

Figure 3: The visualization of the causal relationship among perspectives of TICs.

3.2.5.2 Applying DEMATEL method on the sixteen criteria

Under each perspective, the significant criteria were determined by using the same

procedures as described in (1) to (6) above. Both direct and indirect effects of the criteria under

six perspectives were summarized in Table 3 and the cause and effect diagrams among criteria

under each perspective were shown in Figure 4 to Figure 9.

P3

P1

P2 P4

P5

P6

15.0 15.5 16.0 16.5 17.0 17.5

3

-1

-2

-3

2

1

0

r-c

r+c

Threshold Value (α) = 1.351


93

4. Discussion and Results

4.1 Results on the Perspectives The important of evaluation perspectives was determined by (r+c) values. Based on Table

3, Collective Learning Capability (P2) was the most important evaluation perspective with the

largest (r+c) value = 17.5087, whereas Innovation Sourcing Capability (P3) was the least

important perspective with the smallest (r+c) value = 14.9519. Regarding to (r+c) values, the

prioritization of the importance of six evaluation perspective was P2 > P4> P1> P5> P6> P3.

Table 3: The direct and indirect effects of the criteria under each perspective. Criteria (r+c) (r-c)

The overall effects of the four criteria of Innovation Management Capability perspective

Strategic Management Capability (C1) 17.8235 2.1543

Organization Capability (C2) 19.7564 0.1836

Resource Allocation Capability (C3) 17.0986 -1.3492

Risk Management Capability (C4) 17.7238 -0.9887

The overall effects of the three criteria of Collective Learning Capability perspective

Learning Capability (C5) 9.3937 0.3263

Absorptive Capability (C6) 9.3174 1.3097

Knowledge Management Capability(C7) 9.0972 -1.6360

The overall effects of the two criteria of Innovation Sourcing capability perspective

Network Linkage Capability (C8) 6.3333 1.0000

Technology Acquisition Capability (C9) 6.3333 -1.0000

The overall effects of the three criteria of Technology Development Capability perspective

R&D Capability (C10) 71.7604 2.8446

Project Cross Functional Team Integration Capability (C11) 68.7422 1.7228

Technology Change Management Capability(C12) 65.4969 -4.5675

The overall effects of the two criteria of Robustness Product & Process Design Capability perspective

Product Structural Design and Engineering Capability (C13) 7.8000 1.0000

Process Design and Engineering Capability (C14) 7.8000 -1.0000

The overall effects of the two criteria of Technology Commercialization Capability perspective

Manufacturing Capability (C15) 21.000 1.0000

Market Capability (C16) 21.000 -1.0000


Figure 5: The cause and effect diagram of the three

criteria of Collective Learning Capability

9.6

C5

C7

C6

9.0 9.2 9.4 -1

-2

2

1

0 r+c

r- c


Figure 4: The cause and effect diagram of the four

criteria of Innovation Management Capability

20 17 18 19

-1

-2

2

1

0

C3

C1

C2

C4

r+c

r- c


Figure 7: The cause and effect diagram of the three

criteria of Technology Development Capability

70 64 66 68 -2

-4

4

2

0

-6

6

C10 C11

C12

r+c

r- c


Figure 6: The cause and effect diagram of the two

criteria of Innovation Sourcing Capability.

8.0 2.0 4.0 6.0

r+c

r- c

C9

C8

-1

-2

2

1

0


10 4.0 6.0 8.0 r+c

r- c

-1

-2

2

1

0

C14

C13


Figure 8: The cause and effect diagram of the two criteria

of Robustness Product & Process Design Capability

Figure 9: The cause and effect diagram of the two

criteria of Technology Commercialization Capability

r+c

r- c

24 18 20 22 -1

-2

2

1

0

C16

C15



95

Based on (r-c) values, the six perspectives were divided into (i) cause group and (ii) effect

group.

(i) If the value of (r-c) was positive or net cause, such perspective was classified in the

cause group, and directly affected the others. The highest (r-c) factors also had the greatest

direct impact on the others. In this study, Innovation Management Capability (P1), Collective

Learning Capability (P2), and Technology Development Capability (P4) were classified in the

cause group, having the (r-c) values of 2.5733, 0.8882, and 0.2151, respectively. It also

indicated that P1 (Innovation Management Capability) was the most critical impact factor on

the others. Based on the matrix T in Table 2, it was found that P2 (Collective Learning

Capability) and P4 (Technology Development Capability) had a mutual interaction as both the

value of t24 (1.6366) and t42 (1.5575) were greater than α (1.351).

(ii) If the value of (r-c) was negative or net receive, such perspective was classified in the

effect group, and largely influenced by the others. For this study, Technology

Commercialization Capability (P6), Innovation Sourcing Capability (P3) and Robustness

Product and Process Design Capability (P5) were categorized in the effect group, with the (r-c)

values of -2.1127, -1.0153 and -0.5484, respectively. And P6 (Technology Commercialization

Capability) was the most affected by the other factors (P1), (P2), (P4), and (P5).

4.2 Results on the Criteria According to Table 3, under Innovation Management Capability perspective (P1), this

study found that Organization Capability (C2) and Strategic Management Capability (C1) were

the two most important criteria based on first and second highest (r+c) values of 19.7564 and

17.8235, respectively. Whereas both Strategic Management Capability (C1) and Organization

Capability (C2) were in the cause group based on their positive (r-c) values of 2.1543 and

0.1836, respectively. For Resource Allocation Capability (C3) and Risk Management

Capability (C4) were in the effect group, given negative (r-c) values of -1.3492 and -0.9887,

respectively. From Figure 4, Strategic Management Capability (C1) was the most critical

criteria because it directly influenced on the other three criteria. Organization Capability (C2)

had a direct impact on Resource Allocation Capability (C3) and a mutual interaction on Risk

Management Capability (C4).


For the perspective of Collective Learning Capability (P2), Learning Capability (C5) and

Absorptive Capability (C6) were the two most important criteria based on higher (r+c) values

of 9.3937 and 9.3174, respectively. They were also the net cause group with higher positive

(r-c) values of 0.3263 and 1.3097, respectively. For Knowledge Management Capability (C7)

was net receive with the (r-c) value of -1.6360. From Figure 5, Absorptive Capability (C6)

presented as the most significant criteria given impact to the other two criteria.

For the perspective of Innovation Sourcing capability (P3) in Table 3, Network Linkage

Capability (C8) and Technology Acquisition Capability (C9) showed the same importance level

of the (r+c) values 6.3333. However, based on the (r-c) value of 1.0 (Figure6), Network

Linkage Capability (C8), was a net cause and largely impacted Technology Acquisition

Capability (C9).

According to Technology Development Capability perspective (P4), R&D Capability (C10)

and Project Cross Functional Team Integration Capability (C11) were the two most important

criteria with highest (r+c) values of 71.7604, and 68.7422, respectively. Both of them were net

cause. As shown in Figure7, R&D Capability (C10) had the greatest (r-c) value of 2.8446, which

directly affected Technology Change Management Capability (C12) and had a mutual

interaction on Project Cross Functional Team Integration Capability (C11).

For the perspective of Robustness Product & Process Design capability (P5), both criteria

Product Structural Design and Engineering Capability (C13) and Process Design and

Engineering Capability (C14) had the same importance level of the (r+c) values equaling to

7.80. However, as Figure 8, Product Structural Design and Engineering Capability (C13) was

net cause with the (r-c) value of 1.0, and affected Process Design and Engineering Capability

(C14).

For the perspective of Technology Commercialization Capability (P6), there were the same

importance level of the two criteria i.e. Manufacturing Capability (C15) and Market Capability

(C16), based on their equal (r+c) values of 21.0. However, as Figure 9, Manufacturing

Capability (C15) was a net cause having the (r-c) value of 1.0 and affected Market Capability

(C16).


97

5. Conclusion This study applied DEMATEL method not only to analyze the TIC evaluation perspectives

and criteria, consisting of six perspectives and sixteen criteria for Thai technology-based firms’

but also to describe the cause and effect relationship among them. The result implied that the

management should concentrate on improving the three core perspectives in the cause group

i.e. Innovation Management Capability, Collective Learning Capability, and Technology

Development Capability. The three remaining perspectives were found in the effect group i.e.

Technology Commercialization Capability, Innovation Sourcing Capability and Robustness

Product and Process Design Capability, which they were also affected by the ones in the cause

group.

By the aspect of prioritizing the importance of criteria and the cause and effect relationship

among criteria under the three core perspectives, this study found that the Strategic

Management Capability, Absorptive Capability and R&D Capability were the most critical

criteria. Therefore, in order to enhance the overall competitive advantage in term of TICs, Thai

technology-based firms should allocate more resources in these core perspectives. In the case of

having limited resources, firms should emphasize on their Strategic Management Capability

since it is the main critical criteria in the adjustment of corporate planning and would yield

highest positive results on TICs.

6. Appendix The definitions of criteria are identified in Table A:

Table A: terms and definitions of criteria used in this study Terms Definitions

Strategic Management

Capability

The firm’s ability to identify internal strengths and weaknesses and external opportunities and

threats, to formulate plans in accordance with the corporate vision and missions, and to adjust the

plans for implementation (Yam et al., 2004).

Organization Capability The firm’s ability to secure the organizational mechanism and harmony, to cultivate the organization

culture, and to adopt the better management practices (Yam et al., 2004).

Resource Allocation

Capability

The firm’s ability to acquire and to allocate appropriately capital, exercise and technology in the

innovation process (Yam et al., 2004).

Risk Management Capability The firm’s ability to assess the risk of technological innovation and to take the risk of technological

innovation adoption (Forsman, 2011).


Table A: terms and definitions of criteria used in this study (continue) Terms Definitions

Learning Capability The firm’s ability to identify, to assimilate, and to exploit the knowledge from internal organization

(Yam et al., 2004).

Absorptive Capacity The firm’s ability to recognize, to assimilate, and to apply the value of new external information to

commercial ends (Cohen and Lavinthal, 1990).

Knowledge Management

Capability

The firm’s ability to accumulate critical knowledge resources and to manage its assimilation and exploitation (Miranda et al., 2011).

Network Linkage Capability The firm’s ability to transmit information, skills and technology, and to receive them from other

departments of the firm, including third parties such as the clients, the suppliers, the consultants, the

technological institutions (Shan and Jolly, 2010).

Technology Acquisition

Capability

The firm’s ability to acquire and to adopt external technology from other parties (Hemmert, 2004).

R&D Capability The firm’s ability to integrate R&D strategy, project implementation, project portfolio management,

and R&D expenditure (Yam et al., 2004).

Project Cross functional

team integration capability

The firm’s ability to coordinate and to integrate all phases of the R&D process and the inter-relations

with the functional tasks of engineering, production and marketing (Camisón and Forés, 2010).

Technology Change

Management Capability

The firm’s ability to accurately predict future technological trends and to response the technology

changes (Jansen et al., 2005).

Product Structural Design

and Engineering Capability

The firm’s ability to design product structure, to build product modularization and to make product

and process compatible (De Toni and Nassimbeni, 2001).

Process Design and

Engineering Capability

The firm’s ability to design process for supporting the manufacturing design and to design the

assembly activities (De Toni and Nassimbeni, 2001).

Manufacturing Capability The firm’s ability to transform R&D result into products, which meet the market’s need as required

design, and able to produce (Yam et al., 2004).

Market Capability The firm’s ability to sell products on the basis of the understanding of customers’ need, the

competitive environment, costs and benefits, and the acceptance of the innovation (Yam et al.,

2004).

7. References Alegre, J. and Chiva, R. (2008). Assessing the impact of organizational learning capability on

product innovation performance: An empirical test. Technovation, 28(6), 315–326.

Antony, J., Leung, K., Knowless, G. and Gosh, S. (2002). Critical success factors of TQM implementation in Hong Kong industries. International Journal of Quality and Reliability Management, 19(5), 551–566.

Amabile, T., Conti, R., Coon, H., Lazenby, J. and Herron, M. (1996). Assessing the work environment for creativity. Academy of Management Journal, 39 (5), 1154–1184.

Baark, E., Lau, Antonio, K.W., William, L. and Sharif, N. (2011). Innovation Sources, Capabilities and Competitiveness: Evidence from Hong Kong Firms. Paper presented at the DIME (Dynamics of Institutions & Markets in Europe) Final Conference, 6-8 April, 2011, Maastricht, 1-40.


99

Barney, J.B. and Clark, D.N. (2007). Resource-based Theory Creating and Sustaining Competitive Advantage, Oxford University Press, Oxford.

Bessant, J., Lamming, R., Noke, H. and Phillip, W. (2005). Managing innovation beyond the steady state. Technovation, 25(12), 1366-1376.

Burgelman, R., Maidique, M.A. and Wheelwright, S.C. (2004). Strategic Management of Technology and Innovation. McGraw-Hill, New York, 8-12.

Camisón, C. and Forés, B. (2010). Knowledge absorptive capacity: New insights for its conceptualization and measurement. Journal of Business Research, 63, 707–715.

Camisón, C. and Villar-López, A. (2012). Organizational innovation as an enabler of technological innovation capabilities and firm performance. Journal of Business Research. In Press, corrected proof, available online 22 Jun, 2012.

Ceylan, C. and Koc, T. (2007). Factors impacting the innovative capacity in large-scale companies. Technovation, 27, 105-114.

Cohen, W.M. and Lavinthal, D.A. (1990). Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly, 35, (1), 128-152.

Chesbrough, H.W. (2004). Managing open innovation. Research Technology Management, 47, (1), 23–26.

Chiesa, V., Coughlan, P. and Voss, C.A. (1996). Development of a technical innovation audit. Journal of Product Innovation Management, 13(2), 105–136.

Chiesa, V., Coughlan, P. and Voss, C.A. (1998). Development of a technical innovation audit. IEEE Engineering Management Review, 26(2), 64-91.

Chiva, R. and Alegre, J. (2007). Measuring organizational learning capability among the workforce. International Journal Manpower, 28(3/4), 224-242

Chiu, Y.J., Chen, H.C., Tzeng, G.H. and Shyu, J.Z. (2006). Marketing strategy based on customer behavior for the LCD-TV. International Journal of Management and Decision Making, 7 (2/3), 143-165.

Christensen, J.F. (1995). Asset profiles for technological innovation. Research Policy, 24, 727–745.

De Toni, A. and Nassimbeni, G. (2001). A method for the evaluation of suppliers’ co design effort. International Journal of Production Economics, 72(2), 169-180.

Dobni, C.B. (2008). Measuring innovation culture in organizations, The development of a generalized innovation culture construct using exploratory factor analysis. European Journal of Innovation Management, 11(4), 539-559

Forsman, H. (2011). Innovation capacity and innovation development in small enterprise, A comparison between the manufacturing and service sector. Research Policy, 40,


739-750.

Garrison, G. (2009). An assessment of organizational size and sense and response capability on the early adoption of disruptive technology. Computers in Human Behavior, 25, 444–449.

Gebauer, H., Worch H. and Truffer B. (2012). Absorptive capacity, learning processes and combinative capabilities as determinants of strategic innovation. European Management Journal, 30, 57-73.

Guan, J. and Ma, N. (2003). Innovation capability and export performance of Chinese firms. Technovation, 23(9): 737-747.

Guan, J.C., Yam, R.C.M., Mok, C.K. and Ma N. (2006). A study of the relationship between competitiveness and technological innovation capability based on DEA models. European Journal of Operational Research, 170, 971–986.

Hansen, J. A. (2001). Technology innovation indicators, a survey of historical development and current practice. In: Feldman, M.P., Link, A. N. (Eds.). Innovation Policy in the Knowledge-Based Economy, 73-103.

Hemmert, M. (2004).The influence of institutional factors on the technology acquisition Performance of high-tech firms: survey results from Germany and Japan. Research Policy, 33, 1019–1039.

Ho, Y.C., Fang, H.C. and Lin, J.F. (2011). Technological and design capabilities: is ambidexterity possible? Management Decision, 49 (2), 208 – 225.

Huang, H.C. (2011). Technological innovation capability creation potential of open innovation: a cross-level analysis in the biotechnology industry. Technology Analysis & Strategic Management, 23(1), 49-63.

Isaksen, S.G., Lauer, K.J. and Ekvall, G. (1999). Situational outlook questionnaire: a measure of the climate for creativity and change. Psychological Reports, 85, 665–674.

Jansen, J., Van den Bosch, F. and Volberda, H. (2005). Managing potential and realized absorptive capacity: how do organizational antecedents matter. The Academy of Management Journal, 48(6), 999-1015.

Kim, C. and Song, J. (2007). Creating new technology through alliances: an empirical investigation of joint patents. Technovation, 27 (8), 461–470.

Kim, K.K., Lee B.G., Park, B.S. and Ohm, K.S. (2011). The effect of R&D, technology commercialization capabilities and innovation performance. Technological and Economic Development of Economy, 17(4), 563-578.

Lane, P.J. and Koka, B.R. (2006). The reification of absorptive capacity: a critical review and rejuvenation of the construct. Academy of Management Review, 31 (4), 833–863.

Lee, H., Lee, S. and Park, Y. (2009). Selection of technology acquisition mode using the


101

analytic network process. Mathematical and Computer Modelling, 49, 1274-1282.

Lee, H.I., Kang, H.Y., Hsu, C.F. and Hung, H.C. (2009). A green supplier selection model for high-tech industry. Expert System Application, 36 (4), 7917-7927.

Levitas, E. and McFadyen, M.A. (2009). Managing liquidity in research-intensive firms: signaling and cash flow effects of patents and alliance activities. Strategic Management Journal, 30 (6), 659–678.

Lin, B.-W. (2004). Original equipment manufacturers (OEM) manufacturing strategy for network innovation agility: the case of Taiwanese manufacturing networks. International Journal Production Research, 42(5), 943–957.

Lin, C., Wu, Y-J., Chang, C., Wang, W. and Lee, C-Y. (2012). The alliance innovation performance of R&D alliances - the absorptive capacity perspective. Technovation, 32(5), 282–292.

Lin, C. L. and Tzeng, G. H. (2009). A value-created system of science (technology) park by using DEMETEL. Expert Systems with Applications, 36, 9683-9697.

Liou, J.J.H., Tzeng, G.H. and Chang, H.C. (2007). Airline safety measurement using a novel hybrid model. Journal of Air Transport Management, 13 (4), 243-249.

Martin, E.C. and Terblanche, F. (2003). Building organizational culture that stimulates creativity and innovation. European Journal of Innovation, 6(1), 64-74.

Miranda, S.M., Lee J-N. and Lee J-H. (2011). Stocks and flows underlying organizations’ knowledge management capability: Synergistic versus contingent complementarities over time. Information & Management, 48 (8), 382-392.

Mu, J. and Benedetto, C.A.D. (2011). Strategic orientations and new product commercialization: mediator, moderator, and interplay. R&D Management, 41(4), 337-359.

Nassimbeni, G. and Battain, F. (2003). Evaluation of supplier contribution to product development: fuzzy and neuro-fuzzy based approaches. International Journal of Production Research, 41(13), 2933-2956.

O’Regan, N., Ghobadian, A. and Sims, M. (2006). Fast tracking innovation in manufacturing SMEs. Technovation, 26(2), 251–261.

Perdomo-Ortiz, J., González-Benito, J. and Galende, J. (2009). The intervening effect of business innovation capability on the relationship between total quality management and technological innovation. International Journal of Production Research, 47(18), 5087–5107.

Romijin, H. and Albaladejo, M. (2002). Determinants of innovation capability in small electronics and software firms in Southern England. Research Policy, 31, 1053-1067.


Shan, J. and Jolly, D. R. (2010). Accumulation of Technological Innovation Capability and Competitive Performance in Chinese firms: A quantitative study. The International Association for Management of Technology (IAMOT) in Cairo, Egypt, 1-21.

Shieh, J. I., Wu, H. H. and Huang, K. K. (2010). A DEMATEL method in identifying key success factors of hospital service quality. Knowledge-Based Systems, 23, 277-282.

Spithoven, A., Clarysse, B. and Knockaert, M. (2010). Building absorptive capacity to organize inbound open innovation in traditional industries. Technovation, 30(2), 130–141.

Spyropoulou, S. and Kyrgidou, L.P. (2012). Drivers and Performance Outcomes of Innovativeness: An Empirical Study. British Journal of Management.

Teece, D. J. (2007). Explicating dynamic capabilities: The nature and micro-foundations of (sustainable) enterprise performance. Strategic Management Journal, 28(8), 1319–1350.

Tidd, J. (2006). A review of innovation models. Discussion Paper, Imperial College London: 1-16.

Türker, M.V. (2012). A model proposal oriented to measure technological innovation capabilities of business firms – a research on automotive industry. Procedia - Social and Behavioral Sciences, 41, 147 – 159.

Tzeng, G.H., Chiang, C.H. and Li, C.W. (2007). Evaluating intertwined effects in e-learning programs: a novel hybrid MCDM model based on factor analysis and DEMATEL. Expert Systems with Applications, 32 (4), 1028-1044.

Voudouris, I., Lioukas, S., Iatrelli, M. and Caloghirou, Y. (2012). Effectiveness of technology investment: Impact of internal technological capability, networking and investment’s strategic importance. Technovation, 32(6), 400-414

Wang, C.H., Lu, I.Y. and Chen, C. B. (2008). Evaluating firm technological innovation capability under uncertainty. Technovation, 28(6), 349-363.

Wonglimpiyarat, J. (2010). Innovation index and the innovation capacity of nation. Futures, 42 (3), 788-801.

Wu, W.W. and Lee, Y.T. (2007). Developing global managers’ competencies using the fuzzy DEMATEL method. Expert Systems with Applications, 32 (2), 499-507.

Yam, R.C.M., Guan, J.C., Pun, K.F. and Tang, E.P.Y. (2004). An audit of technological innovation capabilities in Chinese firms: some empirical finding in Beijing, China. Research Policy, 33(8), 1123-1140.

Yam, R.C.M., Lo, W., Tang, E.P.Y. and Lau A.K.W. (2011). Analysis of sources of innovation, technological innovation capabilities, and performance: An empirical study of Hong Kong manufacturing industries. Research Policy, 40, 391-402.

Yang, J. (2012). Innovation capability and corporate growth: An empirical investigation in


103

China. Journal of Engineering and Technology Management, 29(1), 34–46.

Yang, L-R. (2013). Key practices, manufacturing capability and attainment of manufacturing goals: The perspective of project/engineer-to-order manufacturing. International Journal of Project Management, 31(1), 109-125.

Yang, Y.P., Shieh, H. M., Leu, J.D. and Tzeng, G.H. (2008). A novel hybrid MCDM model combined with DEMATEL and ANP with applications. International Journal Operational Research, 5(3), 160-168.

Yu, R. and Tseng, G.H. (2006). A soft computing method for multi-criteria decision making with dependence and feedback. Applied Mathematics Computation, 180(1), 63-75.

Zahra, S.A. and George, G. (2002). Absorptive capacity: a review, reconceptualization, and extension. Academy of Management Review, 27(2), 185–203.

Zeng, S.X., Xie, X.M. and Tam, C.M. (2010). Relationship between cooperation networks and innovation performance of SMEs. Technovation, 30(3), 181-194.

Zhang, Z., Waszink, A. and Wijngaard, J. (2000). An instrument for measuring TQM implementation for Chinese manufacturing companies. International Journal of Quality & Reliability Management, 17 (7), 730–755.

D. Sumrit is a Ph.D. Candidate of Technopreneurship and Innovation Management Program, Graduate School, Chulalongkorn University, Bangkok, Thailand. He received his B.Eng in Industrial Engineering from Kasetsart University, an M.Eng from Chulalongkorn University and MBA from Thammasat University.

Dr. P. Anuntavoranich is an Assistant Professor of Department of Industrial Design at Faculty of Architecture, Chulalongkorn University, and he is Director of Technopreneurship and Innovation Management, Chulalongkorn University. He received his Ph.D. (Art Education) from the Ohio State University Columbus, OH., USA. His specialist is creative design and innovation management.

Peer Review: This article has been internationally peer-reviewed and accepted for publication according to the guidelines given at the journal’s website.


http://TuEngr.com

Call-for-Papers:

ITJEMAST invites you to submit high quality papers for full peer-review and possible publication in areas pertaining to our scope including engineering, science, management and technology, especially interdisciplinary/cross-disciplinary/multidisciplinary subjects.

Next article continue on next page

http://tuengr.com/index.html


*Corresponding author (A. Jala). Tel/Fax: +66-2-5644440-59 Ext. 2450. E-mail address: [email protected]. 2013 International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/105-110.pdf

105


http://TuEngr.com

The Effect of the 2,4-Dichlorophenoxy Acetic Acid, Benzyl Adenine and Paclobutrazol, on Vegetative Tissue-Derived Somatic Embryogenesis in Turmeric (Curcuma var. Chattip)

Anchalee Jala a* a Department of Biotechnology, Faculty of Science and Technology, Thammasat University, THAILAND A R T I C L E I N F O

A B S T RA C T

Article history: Received 09 October 2012 Received in revised form 06 November 2012 Accepted 16 January 2013 Available online 22 January 2013 Keywords: Paclobutrazol; 2,4-dichlorophenoxy acetic acid (2,4-D); Benzyl adenine (BA), Somatic embryogenesis.

The nodal explants of Curcuma var. Chattip could develop callus after in vitro culturing and transplanting within 4 weeks in MS medium supplemented with 1.0 mg l-1 2,4-D, although this optimal if the media was further supplemented with 5.0 mg l-1 BA, obtaining the highest number of new shoots in 6 weeks. MS medium supplemented with 0.01 mg l-1 paclobutrazol and 15% (v/v) coconut water was found suitable for regenerating the highest number of new shoots (7.25 shoots). The number of leaves per plantlet, length of leaves, and length of petrioles were significantly reduced (p≤ 0.05) when increased concentrations of paclobutrazol and especially paclobutrazol with 15 % (v/v) coconut water. However further experimentation is required to evaluate the dose-response and the interaction between coconut water and paclobutrazol. In contrast, there were no significant difference in leaf width in all treatments.


1. Introduction Turmeric (Curcuma var. chattip), a herbaceous plant of the Zingiberaceae (ginger) family

(Purseglove,1972), is an economically important cultivated species in Thailand being grown

for cut flowers, decoration and landscaping. Although propagated by underground rhizomes,


106 Anchalee Jala

the rate of rhizome multiplication and growth is very low, making this non-viable for large

scale economic production. Moreover, many diseases and pests, particularly soft rot caused

by Pythium spp. (Jantan et al., 2003) as well as bacterial wilts are consistently threatening

Curcuma var. chattip. This problem is compounded by its slow propagation rate which

seriously impedes on the ability to replace diseased plants quicker than infection rates.

To aid the rapid and large scale propagation of this plant, in vitro formation of storage

organs such as rhizomes that can be directly transferred to the field without any

acclimatization has been reported (Balachandran et al., 1990). However, further improvement

of the protocol to obtain larger and more vigorous plantlets is required in order to approach an

economically or logistically viable method.

The presented investigation was carried out to examine the effects of the plant growth

regulators palcolbutrazol and benzyl adenine (BA) in the presence of napthaleneacetic acid

(NAA) and coconut water, upon the formation of callus and multiply new plantlets.

2. Materials and Methods Spouted immature shoots (ca. 1 cm. long) were collected and used as explants. They

were washed thoroughly in running tap water and soaked with liquid detergent (teepol

solution) followed by rinsing in tap water for 2 min. For surface sterilization, explants are

rinsed in 10 % (v/v) Clorox solution for 10 min and 5 % (v/v) Clorox solution for 10 min and

finally soaked with sterile distilled water three times to remove traces of Clorox. Immature

shoots are trimmed to remove excess tissue. Murashige and Skoog medium (MS) fortified

with 0.25 % (w/v) and 4% (w/v) final concentration of gelrite and sucrose, respectively. MS

medium supplemented with 2,4-dichlorophenoxy acetic acid (2,4-D), BA, and coconut water

is used as basal medium for callus. The pH of medium was adjusted to 5.6 by using 0.1 N

NaOH or 0.1 HCl and autoclaving at 121 ํC for 20 min to sterilize. All cultures were

incubated under 16hours photoperiod (irradiance of 36 µmole m-2 S-1) with temperature 25 ±

1°C. We observed samples at regular intervals and scored for callus and shoot growth, with

10 independent replicates being used for each experimental treatment.

3. Statistical Analysis This experiment used CRD (Completely Randomized Design). The analysis of variance

between the means was conducted by using Duncan’s multiple range test (Duncan, 1955).


107

4. Result and Discussion After 4 weeks of in vitro culture, MS medium with 1.0 mg l-1 2,4-D and no BA showed

a remarkable degree of callus formation. Indeed, the callus obtained from in vitro culture was faster growing, delicate, mostly spongy, and white creamy in color (Table 1). In addition, with 0.5 mg l-1 2,4-D concentration, slightly more explants grew callus, and grew more callus per explants, when using MS medium supplemented with 0.5 mg l-1 and 1.0 mg l-1 BA. This studies is in broad agreement with Nayak (2000) and Hosoki andSagawa(1977) who reported the callus induction from leaf base as explants of curcuma aromatica using 1.5 mg l-1 2,4-D and 1.0 mg l-1BA. However, in this system it is yet to be evaluated if BA will enhance the highest level of callus production seen with 1.0 mg l-1 2,4-D.

Table 1: The average number of explants induced to form callus (as 5%) and

degree of callus per explants after culturing for 4 weeks. MS medium plus % of explants

induced callus Degree of Callus response

2,4-D 0.5 mg l-1 53.4 a 2,4-D 1.0 mg l-1 78.2 c 2,4-D 2.0 mg l-1 56.7 a

2,4-D 0.5mg l-1+BA 0.1mg l-1 49.86 a 2,4-D 0.5 mg l-1+BA 0.5mg l-1 66.67 b 2,4-D 0.5 mg l-1+BA 1.0mg l-1 61.34 b

a – Slight callusing, b – more callusing, and c – Profuse callusing

Callus obtained from in vitro cultivation (Table 1) were used to investigate the influence

of growth regulators on the induction of somatic embryogenesis.

Four-week old callus were used for multiplying new shoot by culturing them on MS basal

medium supplemented with 0.1 mg l-1NAA, 15% (v/v) coconut water and varying levels

(viz.1.0, 2.0,3.0, 4.0, and 5.0 mg l-1) of BA. All independent in vitro cultures suggested that

the observed effect of BA was likely to be mediated via inducing new shoots, as summarized

in Table 2. After six weeks, new shoots regenerated from callus. MS medium supplemented

with 0.1 mg l-1,15% CW and 5.0 mg l-1 BA gave the highest average levels of new shoots

attained per callus (5.1 ± 0.54) as worked of Dekker et al. (1991) did with ginger but Jala

(2011) cultured Curcuma longa in MS medium supplemented with 2 mg l-1 BA.

Callus obtained from above mentioned medium are used to investigate the influence of

the growth retardant (paclobutrazol) on the growth of somatic embryogenesis.

108 Anchalee Jala

After six weeks, callus was regenerated to new shoots on MS medium supplemented with

0.1 mg l-1 NAA and 5.0 mg l-1 BA. On this condition, callus could form somatic embryos and

germinated (Table 2) within 6 weeks, allowing this system to investigate the effect of

paclobutrazol.

Table 2: Number of new shoots regenerated from callus that are cultured on MS medium

supplemented with 0.1 mg l-1 NAA 15% (v/v) coconut water with varied BA concentrations. NAA (mg l-1) CW(%) BA(mg l-1) Number of new shoots*

0.1

15

1.0 1.2 e 2.0 1.8 d 3.0 2.5 c 4.0 3.2 b 5.0 5.1 a

* Means within columns with different letters are significantly different by using DMRT at the ( p ≤ 0.05) level.

Table 3: The effect of paclobutrazol and coconut water on somatic embryo regeneration from callus within 8 weeks of in vitro culturing.

MS+NAA0.1mgl-1, BA5.0mgl-1, 4 %sucrose (control) plus

No. of multiple shoots*

No. of leaves

per plantlet *

The leaf length (cm)*

The leaf width (cm)ns

Length of leaf petriole (cm)*

Control ( no Paclobutrazol ) 4.00 c 7.82 a 3.80ab 0.73 5.53 a Paclobutrazol 0.01 mg l-1 3.00 d 7.67 a 3.65 b 0.84 4.11 b Paclobutrazol 0. 1 mg l-1 5.00 b 6.92 ab 3.37 bc 0.78 3.36 cd Paclobutrazol 1.0 mg l-1 4.00 bc 7.08 ab 3.53 b 0.86 3.97 bc Paclobutrazol 5.0 mg l-1 3.75 c 5.86 c 3.97 a 0.75 4.40 b Paclobutrazol 10.0 mg l-1 4.50bc 7.54 a 3.25 c 0.85 3.61 c Paclobutrazol 0.01 mg l-1+ cw15% 7.25 a 4.57 c 4.33 a 0.87 3.60 c Paclobutrazol 0.1 mg l-1 + cw15% 3.75 c 5.48 c 3.66 b 0.82 3.70 c Paclobutrazol 1.0 mg l -1 +cw15% 6.25 a 5.33 c 4.08 a 0.82 3.41 cd Paclobutrazol 5.0 mg l -1 +cw15% 3.00 de 6.60 bc 3.93 a 0.79 2.51 de Paclobutrazol 10.0 mgl-1 +cw15% 2.5 e 2.65d 2.15 d 0.91 e *Means within columns with different letters are significantly different from each other (p ≤ 0.05),

ns - no significant difference for that trait across all culture conditions.

4.1 Effect of plant growth (retardant – paclobutrazol) on somatic embryo The callus were used to investigate the influence of the growth retardant (paclobutrazol)

on growth of somatic embryogenesis by cultured them in MS basal medium supplemented

with 0.1 mg l-1 NAA, 5 mg l-1 BA, with and without 15% (v/v) coconut water (CW) and

varied concentrations of paclobutrazol (viz. 0.01, 0.1, 1.0, 5.0 and 10 mg l-1) for 8 weeks.

During this time the number of multiple shoots, number of leaves, leaf length, leaf width, and

length of each leaf petriole were measured as shown in Table 3.


109

Results indicated that paclobutrazol and 15% (v/v) CW both induced a significant

difference (p≤ 0.05) on plantlet regeneration as measured by four out of five traits (Table 3).

Only the average leaf width showed no statistically significant differences between the

different treatments. However, length of leaf petriole and length of leaf were not significantly

different. The numbers of leaves per plantlet were all broadly decreased by increasing

paclobutrazol concentrations, especially in the presence of coconut water where the broadly

similar effects at lower concentrations of paclobutrazol in the presence of 15% (v/v) coconut

water.

However, with respect to the number of adventitious shoots the situation is less clear with

no clear trend. In general, Paclobutrazol alone showed no clear if any concentration dependent effect upon the number of multiple shoots per explants but, the data for Paclobutrazol in the presence of 15% (v/v) coconut water is more erratic with the two highest average number of shoots per explants (7.25 shoots and 6.25 shoots for 0.01 and 1.0 mg/l Paclobutrazol, respectively) intermixed amongst lower values with the lowest attained with 10 mg/l Paclobutrazol. Direct somatic embryogenesis has been reported in many other plants (e.g. Nadguada et al.1978, Salvi et al. 2000, Shiqurkar et al. 2001, Yasuda et al. 1988).

5. Conclusion Immature nodal explants of Curcuma var. Chattip could develop callus after in vitro

culturing in MS medium supplemented with 1.0 mg l-1 2,4-D within 4 weeks. Subculturing callus in MS basal medium supplemented with 0.1 mg l-1 NAA, 15% coconut water and 5.0

mg l-1 BA yielded the highest number of new shoots. These could be further induced somatic embryo regeneration and leaf formation by in vitro culturing in MS medium supplemented with 0.1 mgl-1 NAA, 5.0 mgl-1 BA, 15% (v/v) coconut water and (0.01 to10.0 mg l-1) paclobutrazol. Although equivocal as complex interactions may be occurring, the data suggests that 0.01 and 1.0 mg l-1 paclobutrazol and 15% (v/v) CW were the most suitable conditions leaded to formation of the highest number of new shoots (7.25 and 6.25 shoots per explants, respectively) within 6 weeks.

6. References Balachandran S.M., Bhat S. and Chandel K. 1990. In vitro clonal multiplication of turmeric

(Curcuma spp.) and ginger (Zingiber officinale Rosc.). Plant Cell Rep, 8: 521-524.

110 Anchalee Jala

Dekkers, A.J.; Rao A.N. and Goh C.J. 1991 In vitro strorage of multiple shoot cultures of gingers of ambient temperatures of 24-29 C, Sci Hort, 47:157–168.

Hosoki, T. and Y. Sagawa. 1977. Clonal propagation of ginger (Zingiber Rosc.) Through tissue culture. Sci Hort., 12: 451–452.

Jala, A. 2011. Effects of NAA BA and Sucrose on Shoot Induction and Rapid Micropropagation by Trimming Shoot of Curcuma Longa L. INT TRANS J ENG MANAG SCI TECH, 3(2):101-109.

Jantan, I.B., Yassin M.S.M., Chin C.B., Chen L.L., Sim N.L. 2003. Antifungal activity of the essential oils of nine Zingiberaceae species. Pharmaceutical Biology, 41: 392-397.

Jen-kun Lin, Shoei-Yn and Lin-Shiau. 2000. Mechanisms of Chemoprevention by Curcumin. Proc. Natl. Sci. Counc. ROC (B), 25(2): 59 – 66.

Jitoe, A., Toshiya. Masuda, I. G. P. Tengah, Dewa N. Suprapta, I. W. Gara, Nobuji. Nakatani. 1992. Antioxidant activity of tropical ginger extracts and analysis of the contained curcuminoids. J. Agri. Food Chem., 40:1337- 1340.

Kikuzaki, H. and Nakatani, N. 1993. Antioxidant effects of some ginger constituent. J. Food Sci, 58:1407-1410.

Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays with Tobacco tissue cultures. Physiol. Plant, Compenhagem, 15:473–479.

Nadguada, R.S.Mascarenhas,A.F., Hendre, R.R. and Jagannathan, V. 1978. Rapid multiplication of turmeric (Curcuma longa L). Ind. J. Exp. Biol., 16:120-122.

Nayak, S. 2000. In vitro multiplication and microrhizome induction in Curcuma aromatica Salisb. Plant Growth Regul. 32:41-47.

Salvi, N.D., George, L. and Eapen, S. 2000. Direct regeneration of shoot from immature inflorescence culture of turmeric. Plant Cell Tiss. Org. Cult., 62:235-238.

Shigurkar, M.V., John, C.K. and Nadgauda, R.S. 2001. Factors affecting in vitro microrhizome production in turmeric. Plant Cell Tiss. Cult. 64:5-11.

Yasuda K., Tsuda T., Shimizu H., and Sugaya A. 1988. Multiplication of Curcuma species by tissue culture, Planta Medica, 54: 75-79.

Dr.Anchalee Jala is an Associate Professor in Department of Biotechnology, Faculty of Science and Technology, Thammasat University, Rangsit Campus, Pathumtani , THAILAND. Her teaching is in the areas of botany and plant tissue culture. She is also very active in plant tissue culture research.


*Corresponding author (S.Thammaniwit). Tel/Fax: +66-2-5643001 Ext.3095, 3038. E-mail address: [email protected]. 2013 International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/111-128.pdf

111


http://TuEngr.com

Application of Bender’s Decomposition Solving a Feed–mix Problem among Supply and Demand Uncertainties Somsakaya Thammaniwit a*and Peerayuth Charnsethikul a

a Department of Industrial Engineering, Faculty of Engineering, Kasetsart University, THAILAND.

A R T I C L E I N F O

A B S T RA C T

Article history: Received 23 August 2012 Received in revised form 19 December 2012 Accepted 16 January 2013 Available online 24 January 2013 Keywords: Large-scale Feed-mix Problem; Bender’s Decomposition; Two-stage Stochastic Programming.

The feed-mix problem is primarily transformed into a mixing situation applying a mathematic formulation with uncertainties. These uncertainties generate the numerous expansions of alternative constraint equations. The given problem has been formulated as mathematic models which correspond to a large-scale Stochastic Programming that cannot be solved by the most popular ordinary calculation method, Simplex Method: LINPROG. This research aims to investigate effective methodology to reveal the optimal solution. The authors have examined the method of Bender’s decomposition: BENDER and developed both methods into MATLAB® program and calculated comparatively. The results revealed that the nearest optimal solutions can be determined by means of a Two-stage Stochastic Programing incorporated with Bender’s decomposition at the most intensive number of uncertainties and take less calculation time than by LINPROG.


1 Introduction

Many animal food mixing industries are confronted with a decision making problem on

an appropriate recipe. That is to say that the determination of raw materials which contain

various kinds of feed ingredients added to the process are influenced by expectations of


112 Somsakaya Thammaniwit, and Peerayuth Charnsethikul

obtaining a product with lower costs, standardization, or surpass nutritive requirement. Such

a feed mix problem is complex and cannot be solved by traditional calculation methods.

There are many approaches which have been applied, such as Pearson’s square method which

is very suit for only two-feed ingredients to be mixed [9] and Trial and Error which is one of

the most popular means for feed formulation but it consumes a lot of time for calculation

[4],[9]. Classical Linear Programming (LP) is widely used for modeling the animal feed

problem. The normal objective in formulating the feed mix is to minimize cost subject to

adequate nutrient ingredients (input raw materials) and the required nutrient constraints

(output nutrient values) [1]. However, due to the various constraints that need to be

conserved, the problem has been extended to become a large-scale problem with

uncertainties. Therefore, when using LP method, it is difficult to determine a good balance of

nutrients in the final solution. The constraints in LP are also rigid leading to an infeasible

solution [1], [2]. However, LP has a positive highlight as a deterministic approach, because it

can provide the best solution of hundreds of equations simultaneously [14]. By Stochastic

approaches, there are also various methods that have been applied for such a complex

problem e. g: Chance Constraint (CCP) and Quadratic Programming (QP), Risk Formulation,

and Genetic Algorithm (GA). In addition to these mentioned methods, there are also some

methods with other possible algorithms such as Integrated LP and Dynamic Programming

(DP), Integrated LP and Fuzzy, Integrated GA and Fuzzy, Integrated GA and Monte Carlo

Simulation. All of these methods are arranged as Integrated approaches [13].

This research does not take into consideration all the above mentioned issues, but aims to

investigate another new effective calculation method for the feed mix problem and proposes

the application of Two-stage Stochastic Linear Programming (TSL) incorporated with the

method of Bender’s Decomposition (BENDER). Hence, this paper describes the preliminary

stage of mathematic formulation, the setup of matrix systems and program development,

MATLAB@ program, and represents the optimization results of a case study.

2 Problem Analysis and Methodology

The classical diet problem is considered as a Linear Programming problem with general

LP matrix: TMin Z = C X, Subject to AX = b, and X 0 for all.≥ Because of the limitation of

the calculation devices, the prior results were revealed without regard for some variables with


113

high variance constraint coefficients. Nowadays, because of the higher performance of

computational calculation, the development of the mentioned LP model when the system

uncertainties are taken into account can be written as T T TMin Z = C X + g U + h V

subject to AX + U - V = b x , u, v 0, ≥

(1)

where TC X represents the main cost and T Tg U + h V the additional corrective costs

of materials supplied subject to AX + U - V = b , where A is the coefficient of the

decision variable X (material quantity), U and V stand for the least and the excess mixed

output quantities respectively. Awareness of nutrition values contained in U and V have an

effect on the RHS as well. This issue will be discussed in the subtopic 2.3 later. Meanwhile

the Two-stages Linear Programming Model [6], [7], [10], [11], [16] can be written as:

Maximize Qk n

z = E [c ] x + P [ c x ]qj j qj qjq=1j=1 j=k+1∑ ∑ ∑ (2)

Subject to

1st.Stage k

a x = b for i = 1, 2,...,ij j ij=1r∑ (3)

2nd.Stage k k

a x + a x = bqij j qij qj qij=1 j=1∑ ∑ (4)

i = r +1,...m for q = 1,2,...,Q x 0; x 0,j qj≥ ≥

where Pq probability of occurrences of scenerio q (q=1,2,…,Q) xqj extent variable in the 2nd.stage at constraint j by event q

Notation 1. The value of each random element is independent of the levels of all x j

2. The levels of x j for j = 1, 2 ,…, k ≤ n must be fixed at the 1st. stage

3. The constraint i = 1, 2… r contains only the 1st.stage variables, and the associated aij


and bi are known with certainty.

4. The variables of xqj in the 2nd. Stage, where i = r +1, 2, …, m and q = 1, 2, …, Q

5. The values of cqj , aqij and f bqi , for i = r+1, 2, …, m and j = 1, 2 ,…, n are

represented by the set of ( cqj , aqij , bqi ) with probability Pq , for q = 1, 2, …, Q

2.1 The problem analysis The animal food mixing as shown below in Table 1 was discussed by a production team.

The problem is to determine the optimal quantities of the three main input raw materials to be

added to the mixing process.

Table 1: The input ingredient amounts represented as x1, x2, and x3 were unknown. Protein (%) Calcium (%) cost / unit

x1 Crushed dried fish 51-53 10-11 70Baht x2 Tapioca 2-3 5-7 40Baht x3 Sorghum - 8-9 23Baht m Market required between19-20 Not less than 8 Note: Baht is the currency used in Thailand (As of January 2013, 30Baht = US$1).

The market demand d = 1.000 ± 0.001 unit weigh t

Let x1, x2 and x3 be the non-negative quantities of the crushed dried fish, tapioca and

sorghum respectively. They are mixed to yield 1 unit of the minimum cost diet that satisfies

all the specified nutritional requirements m =2 prescribed as following:

protein not in excess of 20 % (Upper Boundary)

not less than 19 % (Lower Boundary)

calcium not less than 8 %

Incremental corrective action cost of nutrient value, respectively

FEXD = fexd = 7 Baht/ Unit of protein

FLES = fles = 5 Baht/ Unit of protein

FEXD = fexd = 7 Baht/ Unit of calcium

FLES = fles = 4 Baht/ Unit of calcium

Incremental corrective action cost of ingredient (raw material), respectively

FEXDD = fexdd = 30 Baht/ kg

FLESD = flesd = 5000 Baht/ kg


115

2.2 Methodology Such a problem is a typical large-scale stochastic linear programming with full system

uncertainties (tolerances). The decision values of variables xj are decided by the coefficient

of a ± tol, a ± tol, ... , a ± toln1 2 , right-hand-side ( RHS) parameter vector, the nutrition values

bi of b ±tol, b ±tol, ... , b ±tolm1 2 and moreover, the size of the market demand d ±tol as

Figure1 below:

Figure 1: The problem type: A-B-D Uncertainty [15].

2.3 Model Formulation To formulate the given problem in the form of TSL_ model and to allocate the

calculation matrix system, some occurrence possibilities were assumed as follows:

Assumptions A):

PN the occurrence possibility for each nutrient constraint

PD the occurrence possibility for each demand constraint

PN = PD = P (Point) for this case study, the distributions of the probability of PN and PD are

assumed to be uniform distributions. Thus, the possibilities PN and PD will be

equal and also equal to P (Point) where the P (Point) is the initial input number

for allocating the division number of all system uncertainty intervals.

E incremental event step, for this study, E = PN x PD

Constr Constraint, C = m x Event + PD

Var Variable, Var = n + 2Constr


Assumptions B):

The lower and upper boundaries of all constraint variables are allocated from the middle

point of their tolerances:

(mid) (tol) (mid) (tol)ij ij ij ij ij

(mid) (tol) (mid) (tol)i i i i i

(mid) (tol) (mid) (tol)k

(mid) (tol) (mid) (tol) (mid) (tol)ij ij i i

a a - a ; a + a

b b - b ; b + b

d d - d ; d + d

a ; a ; b ; b ; d and d R

⎡ ⎤∈ ⎣ ⎦⎡ ⎤∈ ⎣ ⎦⎡ ⎤∈ ⎣ ⎦

∈

(5)

Referring to the previous notations (1), (2), (3), (4) and (5) including analyzing the above

given diet problem in Table 1, the calculation models were developed to be (6), (7) and (8).

The objective function of the given problem was to minimize the total cost z min. which is the

sum of the raw material unit cost cj multiplied by the amount of xj, plus the sum of all

necessary incremental corrective action costs for both qualitative and quantitative values not

meeting the specification, Equation (6). The sum of the product of the quantity xj and its

uncertain coefficients ija l including the sum of nutritive slack ui kl and surplus vi kl of each

calculation scenario have to be balanced to the right-hand-side RHS , i.e. the vectors of

required nutritive value of ib l multiplied by the demand d , in Equation (7). Simultaneously,

the LHS, the sum of xj and addition of the slacks uk and surplus vk has to be associated with

the quantity requirement of d. In practice, the demand d cannot be exactly equal to 1 unit, but

rather it comprises an allowance of ± 0.001 in unit weight (0.1% error).

Minimize z n m P P P

c x + (g u + h v ) + (g u + h v )j j i k i k i k i k k k k k=1j=1 i=1k=1 k=1′ ′∑ ∑ ∑ ∑ ∑l l l ll

(6)

Subject to

n

a × x + u - v = b × dij j i k i k i kj=1∑ l l l l ; i k∀ l (7)

k k k

nx + u v = djj=1

′ ′−∑ ; k∀ (8)

x ; u ; v ; u ; v 0j i k i k k k′ ′ ≥l l ; i, j, k∀

Where


117

k denotes the constraint alternatives

j denotes the type of ingredient to be input to mixing process ( j=1,2,…,n ).

i denotes the type of nutrient composition which the market needs ( i=1,2,…,m)

cj denotes the cost factor of raw material j- type (cost/ unit weight = fx )

xj denotes the quantity of raw material j- type (weight unit = kg ).

aij denotes the nutrient value type i in material type j

bi denotes the nutrient value (of product) type i /unit weight.

ui kl denotes the nutrient value (of product) i , at the event l which misses (slack )

the target in the alternative k

vi kl denotes the nutrient value (of product) i , at the event which exceeds

(surplus ) the target in the alternative k

i kg l denotes the expected cost of nutrient value i / unit which misses the target

alternative k . FLES = fles

i kh l denotes the expected cost of nutrient value i / unit which exceeds the target

alternative k . FEXD = fexd

gk denotes the expected cost of ku′ (by lack of demand) = FLESD

hk denotes the expected cost of kv′ (by exceeding demand) = FEXDD

ku′ denotes the lower quantity of raw material in the alternative k

kv′ denotes the excess quantity of raw material in the alternative k

dk denotes the market demand 1.000 weight unit with the standardized allowance

of ± 0.001 in unit weight (0.1 % error) for animal food production.

2.4 Minimum Cost Calculation Model Minimize z =

mFLES FEXD FLESD FEXDD

P P P(fx x + fx x + fx x ) + ( u + v ) + ( u + v )1 1 2 2 3 3 i k i k k k=1i=1 k=1 k=1

′ ′⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅∑ ∑ ∑ ∑l ll (9)

Subject to

n

a × x + u - v = b × dij j i k i k i kj=1∑ l l l l ; i k∀ l

l


k k k

nx + u - v = djj=1

′ ′∑ ; i,k∀

x ; u ; v ; u ; v 0j i k i k k k′ ′ ≥l l ; i, j, k∀

Substitute the given data from Table 1 into the minimal cost calculation model above

( )

2 P P(70x + 40x + 23x ) + ((5,4) u + (7,7) v ) +1 2 3 i k i k=1i=1 k=1P

5000 u + 30 vk kk=1

Minimize Z ⋅ ⋅∑ ∑ ∑

′ ′⋅ ⋅∑

= l ll

Subject to

[51--53]x + [2--3]x + [0]x +u - v = [19--20] ,k1 2 3 1 k 1 k[10--11]x + [5--7]x + [8--9]x + u - v = 8 ,k1 2 3 2 k 2 k

x + x + x + u - v = 1±0.001 k1 2 3 k k' 'x , x , x 0, u , v 0 i, ,k1 2 3 i k i k

∀

∀

′ ′ ∀

≥ ≥ ∀

ll l

ll l

ll l

1st.Event; for 1, =1, k =1i = l (at lower boundary)

51x + 2x + 0x +u v = 19 (0.999)1 2 3 111 11110x + 5x + 8x + u - v = 8 (0.999)1 2 3 211 211x + x + x + u - v = 0.9991 2 3 1 1

− ⋅

⋅

′ ′

2nd.Event; for 2, = 2, k = 2i = l (at middle value)

52x + 2.5x + 0x +u - v = 19.5 (1.000)1 2 3 122 12210.5x + 6x + 8.5x + u - v = 8 (1.000)1 2 3 222 222x + x + x + u - v = 1.0001 2 3 2 2

⋅

⋅

′ ′

3rd.Event; for 3, = 3, k = 3i = l (at upper boundary)

53x + 3x + 0x + u - v = 20 × (1.001)1 2 3 133 13311x + 7x + 9x + u - v = 8 × (1.001)1 2 3 233 233x + x + x + u - v = 1.0011 2 3 3 3′ ′

The above formulations just demonstrate how to set up only three events (l): lower

boundary (LB), middle value, and upper boundary (UB). But, in this research, the calculation

models were planned to be set up throughout the tolerance intervals of all relevant variables


119

and divide those with the same number of P, by which the resolution (res), through all

uncertainty intervals can be definable. One must beware of the LB and UB of each event

which correspond with the given tolerances.

2.5 The TSL incorporated with Bender’s Decomposition The concept of the Bender’s decomposition is to predict the second-stage costs by a

scalar θ and replace the second-stage constraints by cuts, which are necessary conditions

expressed in terms of the first-stage variables x and θ [6], [10], [12]. The initial model can be

modified, and written as shown below:

The 1st.stage: Primal problem

T T TMin Z = C X + g U + h V

subject to AX + U - V = b

x , u, v 0, ≥ (10)

The 2nd.stage: The Dual Bender’s Decomposition

( )

( )

( )

U - V = b - AX

b - AX 0T TMax ω = b - AX Y + C X

Tθ = b - AX

T T Tcond = cond; θ+ Y A-C X b Y

≥

∑

≥⎡ ⎤⎣ ⎦

(11)

( )Min θ+X;

T T TSubject to θ + Y A-C X b Y≥ (12)

Stop condition Check θ = ? No Get 1X, 2X,.....

Yes Get θ and X Min Z = Max ω Stop

Xnewω → →

→ ⇒ ⇒

The selected algorithm to attain a satisfactory solution was the integration between

TSL and Bender’s decomposition. As shown in Figure 2, the main concept of Bender’s

decomposition is to split the original problem into a master problem and a sub-problem,

which in turn decomposes into a series of independent sub problems, one for each. The

latter are used to generate cuts. The X initial is to be randomly selected to substitute in terms of


the constraints inequality equation. If the result, by substitution of X initial, is equal to or

greater than 0, then iy = gi. If the result (substitution of X initial) is less than 0, then iy = - h.

The xinitial and the selected yi are substituted in the Dual equation and in the inequality (to

generate the optimal cut) to attain the maximum value of ω and θ respectively. By

minimization of the + 0Xnew ; subject to newT T Tθ (y A-c ) x b y+ ⋅ ≥ , Equation (12) renders

θ and x new . If the obtained and are equal at the step of convergence test, and are

to be approximately equivalent to zDual which can be obtained as an optimal solution.

Figure 2: TSL incorporated with Bender’s decomposition [6], [10], [11], [15].

2.6 Calculation Tools

The mathematical calculation tool, MATLAB® Program was selected to solve this

problem. The MATLAB®_Software / Verion.2006a and a HP_Pavillion_IntelCore_2Qurd

Inside, No.: 016-120610000, personal computer, at the Department of Industrial Engineering,

θ

ω θ ω θ


121

Thammasat University, Pathumtani were used. The calculation steps were as follows

Step1. In accordance with the mathematic formulation, previous topic 2.4, the matrices

systems were established according to the ordinary simplex method. The initial input

variable matrix was expressed in two possible substitutions: randomization and dual-

simplex algorithm.

Step2. Model building according to the method of Bender’s decomposition followed the

programming flowchart as shown in Figure 2.

Step3. Preparation of MATLAB® programming named: lstart, start, setupmodel, solvemodel;

linMAT, bender, dataABD_Uncertainty.mat, result, pline, selfcat, speye, and vspace

Step4. Creation of data file: a, a_tol, b, b_tol, d, d_tol, fx, fles, fexd, flesd and fexdd

Step5. Program execution by the program named: lstart (input parameter is P-Number).

2.7 Computational Calculation To solve the above formulated problem, the primal-dual Simplex method /LINPROG,

Equations (1) was applied to compare with two-stage stochastic linear programming

incorporated with the method of Bender’s Decomposition, Equations (9), (10), (11), (12)

referred to Equations (2), (3), (4), (6), (7), and (8).

Assumptions:

1. The quantity of each nutrient value in each type of raw material is a continuous

event. The values, which are independent of each other, are uncertain.

However, the values intervals are recognized to be uniform distributions.

2. All scenarios are also uniformly distributed and independent on each other.

As the solving tools for comparative calculation, the results of both selected method can

be collected after the calculation iteration has terminated. The programs for primal-dual

simplex and Bender’s decomposition were developed. The starting program referred to as

‘Start.m’ constitutes the main application to perform the execution of all relevant functions.

The matrix systems corresponding with Equations (9), (10), (11), (12) are established. The

program referred to as ‘linMAT.m’ is a matrix to receive all loaded input data whereas the


program name: ‘Speye.m’ serves to construct a large identity matrix system with the

‘Selfcat.m’ application to await the parameter patterns modification prior to the continuation

of the Bender’s decomposition program ‘Bender.m’ [15].

Table 2: Lower section cut off at the uncertainty number of P=2:2:50.

P Event Constr Var x1_BEN x2_BEN x3_BEN Sx_BEN Z_BEN Ti_BEN x1_LIN x2_LIN x3_LIN Sx_LIN Z_LIN Ti_LIN2 4 10 23 0.3631 0.2420 0.3949 1.0000 48.61 0.4924 0.3631 0.2420 0.3949 1.0000 48.61 2.33994 16 36 75 0.3633 0.2422 0.3946 1.0000 48.60 0.1767 0.3633 0.2422 0.3946 1.0000 48.60 0.05586 36 78 159 0.3632 0.2422 0.3947 1.0000 48.59 0.1995 0.3632 0.2422 0.3947 1.0000 48.59 0.08688 64 136 275 0.3617 0.2761 0.3622 1.0000 48.57 0.1284 0.3617 0.2762 0.3621 1.0000 48.57 0.2207

10 100 210 423 0.3617 0.2768 0.3615 1.0000 48.54 0.1818 0.3617 0.2766 0.3617 1.0000 48.54 0.362012 144 300 603 0.3617 0.2769 0.3614 1.0000 48.52 0.2010 0.3617 0.2769 0.3615 1.0000 48.52 0.877414 196 406 815 0.3618 0.2739 0.3643 1.0000 48.52 0.1472 0.3618 0.2739 0.3643 1.0000 48.52 1.467016 256 528 1059 0.3617 0.2762 0.3621 1.0000 48.51 0.1660 0.3617 0.2762 0.3621 1.0000 48.51 1.192018 324 666 1335 0.3617 0.2761 0.3621 1.0000 48.51 0.1456 0.3617 0.2765 0.3618 1.0000 48.51 0.950120 400 820 1643 0.3617 0.2766 0.3617 1.0000 48.51 0.1393 0.3617 0.2766 0.3617 1.0000 48.51 0.260722 484 990 1983 0.3617 0.2764 0.3619 1.0000 48.50 0.1514 0.3617 0.2768 0.3616 1.0000 48.50 0.361124 576 1176 2355 0.3617 0.2768 0.3615 1.0000 48.50 0.1820 0.3617 0.2769 0.3614 1.0000 48.50 0.704526 676 1378 2759 0.3617 0.2763 0.362 1.0000 48.50 0.1695 0.3617 0.2763 0.362 1.0000 48.50 1.548028 784 1596 3195 0.3617 0.2763 0.362 1.0000 48.50 0.1490 0.3617 0.2765 0.3618 1.0000 48.50 0.627930 900 1830 3663 0.3617 0.2763 0.362 1.0000 48.50 0.1589 0.3617 0.2766 0.3617 1.0000 48.50 0.717232 1024 2080 4163 0.3617 0.2769 0.3614 1.0000 48.49 0.1537 0.3617 0.2767 0.3616 1.0000 48.49 0.502334 1156 2346 4695 0.3617 0.2767 0.3616 1.0000 48.49 0.2377 0.3617 0.2768 0.3615 1.0000 48.49 0.661336 1296 2628 5259 0.3617 0.2769 0.3614 1.0000 48.49 0.1867 0.3617 0.2769 0.3614 1.0000 48.49 2.270738 1444 2926 5855 0.3617 0.2765 0.3618 1.0000 48.49 0.1819 0.3617 0.2765 0.3618 1.0000 48.49 1.272740 1600 3240 6483 0.3617 0.2770 0.3614 1.0000 48.49 0.1650 0.3617 0.2766 0.3617 1.0000 48.49 0.864842 1764 3570 7143 0.3617 0.2764 0.3619 1.0000 48.49 0.1649 0.3617 0.2767 0.3616 1.0000 48.49 1.172444 1936 3916 7835 0.3617 0.2767 0.3616 1.0000 48.49 0.1892 0.3617 0.2768 0.3615 1.0000 48.49 0.841146 2116 4278 8559 0.3617 0.2768 0.3615 1.0000 48.49 0.1923 0.3617 0.2768 0.3615 1.0000 48.49 3.180748 2304 4656 9315 0.3617 0.2768 0.3615 1.0000 48.49 0.1462 0.3617 0.2769 0.3614 1.0000 48.49 3.491950 2500 5050 10103 0.3617 0.2767 0.3616 1.0000 48.49 0.1504 0.3617 0.2766 0.3617 1.0000 48.49 2.8000

Figure 3: The congruence of x1_BEN vs. x1_LIN; x2_BEN vs. x2_LIN and x3_BEN vs.

x3_LIN, x3_LI


123

3 Results and Discussion

The calculation results were enumerated to check the calculation efficiencies of both the

applied methodologies and the programming development. To be discussed in this research

paper are the expected values of all response factors and their calculation times on a limited

set of personal computers. Hence, there are three sections discussed as follow:

3.1 The result at the lower section (P = 2:2:50) P = 2:2:50 denotes for the value of P starting at 2: increasing step 2: and ending at 50.

According to assumption a) on page 6, the incremental event step E = P2 for uniform

distribution and constraint number C = m × Event + PD. The values of x1_BEN, x1_LIN;

x2_BEN, x2_LIN; x3_BEN, x3_LIN are congruent and consistent variants as represented in

Table 2 and in Figure 3. The results of Zmin_BEN and Zmin_LIN represent their respective

congruencies and at P= 32 (in Figure 4, at 16 on the axis) . However, consideration of the

series plot of Ti_BEN and Ti_LIN in Figure 5 shows the calculation time fluctuations of

Ti_LIN, but not for Ti_BEN.

Number of point(P=2:2:50) divided in all uncertainties intervals

Calc

ulat

ion

tim

e_ T

i

24222018161412108642

4

3

2

1

0

VariableTi_BENTI_LIN

Time Series Plot of Ti_BEN, TI_LIN

Figure 4: Zmin_BEN and Zmin_LIN Figure 5: Ti_BEN and Ti_LIN

3.2 The results at the middle section (P = 50:2:106) At the intensive computational events, the values of x1_BEN, x3 _BEN, z_BEN and

x1_LIN, x3 _LIN, z_LIN are stable congruent with a very small consistent variance as

reported in Table 3 below:


Table 3: Upper section cut off at the uncertainty number P = 50:2:106. P Event Constr Var x1_BEN x2_BEN x3_BEN Sx_BEN Zmin_BEN Ti_BEN x1_LIN x2_LIN x3_LIN Sx_LIN Zmin_LIN Ti_LIN50 2500 5050 10103 0.3617 0.2767 0.3616 1.0000 48.49 0.2884 0.3617 0.2766 0.3617 1.0000 48.49 3.373752 2704 5460 10923 0.3617 0.2766 0.3617 1.0000 48.49 0.1726 0.3617 0.2767 0.3616 1.0000 48.49 1.256954 2916 5886 11775 0.3617 0.2772 0.3611 1.0000 48.49 0.2258 0.3617 0.2767 0.3616 1.0000 48.49 1.769956 3136 6328 12659 0.3617 0.2771 0.3612 1.0000 48.49 0.1420 0.3617 0.2768 0.3615 1.0000 48.49 1.322958 3364 6786 13575 0.3617 0.2768 0.3615 1.0000 48.48 0.1641 0.3617 0.2768 0.3615 1.0000 48.48 2.875560 3600 7260 14523 0.3617 0.2771 0.3612 1.0000 48.48 0.2000 0.3617 0.2769 0.3614 1.0000 48.48 2.325462 3844 7750 15503 0.3617 0.2768 0.3615 1.0000 48.48 0.2491 0.3617 0.2767 0.3616 1.0000 48.48 5.546664 4096 8256 16515 0.3617 0.2767 0.3616 1.0000 48.48 0.2022 0.3617 0.2767 0.3616 1.0000 48.48 1.580366 4356 8778 17559 0.3617 0.2766 0.3617 1.0000 48.48 0.2676 0.3617 0.2768 0.3615 1.0000 48.48 2.535368 4624 9316 18635 0.3617 0.2768 0.3604 0.9989 48.48 0.1525 0.3617 0.2766 0.3607 0.9990 48.48 3.006070 4900 9870 19743 0.3617 0.2764 0.3567 0.9947 48.47 0.2050 0.3617 0.2762 0.3568 0.9947 48.47 6.688472 5184 10440 20883 0.3742 0.0000 0.5166 0.8908 48.39 0.1865 0.3741 0.0000 0.5166 0.8908 48.39 3.082474 5476 11026 22055 0.3742 0.0000 0.5141 0.8882 48.18 0.1798 0.3742 0.0000 0.5140 0.8882 48.18 2.044076 5776 11628 23259 0.3742 0.0000 0.5116 0.8858 47.98 0.1804 0.3742 0.0000 0.5116 0.8858 47.98 1.480678 6084 12246 24495 0.3742 0.0000 0.5094 0.8835 47.78 0.2126 0.3742 0.0000 0.5094 0.8835 47.78 2.000980 6400 12880 25763 0.3742 0.0000 0.5085 0.8827 47.60 0.1892 0.3742 0.0000 0.5085 0.8827 47.60 1.509482 6724 13530 27063 0.3742 0.0000 0.5062 0.8804 47.42 0.1766 0.3742 0.0000 0.5064 0.8806 47.42 1.455084 7056 14196 28395 0.3742 0.0000 0.5045 0.8787 47.24 0.1828 0.3742 0.0000 0.5045 0.8787 47.24 3.014286 7396 14878 29759 0.3742 0.0000 0.5026 0.8768 47.07 0.1794 0.3742 0.0000 0.5026 0.8768 47.07 6.036188 7744 15576 31155 0.3742 0.0000 0.5009 0.8750 46.91 0.1692 0.3742 0.0000 0.5008 0.8750 46.91 5.982990 8100 16290 32583 0.3742 0.0000 0.4991 0.8733 46.75 0.1697 0.3742 0.0000 0.4991 0.8733 46.75 4.676992 8464 17020 34043 0.3742 0.0000 0.4987 0.8729 46.59 0.1991 0.3742 0.0000 0.4987 0.8729 46.59 1.961594 8836 17766 35535 0.3742 0.0000 0.4971 0.8713 46.45 0.1917 0.3742 0.0000 0.4971 0.8713 46.45 4.871796 9216 18528 37059 0.3742 0.0000 0.4956 0.8698 46.30 0.1981 0.3742 0.0000 0.4956 0.8698 46.30 10.819798 9604 19306 38615 0.3742 0.0000 0.4942 0.8684 46.16 0.2047 0.3742 0.0000 0.4942 0.8684 46.16 4.3436

100 10000 20100 40203 0.3742 0.0000 0.4929 0.8671 46.03 0.2016 0.3742 0.0000 0.4929 0.8670 46.03 3.0834102 10404 20910 41823 0.3742 0.0000 0.4925 0.8668 45.90 0.2233 0.3742 0.0000 0.4925 0.8667 45.90 3.4999104 10816 21736 43475 0.3742 0.0000 0.4913 0.8655 45.77 0.1791 0.3742 0.0000 0.4913 0.8655 45.77 7.3899106 11236 22578 45159 0.3742 0.0000 0.4901 0.8643 45.65 0.1778 0.3742 0.0000 0.4901 0.8643 45.65 3.6450

According to the numerical consideration shown in Table 3, the values of x1_BEN ,

x1_LIN; x2_BEN, x2_LIN ; x3_BEN, x3_LIN are congruent and consistent variants. The

optimal values of Zmin_BEN and Zmin_LIN are also congruent and have tendencies to

converge to the most optimal solution. By contrast with this, the calculation times Ti_LIN

and Ti_BEN are very different. The ratio of them is 2.5353:0.2676 = 9.474:1 at P = 66. The

sums of Sx_BEN and Sx_LIN start to decrease after P = 66 as do the Zmin values also.

Figure 6, 7 and 8 show that the components of x3_BEN and x3_LIN were selected , but not

x2_LIN and x2_BEN at P=72.

In an actual factory, the producers can make a decision at this stage with Sx = 0.8908 unit

weight and Zmin = 48.39. If they want to obtain the sum Sx = 1.000 unit weight to meet

demand, they can select the status of P = 66 with the optimal Zmin = 48.48 (higher cost). A

further perspective, Figure 9 represents the series plot of Ti_BEN and Ti_LIN with major

fluctuations. the calculation times of Ti_LIN are extremely variable, whereas the calculation

time Ti_BEN tends to be constant.


125

Numer of Point [50:2:106] divided in uncertainty intervals

Mar

ket

dem

and

[Un

it W

eigh

t]

272421181512963

1.000

0.975

0.950

0.925

0.900

0.875

0.850

VariableSx_BENSx_LIN

Time Series Plot of Sx_BEN, Sx_LIN

Numer of Point [50:2:106] divided in uncertainty intervals

Min

imum

cos

t Z

[Un

it C

urre

ncy]

272421181512963

48.5

48.0

47.5

47.0

46.5

46.0

45.5

VariableZmin_BENZmin_LIN

Time Series Plot of Zmin_BEN, Zmin_LIN

Figure 6: Sx_BEN and Sx_LIN Figure 7: Min.cost Zmin_BEN and Zmin_LIN

Numer of Point [50:2:106] divided in uncertainty intervalsSele

cted

x1_

LIN,

x1_

BEN,

x3_

LIN,

x3_

BEN

[Uni

t w

eigh

t]

272421181512963

0.525

0.500

0.475

0.450

0.425

0.400

0.375

0.350

Variable

x1_LINx3_LIN

x1_BENx3_BEN

Time Series Plot of x1_BEN, x3_BEN, x1_LIN, x3_LIN

Number of point(P=50:2:106) divided in all uncertainties intervals

Calc

ulat

ion

tim

e_Ti

272421181512963

12

10

8

6

4

2

0

VariableTi_BENTI_LIN

Time Series Plot of Ti_BEN, TI_LIN

Figure 8: The congruence of x1_BEN, Figure 9: Ti_LIN and Ti_BEN.

x3_BEN,x1_LIN, x1_BEN

3.3 The results at the upper section Continuing the computational calculation by the Bender’s decomposition method with

P = (106:2:1584) until the calculation terminated (out of memory on the HP_Pavillion_

IntelCore_2Quard Inside, No.: 016-120610000, personal computer), the nearest optimal

solution can be accepted at P=1584 corresponding to Zmin of 38.63Baht and Sx_BEN of

0.8185 unit weight. Sx_BEN is not equal to 1 as the market demand. It depends upon the cost

factors of flesd and fexdd . If flesd is less than fexdd, it will be reasonable to produce the

mixed product with lower amount from the demand. But, the optimization can reveal the

lowest value of the Zmin as shown in Table 4 below.


Table 4: The upper section cut off at the uncertainty number P = (1560:2:1584) P Events Constr Var x1_BEN x2_BEN x3_BEN Sx_BEN Z_BEN Ti_BEN1560 2433600 4868760 9737523 0.3743 0 0.4445 0.8187 38.64 10.95341562 2439844 4881250 9762503 0.3743 0.0002 0.4442 0.8186 38.64 10.29981564 2446096 4893756 9787515 0.3743 0 0.4442 0.8185 38.64 10.38931566 2452356 4906278 9812559 0.3743 0 0.4442 0.8185 38.64 10.50751568 2458624 4918816 9837635 0.3743 0 0.4453 0.8197 38.63 10.48421570 2464900 4931370 9862743 0.3743 0.0001 0.4444 0.8187 38.63 11.0561572 2471184 4943940 9887883 0.3743 0 0.4442 0.8185 38.63 10.07221574 2477476 4956526 9913055 0.3743 0 0.4445 0.8188 38.63 11.05311576 2483776 4969128 9938259 0.3743 0.0002 0.4445 0.8188 38.63 11.04381578 2490084 4981746 9963495 0.3743 0.0011 0.4446 0.8189 38.63 11.3731580 2496400 4994380 9988763 0.3743 0.0055 0.4441 0.8185 38.63 10.5431582 2502724 5007030 1E+07 0.3743 0 0.4445 0.8187 38.63 11.70561584 2509056 5019696 1E+07 0.3743 0 0.4442 0.8185 38.63 10.5942

OUT OF MEMORY

4 Conclusions

At the first start with a lower division number of points P, the results obtained from the

simplex method-LINPROG and Bender’s decomposition were consistently equivalent. The

calculation results were almost identical. These two algorithms are very suitable for small-

scale problems but when increasing the division numbers (point P) there is, a rise of

uncertainties numbers, the consequence of the enlargement of the constraint numbers. Some

of response factors were found to deviate from the target and thus failed in condition.

Nevertheless, the calculation by both methods can be performed. The LINPROG method is

extensive in calculation time and thus requires a large memory storage. On the personal

computer, the calculation failed to determine the results at the earlier stage.

Conversely, the Bender’s decomposition method can quickly and consistently obtain the

nearest optimal solution up to the calculation termination due to being out of memory. The

same problem and the same calculation tool, MATLAB® program, were also performed on a

high performance computer. The results showed the enormous effect of the system

uncertainties mainly influencing the calculation times. It is noteworthy that the ratio of the

mean time consumption of the LINPROG : BENDER is approximately equal to 232.77 :1 at

P=2:2:500 on a general PC, whereas the results of the other response factors can be

congruent.


127

The authors hope that this calculation method, the integration of Two-stage stochastic linear programming incorporated with the Bender’s decomposition method, compacted in

general form of a MATLAB® programming, can contribute to supporting decision making

in other operations research areas as a low cost effective calculation tool. For future research, this program is to be developed in form of a Graphic User Interface for convenience of use.

5 Acknowledgements

Many thanks go to National Electronics and Computer Technology Center (NECTECH), THAILAND for kindly supporting access into their HPC-calculating system and the TechSourch Co.Ltd. (Thailand) for their research cooperation. Furthermore, the authors thank Mr. Rattaprom Promkham from Mathematics Department, Rachamonkala University of Technology Thanyaburi for programming suggestions.

6 References

[1] A. E. Chappell, (1974): “Linear programming cuts costs in production of animal feeds”, Operation Research quarterly, 25(1): 19-26.

[2] A. G. Munford, (1996): “The use of iterative linear programming in practical applications of animal diet formulation”, Mathematics and computers in Simulation, 42: 255-261.

[3] E. Engelbrecht, (2008): “Optimising animal diets at the Johannesburg zoo”. University of Pretoria, Pretoria.

[4] D. M. Forsyth, (1995): “Chapter 5: Computer programming of beef cattle diet”, in Beef cattle feeding and Nutrition, 2nd ed., T. W. Perry and M. J. Cecava, Academic Press, Inc, pp. 68.

[5] I. Katzman, (1956): Solving Feed Problems through Linear Programming, Journal of Farm Economics, 38(2) (May, 1956): 420-429.

[6] G.Infanger, George B. Danzig, (1993): Planning under uncertainty-Solving Large-Scale Stochastic Linear Programs, Stanford University.

[7] G. Ausiello, P. Crescenzi, G.Gambosi, V.Kann, A. M. Spacecamda, (1998): Complexity and Approximation, Springer, Chapter 2.

[8] Michel X. Goemans, David, P. William son, (1997): The primal-dual method for approximations algorithms and its application to network design problems, PWS Publishing Co., Chapter 4.

[9] M. O. Afolayon and M. Afolayon, (2008) “Nigeria oriented poultry feed formulation software requirements”, Journal of Applied Sciences Research, 4(11): 1596-1602.


[10] P. Charnsethikul, (2009): Theory of Primal/Dual and Benders’ Decomposition, Lecture notes Department of Industrial Engineering, Kasetsart University, Thailand.

[11] P. Charnsethikul, (1996): Linear Programming with Constraint Coefficient Tolerances, IE Network Conference, Industrial Engineering in AD. 2000, 24-25 October 1996, Pattaya, Thailand: 287-298.

[12] Robert M. Freund, (2004): Benders’ Decomposition Methods for Structured Optimization, including Stochastic Optimization, Massachusetts Institute of Technology.

[13] Rosshairy Abd Rahman, Chooi-Leng Ang, and Razamin Ramli, (2010): Investigating Feed Mix Problem Approaches: An Overview and Potential Solution, World Academy of Science, Engineering and Technology.

[14] S. Babu and P. Sanyal, (2009): Food Security, Poverty and Nutrition Policy Analysis: Statistical Methods and Applications. Washington, DC, USA: Academic Press. p.304.

[15] S. Thammaniwit, (2013): A Stochastic Linear Programming Method for the Diet Problem under Uncertainties, Doctoral Thesis, Submitted to the Senate of Kasetsat University, Bangkhen, Bangkok, Thailand, and December 2013.

[16] Wagner, H. M., 1977. Principles of operations research, with applications to managerial decisions, 2nd Ed., Introduction to Stochastic Programming Models, pp.651-699.

S.Thammaniwit is an Assistant Professor of Industrial Engineering Department at Thammasat University, Thailand. He received his Dipl.Ing. (Konstruktionstechnik: Werkzeugsmaschinen) from The University of Applied Science of Cologne, Germany in 1982. He worked and received on-the-job training with at least 14 German companies before working in the government sector in his country. He earned his Master’s degree in Manufacturing System Engineering under Chula/Warwick corporation program at Chulalongkorn University, Bangkok in 1992. Most of his research is involved with tools, machine tools design and construction as well as Engineering Management. Currently, he is pursuing a doctoral degree at Kasetsart University, Bangkhen, Bangkok, Thailand.

Dr.P. Charnsethikul is an Associate Professor of Industrial Engineering Department, Kasetsart University, Bangkok, Thailand. He received his M.S, PhD. (Industrial Engineering) from Texas Technical University, USA. His research interests are in the area of Optimization, Operations Research, Numerical Mathematics & Statistics, Management Science, Production & Operations, Numerical Methods and Analysis with Applications in Safety Engineering. Since 2006 he has been acting as Deputy Dean of the Faculty of Engineering at Kasetsart University, Bangkhen, Bangkok, Thailand.


*Corresponding author (David Kuria). Tel: +254-727-399208. E-mail addresses: [email protected]. 2013 International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/129-143.pdf

129


http://TuEngr.com

Securing Bank Loans and Mortgages Using Real Estate Information Aided by Geospatial Technologies David Kuriaa*, Moses Gacharia, Patroba Oderab and Rogers Mvuriac

a Department of Geomatic Engineering and Geospatial Information Science, Dedan Kimathi University of Technology, KENYA b Department of Geomatic Engineering and Geospatial Information Systems, Jomo Kenyatta University of Agriculture and Technology, KENYA c Lutheran World Federation, Department for World Service, KENYA A R T I C L E I N F O

A B S T R A C T

Article history: Received 10 November 2012 Received in revised form 25 January 2013 Accepted 28 January 2013 Available online 29 January 2013 Keywords: Banking; GIS; Loan appraisal; Information technology; Mortgage.

Due to liberalization within the financial market, there has been increased cash flow in banks. This has resulted in increased competition among banks to secure and increase their customer base, in an effort to remain profitable. Banks are foregoing the multitude of checks that used to be conducted before granting any mortgage facility to customers, in an effort to remain competitive. This has led to a drastic increase in the number of credit card and loan defaulters, leading to increased operation costs and reduction in profit margins. This research proposes an integrated GIS approach enabling banks locate defaulting real estate properties used as collateral. Using data provided by Kenya Commercial Bank (KCB) for a locality in Kenya, a geodatabase was developed and a custom application developed for the bank loan appraiser to use. This application retrieves property information about a client based on his/her bank account information. Based on a series of spatially driven queries embedded in the solution, the appraiser can prepare a detailed appraisal for the client in a very short time, thereby satisfying the client, while not prejudicing the banks position.



130 David Kuria, Moses Gachari, Patroba Odera and Rogers Mvuria

1 Introduction

Geographic Information Systems (GIS) have been defined and conceptualized in a number

of different but related ways. de Man (1988), Goodchild (1992), and Burrough (1986) argue

that GIS is a special type of information system that handles spatial data. Dickinson and Calkins

(1988) take a component view of GIS, arguing that a GIS has three elements: technology

(hardware and software), a database, and infrastructure (staff, facilities, etc.). GIS technology

has a great deal to offer the mortgage finance industry because geographic location and spatial

relationships have a central role in housing and mortgage market outcomes. Housing is fixed in

its location and is durable. As a result, a home’s location relative to employment opportunities,

mortgage finance and housing market intermediaries, public services, and amenities exerts a

strong influence on its price. The location of a home influences the choices and opportunities of

its residents and of those seeking to own it and location is thus a strong influence on mortgage

markets. The location of mortgage suppliers defines the availability of mortgage credit. The

segregation of residential space into discrete geographic submarkets influences the pattern and

nature of mortgage product demand.

The highly segregated nature of residential space demands that mortgage lenders target

their products, services, and marketing efforts to the specific character of different demographic

groups that occupy different market areas Because the nature of the competition faced by

participants in the mortgage finance process varies across different areas, they must develop

different competitive strategies to suit the character of their competition in different areas. GIS

technology has the potential to support a wide range of business applications in mortgage

finance (Alberts and Douglas, 1992). At the most elemental level, it can provide mapping

capabilities to help decision makers visualize the spatial distribution of variables that affect

their business. At a higher level, it can be used to combine multiple variables such as the racial

and income composition of neighborhoods with the location of recent mortgage originations. It

can be used, for example, to select the optimal number, location, and size of branch offices to

service a market area given some decision rules about how far any share of potential borrowers

can be from a branch office (Birkin and Clarke 1998).

In Kenya, the use of GIS is rather limited but growing steadily with the mainstreaming of

GIS in many curricula in the Universities. Acquisition of georeferenced data is also an

expansive undertaking including the data management and dissemination. Another limitation is


131

poor consumer awareness which means less demand for the products and services of GIS.

Regardless of these limitations, GIS has been used in Kenya for several projects with good

result, for instance, in compiling the National Water master plan (Republic of Kenya, 1992).

The Kenya Wildlife Services (Kariuki, 1992) used GIS for managing the large volumes of data

they acquire relating to wildlife census, vegetation and land use dynamics, infrastructure,

security and planning of operations. The department of resource surveys and remote sensing

makes use of GIS and Remote Sensing in the mapping of natural resources (Ottichilo, 1986).

An integrated GIS and remote sensing system has been developed for water resources

management in Kitui county (Kuria, et al, 2012), while Mulaku and Nyadimo (2010) used GIS

in the mapping of schools. Kuria, et al (2011) developed a GIS tool for enhancing efficiency in

distribution of national examinations. GIS has also been used to prepare the National

Environment Action Plan (Ministry of Environment and Natural Resources, 1994) and to

monitor a development programmed in Laikpia District (Hoesli, 1995).

In recent years, the banking industry in Kenya has been undergoing drastic changes

reflecting a number of underlying developments. Significant advancement in surveying,

architectural and information technology (IT) has accelerated and broadened the dissemination

of real property information and financial services and also increased the complexity. Another

key impetus for change has been the increasing competition among a broad range of domestic

and foreign institutions in providing bank loans, mortgages and other related services.

Regulations and computer technology advancement are forcing mortgage institutions to adopt

better operational strategies and upgrade their skills, throwing more challenges to banking

sector.

One of the most tedious tasks in banking is providing mortgage services for their clients.

Banks offer this service to enhance their accessibility to the customers. To cushion themselves,

banks need collateral such as tangible (fixed) property or business assets etc. In order for a bank

to sanction a mortgage, the property is analyzed. Some of these analyses include distance from

the main road or size of the plot, verification of the actual owner of the plot, current land

valuation etc. Currently in Kenya, banks accomplish these analyses manually by going to the

sites or contacting third party to do the tasks. This process at the least takes 30 days to complete

and involves a substantial expenditure in funds while following this due diligence. In case the

same plot is being used by a client to request for other loans to the same bank or different banks


the analyses have to be done for each application cascading the problem.

A GIS simplifies the process by easily accomplishing the following: (i) does the spatial analysis

easily on the computer from the office, (ii) calculates the area or distance from main road or

other features of interest without having to visit the site/plot, (iii) calculates the land value of the

area by analyzing the surrounding area, (iv) creating a centralized or distributed database

system that can be used to detect and prevent multiple loan applications utilizing the same plots,

and (v) perform auction procedures and instructions without much difficulty.

The objective of this study is to develop a pilot database for real property information

within parts of Kilimani, Lavington and Kileleshwa estates that can be used to secure bank

loans and mortgages. To accomplish this objective a geodatabase is developed with capabilities

of analyses, and modeling of (i) present land use (ii) present land value (iii) information of the

plot e.g. past owner, case, previous loan taking condition etc (iv) distance from the main road

and (v) adjacent road width

2 The study area

The area chosen for study (Figure 1) covers extensive parts of Kileleshwa, Kilimani and

Lavington estates. It lies between (9857500m N, 242000m E) and (9859000m N, 254000m E).

It covers an area approximately 3,000,000 m2. The area has approximately 1500 plots and over

1800 permanent buildings. Semi-permanent buildings were however not considered.

Lavington is a suburb green haven lying halfway between the busier areas of Hurlingham

and Westlands. Here the few modern apartment blocks are unobtrusive and the lanes linking busier roads are lined with large houses and bungalows set in an acre or less of well-tended land. Many have swimming pools. Some roads are gated with security staff screening all visitors. Lavington is popular with expatriate families. Kilimani is bounded to the north by Dennis Pritt Road and to the south by Ngong Road, bisected by Argwings Kodhek Road. T hese busy areas host many of Nairobi’s newly built apartment blocks. These are replacing the expansive 1950s and 1960s bungalows which once sat back from the wide streets bordered by high green hedges. Few of the new blocks are taller than five story and many enjoy balconies, communal swimming pools and 24-hour security staff. Kileleshwa is quieter and greener than Kilimani, Hurlingham or Ngong Road and more of the 1950s and 1960s bungalows, set in large mature gardens, have survived the property developers. However, there are some apartments


133

blocks here too.

Figure 1: The study area.

2.1 Mortgage financing The term mortgage refers to the process by which an individual or a business can purchase

either a residential or a commercial property without having to pay the total value upfront. Mortgage is defined as a loan to an individual or a business for purchasing a real estate. In this case the real estate also acts as collateral for the loan. The mortgage contains two parts: (i) mortgage that is also the pledge and (ii) the promissory note which is the promise for repayment.

Virtually all mortgage financing arrangements require one to put in some equity, with the

financing institution funding a portion of the value of one’s new home. Most will require that one saves up for the required down payment with them, which is usually 15% to 30% of the purchase price. Paying more up-front works better in the long run, reducing the costs of the mortgage. One may also pay term deposits and various statutory fees for any transfers, charges and/or registrations to take place. As a rule of thumb, the combined legal and stamp duties work out to about 10% of the purchase price of the new home. Monthly repayments will also typically include the cost of insurance policies on the property offered as security for the loan and on mortgagor’s life.


3 Methodology

The study area was chosen on the basis that it has undergone drastic development changes

for the past few years due to changing zoning ordinances and building regulations. Emergence

of apartments clearly indicates that many people are buying residential houses in this area. The

prices of such apartments range from Ksh. 6,000,000 and some even over Ksh. 10,000,000. It is

thus obvious that mortgage market will continue to be of great concern in this region. The

banking sector thus needs to collect and maintain a spatial database of all spatially referenced

information about the buildings that are on mortgage and all land parcels that acting as

collaterals.

3.1 Data collection In order to have a clear view of the study area some aerial photographs were used. These

had the advantage of depicting the land uses as they appear and brought about clarity.

Figure 2: Digitized property map.

A topo-cadastral map (map showing land subdivisions and the topography) of the area was

scanned at a resolution of 600 dpi (dots per inch) and saved. The target areas of Kilimani,

Lavington and Kileleshwa estates acted as the sample size of the project. The features

considered are (i) transport network, (ii) cadastral subdivision, (iii) existing buildings, and (iv)


135

land use. Figure 2 shows the land parcels in the study area, which can serve as collateral for

mortgages.

Data capture was carried out in the following ways: (i) digitizing by scanning hardcopy

maps(ii) onscreen digitizing and (iii) attribute entries. The topographic maps, survey plans

and building plans were scanned using a flatbed scanner. The resulting images are a digital

copy of the original paper based map or plan. Onscreen digitizing was done with great care

since a deviation in parcel areas would mean a lot in the final analysis. This involved

digitization of line, point or area features. Once all the required features had been digitized

and stored in their corresponding layers, attribute entry was done. Buildings’ photographs and

where possible floor layout plans were incorporated into the database in order to provide as

much information as possible. In the following diagram, one can that the selected building is cut

through by the boundary on the right. From visualization alone several conclusions can be

drawn; number of floors, boundary case, building type, value, floor area, nearest road etc.

3.2 The conceptual system A customer-centric business model can help address these challenges. A mortgage

institution’s primary function is to deliver financial services and products to their customers. In

the modern world they need to be market driven and market responsive. The success of such

an institution depends on its approach to data management, customer relation management.

Such institutions manage a bulk of information about customers, customer profiles and much

more. By incorporating ‘geographical location information of real property’ into their

database, long range planning mixed with geographical modeling will yield tangible benefits to

the mortgage finance institutions community.

Figure 3 shows the mortgage decision making process as conceptualized in this work. A

GIS plays a central role tying the bank’s non-spatial data to spatial data. Utilizing the bank’s

non-spatial data, the financial health of a client’s account can be easily obtained. Using the

spatial data, property information can be retrieved and analyzed based on various attributes and

spatial relationships with other features. From these analyses, the mortgage or loan value can be

determined. Information about encumbrances and other restrictions on a property can also be

retrieved. Since all these bits of information have been stored in a single centralized or

distributed database, a quick decision (in a matter of minutes) can be arrived at on whether to


grant or deny the application. To aid in this decision making process, the logic captured in

Figure 3 has been coded into an extension that calls up the geodabase and based on the

parameters of interest (such as particular parcel, client, mortgage remaining, transaction

histories) a set of decision pathways are proposed and presented to the appraiser using the

system.

Figure 3: Conceptual model of the mortgage decision making process.

4 Results and Discussion

A pilot GIS database for real property information in Kileleshwa, Kilimani and Lavington

has been developed. The database contains information on plots, plot numbers, plot sizes, plot

values. Layers of related geographic features e.g. buildings, access roads and rivers are also

incorporated in this database. Figure 4 shows an example of information retrieved for one land

parcel and the building it contains. Managing such a bulk of geographical data including


137

customer’s bank records is hectic and cumbersome. Prior to the incorporation of the GIS,

decision making required some time and even a third party consultation. Now with the GIS in

place, some of the checks and decisions can be made quite easily and in a cost effective fashion.

Figure 4: Retrieved geodatabase results for parcel and contained building.

Figure 5: Accessibility levels.

In Figure 4, parcels with boundary disputes can be easily identified from the database,

encroaching parcels, existing encumbrances, e.g. court orders, caveat, pending cases and other

restrictions can also be effortlessly retrieved. It is evident that the distance from the main road


affects the value of any property; the further the property is from a main road, the lesser the

value.

Figure 5 shows the accessibility levels. These levels were determined through a 15m

buffering on either side of the road. Parcels in light green color had a high proximity to tarmac

road, other parcels in light blue had a lower proximity to the road. They however had minor

access roads linking to the tarmac roads. For land value computations purposes, the study area

was assumed to have a high accessibility level in the interest of simplicity. Eighteen properties

were sold out recently in the area; these were used to make buffers at distances of 5 km.

Figure 6: Property value distribution.

Buffers are rings drawn around features at specified distance from the features. These buffers

were meant to determine the value of all land parcels in the study area. It was concluded that the

land values were homogeneous throughout the study area. The statistics of these plot values in

shillings and areas in square meters were generated. From frequency distribution analysis of

these recently sold out properties, a mean value of approximately KShs. 14,790,000.00 per acre

was obtained. A similarly obtained mean area of the properties was found as 4990 m2. The

following formulation (eq. 1) was used to obtain the prevailing value for a property.


139

AcACC ××= 4047 (1)

Where Ac = property area (in acres) A = property area (m2) and C = value of property with

the over-bar referring to mean value, with the value 4047 being the conversion factor (from

acres to m2). It is worth mentioning that in the study area, errors in area computation can have

serious value implications since 1 m2 costs KShs. 3,000.00. Based on these computations, a

property value distribution map was prepared (Figure 6). It gives a clear view of prevailing

land values. It is therefore easy to predict the range of any plot value at a glance. The most

expensive parcels belong to Lavington Primary school and Kenton College – dark brown. The

value depends on the size of the property, the larger the size the more expensive.

Figure 7 provides insight to the bank on the distribution of property that the bank needs to

subject to auction – in red. The plots in purple show mortgaged property. Tracking such

property with the use of Global Position System (GPS) is easy and cost effective since their

positional information can be obtained from the map and imported into the GPS devices for

field crews.

Figure 7: Mortgages approved and defaulted.


An ArcGIS extension was developed that can be used to bring out a user interface to

non-GIS appraisers. This was programmed in Visual basic and linked to the GIS database. The

extension generated the interface shown in Figure 8. An account number input generated all

other information related to that parcel. The database end user (bank employee) is only

expected to input the account number for any customer and the program automatically retrieves

all other parcel related information. All this is achieved within a GIS environment with the

map of attributes in the background.

Figure 8: User interface for querying database.

Based on the amount of mortgage applied for and the credit history of a client, the

application is able to flag the credit worthiness of the applicant. A healthy client is flagged with

a green status, while one who either has encumbrances on the property or whose value of

collateral is insufficient to guarantee the mortgage is flagged with a red status. In the case of one

whose case needs further clarification an orange status is presented.


141

5 Conclusion A GIS database for real property information within the study area has been developed.

Attributes related to real estate were input from existing maps, with some being generated by

the software such as areas and prices of land parcels. The geodatabase retrieves geospatial

information from the mortgage appraiser. Performance of analysis, statistics and updating

yielded the desired results. Storage and retrieval of spatial data was convenient, without a large

storage capacity demands. Using the proposed system, a bank can effortlessly and efficiently

manage the administration its financial products touching on spatial elements. This system has

demonstrated capability of identification and tracking of defaulters, multiple loan or mortgage

applications tied to the same property. It is able to process the credit worthiness of any client

applying for a mortgage.

While this study has demonstrated the potential of GIS in the banking industry, banks still

need to decide on the utility of GIS with respect to simplifying their loan and mortgage

processing. The rapid emergence of apartments in high residential areas shows that mortgage

market will continue to thrive and hence handle more geographic data. Such data needs a

central storage with fast digital map retrieval capability; which is accompanied by any other

related non-spatial data. This work considered data from one bank which was not linked to any

other bank’s data and it is imperative that a centralized system for the banking sector can help

deal with rogue defaulters from other banks. Sharing of these or any spatial data among bank

branches can improve efficiency and thus reduce costs.

6 Acknowledgement

The authors would like to acknowledge the Kenya Commercial Bank for providing

valuable information on the loan procurement procedures and data used in this work.

7 References

Alberts, R. J., and Douglas S. B. (1992). “Geographic Information Systems: Applications for the Study of Real Estate”, Appraisal Journal, 60(4), pp. 483–492.


Birkin, M. and Clarke, G. (1998). “GIS, geodemographics and spatial modelling: an example within the UK financial services industry”, Journal of Housing Research, 9(1), pp87-112.

Burrough, P., (1986). Principles of Geographic Information Systems for Land Resources Assessment. Monograph on Soil and Resources Survey, (12), Clarendon Press, Oxford. 194 pp.

Dickinson, H. J. and H. W. Calkins (1988). "The economic evaluation of implementing a GIS", International Journal of Geographical Information Systems, 2(4), pp. 307-327.

Goodchild, M., (1992). “Geographical Information Science”, International Journal of Geographic Information Systems, 6(1), pp. 31-46.

Hoesli, T. (1995). “GIS based impact monitoring of a development programme” Laikipia-Mount Kenya Papers. No.18.

Maguire, D. J., (1991). “An overview and definition of GIS”, In Geographical Information Systems Principles and Applications, edited by D. J. Maguire, M. F. Goodchild, and D. W. Rhind. (New York: Longman Scientific and Technical; John Wiley and Sons, Inc.), pp. 9-20.

de Man, E., (1988). “Establishing a geographic Information System in Relation to its Use”, International Journal of Geographic Information Systems, 2(2), pp. 245-261.

Ministry of Environment and Natural Resources (1994). The Kenya National Environment Action Plan. Summary. Nairobi, Kenya.

Mulaku, G.C. and Nyadimo, E. (2011). “GIS in Education Planning: The Kenyan School Mapping Project” Survey Review, 43(323): 567-578.

Kariuki, A. (1992). “Applications of geographic information systems in the management of the wildlife resource.” In Applications of geographical information systems for efficient data storage and handling in Kenya. Okoth, P.F. (ed.). Proceedings of a symposium. Kenya Soil Survey, Nairobi. Pp.10-19.

Kuria, D. N., Gachari, M. K., Macharia, M. W. and Mungai, E., (2012). “Mapping groundwater potential in Kitui District using geospatial technologies”. International Journal of Water Resources and Environmental Engineering, 4(1), pp. 15 – 22.

Kuria, D. N., Ngigi, M. M., Wanjiku, J. W. and Kasumuni, R. K., (2011). “Managing distribution of national examinations using geospatial technologies: A case study of Pumwani and Central divisions” International Journal of Computer Engineering Research. 2(5), pp. 82 – 92.


143

Ottichilo, W.K. (1986). “Food production, famine and early warning system: The Kenyan experience.” In: Proceedings of the 20th International Symposium on Remote Sensing of the Environment. Nairobi, Kenya. Environmental Research Institute of Michigan, Ann Arbor, Michigan, U.S.A. pp. 177-180.

Republic of Kenya. (1992). “The study on the National Water Master Plan” Sectoral Report (S). GIS based analysis. Ministry of Water Development. Nairobi.

Dr. David Kuria is a Senior Lecturer in the department of Geomatic Engineering and Geospatial Information Science of the Dedan Kimathi University of Technology. He holds a B. Sc (Surveying) with Honors from the University of Nairobi (Kenya), an M. Sc (Photogrammetry and Geoinformatics) from the Stuttgart University of Applied Sciences (Germany) and a PhD from the University of Tokyo (Japan). Dr. Kuria’s current interests are in web mapping, climate research and geospatial application development.

Dr. Moses Gachari is an Associate Professor in the Department of Geomatic Engineering and Geospatial Information Science of the Dedan Kimathi University of Technology. He holds a B.Sc (Surveying and Photogrametry) with Honors from the University of Nairobi (Kenya), an M.Sc., and a PhD degrees from the University of Oxford (UK). Prof. Gachari has research interests in geospatial applications in development and environment, geodesy and surveying in general.

Dr. Patroba Odera is a lecturer in the Department of Geomatic Engineering and Geospatial Information System of Jomo Kenyatta University. He holds a B. Sc. in Surveying with Honors and an M. Sc. in Surveying from the University of Nairobi (Kenya) and a PhD from the Kyoto University (Japan). Dr. Odera’s research interests are in gravity determination, geodesy and geospatial technologies.

Rogers Mvuria is a Geomatics Engineer and GIS analyst with the Lutheran World Federation. He holds a B.Sc. in Geomatic Engineering and Geospatial Information Systems with Honors from the Jomo Kenyatta University of Agriculture and Technology. Mr. Mvuria’s research interests are in development of geospatial applications and surveying.



http://TuEngr.com

Call-for-Papers:

ITJEMAST invites you to submit high quality papers for full peer-review and possible publication in areas pertaining to our scope including engineering, science, management and technology, especially interdisciplinary/cross-disciplinary/multidisciplinary subjects.

Next article continue on next page



*Corresponding author (B. Wanno). Tel/Fax: +66-43-754246. E-mail address: [email protected] 2013 International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/145-156.pdf

145


http://TuEngr.com

Theoretical Investigation of Hetero–Diels–Alder Functionalizations on SWCNT and Their Reaction Properties

Danai Pankhao a, Nongnit Morakot a, Somchai Keawwangchai a, and Banchob Wanno a,b*

a Supramolecular Chemistry Research Unit, Department of Chemistry, Faculty of Science,

Mahasarakham University, THAILAND b The Center of Excellence for Innovation in Chemistry (PERCH-CIC), THAILAND A R T I C L E I N F O

A B S T RA C T

Article history: Received December 2012 Received in revised form 28 January 2013 Accepted 04 February 2013 Available online 07 February 2013 Keywords: DFT; Hetero–Diels–Alder reaction; Nitrosoalkene; ONIOM; SWCNT; Thionitrosoalkene.

Two–layered ONIOM method at the ONIOM(B3LYP/6–31G(d,p):AM1) theoretical level was applied to investigate the hetero Diels-Alder reaction functionalization of various nitrosoalkenes (NAs) and thionitrosoalkenes (TNAs) onto side–wall (5,5) armchair SWCNT. The results indicated that SWCNT can be functionalized with NAs and TNAs. The energy barriers of TNAs funtionalized SWCNT were lower than those of NAs. This implied that TNAs are easier to react with SWCNT than those of NAs. In addition, electronic properties and density of states of SWCNT were modified by the Diels-Alder functionalizations of NAs and TNAs.


1 Introduction During the last decade, many researches have been focused on functionalizations of

single–walled carbon nanotube (SWCNT) to make fascinating new physical and chemical

properties for practical applications (Meyyappan, 2005). Generally, chemical

functionalizations to SWCNT were achieved by covalent functionlizations onto the sidewall of

tube at the sp2 carbon system. The experimental functionlizations using fluorine (Chamssedine


146 D. Pankhao, N. Morakot, S. Keawwangchai, and B. Wanno

et al., 2011), diazonium salt (Bahr et al., 2001), and fuming nitric acid (Kitamura et al., 2011),

and theoretical functionalizations using azomethine ylides (Cho et al., 2008), nitrene (Zhang et

al., 2006), ozone (Yim & Johnson, 2009), ethene (Lawson & Walker, 2012), alanine and

alanine radical (Rajarajeswari et al., 2012), and diazomethyl aromatic compound (Raksaparm

et al., 2012), pyrazinamide (Saikia & Deka, 2010) on SWCNTs were successful studied and

reported. Chemical cycloaddition on buckminsterfullerene (C60) was reviewed by Yurovskaya

and Trushhov (Yurovskaya, & Trushkov, 2002).

Diels–Alder reaction is well known to occur between a conjugated diene and a dieneophile

and is particularly useful chemical modification to construct cyclic compounds. This reaction

has also been explored on the functionalizations of the C60 (Ohno et al., 1993; Ohno et al.,

1995; Yang et al., 2006; Nakahodo et al., 2008; Yang et al., 2009). Interestingly, the

Diels–Alder cycloadditions on the sidewall SWNT were successful studied by experimental

(Delgado et al., 2004; Ménard-Moyon et al., 2006) and theoretical (Lu et al., 2002) methods.

Nitrosoalkenes (Tahdi et al., 2002; Gallos et al., 2003) and thionitrosoalkenes (Bryce et al.,

1994; Reed & Zhang, 2001) are a class of hetero dienes. In principle, the SWCNTs should be

traceable to these reactions with hetero compounds such as nitrosoalkenes (NAs) and

thionitrosoalkenes (TNAs). However, experimental and theoretical studies of the side–wall

addition of nitrosoalkene and thionitrosoalkene to SWCNT have not yet appeared to the best of

our knowledge. In the present work, the hetero-Diels–Alder reactions of nitrosoalkene and

thionitrosoalkene compounds on armchair (5,5) SWCNT have been investigated by using the

quantum calculation.

2 Computational Details

Two-layered ONIOM method at the ONIOM(B3LYP/6–31G(d,p):AM1) theoretical level

was applied to geometry optimizations of all species of cycloaddition functionalization onto

side-wall (5,5) armchair SWCNT. The model of SWCNT (C70H20 model) was chosen with

open ends and the hydrogen atoms were used to saturate the carbon atoms at the two terminated

ends of the tube (Figure 1). The ball atoms of a pyrene (C16) model cluster and those belonging

to nitrosoalkene and thionitrosoalkene molecules were treated at the higher B3LYP/6–31G(d,p)

level, and the remaining SWCNT atoms were treated with the AM1 method. Based on the

two-layered ONIOM approach, a pyrene molecule shown as ball atoms of SWCNT was

selected to be the high level layer. The hetero-Diels–Alder functionalizations were assigned to


147

occur at the C1–C2 bond of SWCNT as shown in Figure 1. All of the structures of reactants,

transition states and products were located by the ONIOM(B3LYP/6–31G(d,p):AM1) model

achieved without any symmetry constraints. All transition states were characterized by single

imaginary frequency.

The vibration frequency computations were performed at 298.15 K and the standard

pressure as applied in our previous works (Wanno & Ruangpornvisuti, 2006). The highest

occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO)

energy gaps and density of states (DOSs) were also been determined. All calculations were

performed with the GAUSSIAN 03 program (Frisch et al., 2008). The molecular graphics of all

related species were generated with the MOLEKEL 4.3 program (Flükiger et al., 2000).

Figure 1: The optimized structures of single walled carbon nanotube (SWCNT),

nitrosoalkene (NA) and thionitrosoalkene (TNA) reactants.

3 Results and discussion

The structural optimizations of (5,5) armchair SWCNT, nitrosoalkene and

thionitrosoalkene, their product and transition state structures were carried out at the

ONIOM(B3LYP/6–31G(d,p):AM1) level of theory. The optimized structures of SWCNT,

nitrosoalkenes, and thionitrosoalkenes are displayed in Figure 1. The selected bond distances

and bond angles of the optimized structures are considered and discussed. The average C1–C2

bond distance of SWCNT was 1.380 Å which is in good agreement with the previous reports

(Raksaparm et al., 2012). For the nitrosoalkene reactants, the C3–C4 bond distances were


1.335, 1.343, and 1.342 Å, respectively whereas the N–O bond distances are 1.220, 1.215, and

1.213 Å. The C3–C4–N bond angles were 123.7, 119.0, and 119.3°, respectively whereas the

C4–N–O bond angles were 114.5, 115.5, and 115.3° for NA, PhNA, and NO2PhNA,

respectively. The C3–C4 bond distances were 1.346, 1.353, and 1.352 Å, respectively whereas

the N–S bond distances were 1.602, 1.596, and 1.594 Å. Moreover, the C3–C4–N bond angles

were 128.2, 122.8, and 123.2°, respectively whereas the C4–N–S bond angle were 121.8, 123.8,

and 123.6° for TNA, PhTNA, and NO2PhTNA, respectively.

Figure 2: The ONIOM(B3LYP/6–31G(d,p):AM1)–optimized transition state structures for

nitrosoalkenes (above) and thionitrosoalkenes (bottom). Imaginary frequencies (in cm–1) are also presented.

Considering the transition state as show in Figure 2, the reaction started from nitrosoalkene

or thionitrosoalkene molecules reacted to C=C bond of SWCNT via the transition state to form

the functionalized SWCNT products. For the transition state structures of nitrosoalkenes, the


149

Figure 3: The ONIOM(B3LYP/6–31G(d,p):AM1) – optimized product structures for

nitrosoalkene (above) and thionitrosoalkene (bottom) functionalizations.

C1–O and C2–C3 bonds were formed at the sidewall of SWCNT, the C3–C4 and N–O bond

distances were then elongated when comparing with its corresponding reactant structures. The


C1–O bond distances were 2.268, 1.939, and 1.969 Å for NA, PhNA, and NO2PhNA,

respectively, whereas the C2–C3 bond distances were 2.059, 2.176 and 2.168 Å for NA, PhNA,

and NO2PhNA, respectively. For the transition state structures of thionitrosoalkenes, when the

C1–S and C2–C3 bonds were formed at the sidewall of SWCNT which the C3–C4 and N–S

bond distances were also elongated. The C1–S bond distances were found to be 2.649, 2.498,

and 2.675 Å for TNA, PhTNA, and NO2PhTNA respectively, whereas the C2–C3 bond

distances were 2.239, 2.088, and 2.221 Å for TNA, PhTNA, and NO2PhTNA, respectively.

Geometrical structures of products are displayed in Figure 3 in which the products were

represented the formation of the newly six-member ring of functionalized SWCNTs. The C1–O

bond distances were 1.485, 1.486, and 1.493 Å for NA, PhNA, and NO2PhNA, respectively,

while the C2–C3 bond distances were 1.583, 1.583, and 1.583 Å for NA, PhNA, and

NO2PhNA, respectively, and C1–S bond distances were 1.928, 1.927, and 1.930 Å for TNA,

PhTNA, and NO2PhTNA, respectively, while the C2–C3 bond distances were 1.583, 1.582, and

1.581 Å for TNA, PhTNA, and NO2PhTNA, respectively. After the functionalization

completed each of C1 and C2 atoms formed 4 chemical bonds with neighboring atoms. This

indicated that hybridizations of C1 and C2 atoms were completely changed from sp2 to sp3.

3.1 Reaction Energies and Energy Profiles Energy profiles based on the ONIOM(B3LYP/6–31G(d,p):AM1) computation for the

hetero-Diels-Alder functionalizations of nitrosoalkenes and thionitrosoalkenes onto SWCNT

are displayed in Figure 4 and the reaction energies, reaction energy barriers, and imaginary

frequencies of the functionalizations are listed in Table 1. The reaction profiles with initial

reactants (R), transition state and reaction products (P) are also represented in Figure 4. The

relative energy profiles showed that the energy barriers for nitrosoalkene functionalizations

were 21.88, 24.76, and 27.64 kcal/mol for the NA, PhNA, and NO2PhNA, respectively. It

should be noted here that the NA addition showed the lowest in the activation barrier. In

addition, the energy barriers of the thionitrosoalkene functionalizations were 15.01, 13.16, and

12.32 kcal/mol for the TNA, PhTNA, and NO2PhTNA, respectively, in which the NO2PhTNA

addition showed the lowest in the activation barrier. Clearly, all of functionalizations were

occurred via exothermic process. In both system, these energy barriers are strongly dependent

on the nature of the heteroatoms and the molecular geometries presented on the reaction.


151

Figure 4: The reaction profiles and relative energy profiles (in kcal/mol) of (a) nitrosoalkene and (b) thionitrosoalkene functionalizations. Where R is reactants, TS is transition state and

P is reaction products.

Table 1: Reaction energies (ΔE), reaction barriers (ΔE ≠) and the imaginary frequencies (vi) for the transition state of hetero Diels-Alder functionalizations computed at the

ONIOM(B3LYP/6–31G(d,p):AM1) level of theory. Reactions ΔE a ΔE ≠,a vi

b

Nitrosoalkene addition SWCNT+NA NA/SWCNT -0.11 21.88 561.4iSWCNT+PhNA PhNA/SWCNT -2.05 24.76 519.8iSWCNT+NO2PhNA NO2PhNA/SWCNT -2.60 27.64 536.8iThionitrosoalkene addition SWCNT+TNA TNA/SWCNT -8.55 15.01 498.5iSWCNT+PhTNA PhTNA/SWCNT -12.34 13.59 450.5iSWCNT+NO2PhTNA NO2PhTNA/SWCNT -12.95 12.32 507.6ia In kcal/mol. b Imaginary frequencies (cm–1).

Table 2: The ELUMO and EHOMO energies and Egap of tube and its adduct complexes computed

at the ONIOM(B3LYP/6–31G(d,p):AM1) level of theory Species ELUMO

a EHOMOa Egap

a, b

SWCNT –2.28 –4.45 2.17 [2.20] c NA/SWCNT –2.31 –4.47 2.16 PhNA/SWCNT –2.29 –4.45 2.16 NO2PhNA/SWCNT –2.61 –4.60 1.99 TNA/SWCNT –2.35 –4.48 2.13 PhTNA/SWCNT –2.33 –4.45 2.12 NO2PhTNA/SWCNT –2.58 –4.61 2.03

a In eV. b Egap = ELUMO – EHOMO. c Computed at B3LYP/6–31G* level (reported by Zhou et al. 2004)


Figure 5: The density of states of the SWCNT,

compared with (a) nitrosoalkene and (b) thionitrosoalkene complexes.

3.2 Electronic properties and density of state The ELUMO and EHOMO energies and frontier molecular orbital energy gaps (Egap) of

SWCNT and its adduct complexes computed at the ONIOM(B3LYP/6–31G(d,p):AM1) level

are displayed in Table 2. The results showed that, Egap for the pure SWCNT was 2.17 eV which

is in good agreement with the previous results (2.20 eV) reported by Zhou et al. (2004). For

NA, TNA, PhNA, and PhTNA complexes with SWCNT, the Egap were slightly different from

SWCNT. On the other hand, for the NO2PhNA and NO2PhTNA complexes with SWCNT, their

Egap values were 1.99 and 2.03 eV, respectively which was different from the other products.


153

Plots of the density of states of the NA, PhNA, and NO2PhNA and TNA, PhTNA, and

NO2PhTNA functionalized SWCNTs, compared with the armchair (5,5) SWCNT are displayed

in Figure 5. It was shown that electronic structure of the SWCNT was sensitive to the hetero

Diels–Alder functionalizations. The band gaps of SWCNT near Fermi level become narrower,

which suggested that the conductivity of SWCNT was modified by nitrosoalkene and

thionitrosoalkene functionalizations.

4 Conclusion

The hetero-Diels-Alder functionalizations of various nitrosoalkenes (NAs) and

thionitrosoalkenes (TNAs) onto side-wall (5,5) armchair SWCNT were investigated by using

the two-layered ONIOM method at the ONIOM(B3LYP/6-31G(d.p):AM) theoretical level.

The results indicated that SWCNT can be functionalized with NAs and TNAs. The energy

barriers of TNAs funtionalized SWCNT were lower than those of NAs. This implied that TNAs

are easier to react with SWCNT than those of NAs. In addition, hetero-Diels-Alder

functionalizations affected to electronic properties and density of states of SWCNT.

5 Acknowledgements

The authors appreciate the Research Affairs, Tungmanee School, Ubonratchathani, for

partial support of this research and the facility provided by Supramolecular Chemistry Research

Unit, Department of Chemistry, Faculty of Science, Mahasarakham University. The Institute

for the Promotion of Teaching Science and Technology, THAILAND, for financial support is

also gratefully acknowledged. The authors are also grateful to Dr. Wandee Rakrai and Dr.

Chanukorn Tabtimsai for their helps.

6 References

Bahr, J.L., Yang, J., Kosynkin, D.V., Bronikowski, M.J., Smalley, R.E. & Tour, J.M. (2001). Functionalization of carbon nanotubes by electrochemical reduction of aryl diazonium salts: A bucky paper electrode. Journal of American Chemical Society. 123, 6536-6542.

Bryce, M.R., Heaton, J.N., Taylor, P.C. & Anderson, M. (1994). Diels–Alder and ene reactions of new transient thionitrosoarenes (Ar-N=S) and thionitrosoheteroarenes (Het–N=S) generated from N-(arylaminosulfanyl) and N-(heteroarylaminosulfanyl)-phthalimides: synthesis of cyclic and acyclic sulfenamides. Journal of the Chemical Society, Perkin Transactions 1, 1935-1944.


Chamssedine, F., Guérin, K., Dubois, M., Disa, E., Petit, E., El Fawal, Z. & Hamwi, A. (2011). Fluorination of single walled carbon nanotubes at low temperature: Towards the reversible fluorine storage into carbon nanotubes. Journal of Fluorine Chemistry. 132, 1072-1078.

Cho, E., Shin, S. & Yoon, Y.G. (2008). First-principles studies on carbon nanotubes functionalized with azomethine ylides. The Journal of Physical Chemistry C. 112, 11667-11672.

Delgado, J.L., Cruz, P., Langa, F., Urbina, A., Casadoc, J. & Navarretec, J.T.L. (2004). Microwave-assisted sidewall functionalization of single-wall carbon nanotubes by Diels–Alder cycloaddition. Chemical Communications. 1734-1735.

Flükiger, P., Lüthi, H.P., Portmann, S. & Weber, J. (2000) Molekel 4.3. Swiss Center for Scientific Computing, Manno.

Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Montgomery, J.A. Jr., Vreven, T., Kudin, K.N,, Burant, J.C., Millam, J.M., Iyengar, S.S., Tomasi, J., Barone, V., Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G.A., Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene, M., Li, X., Knox, J.E., Hratchian, H.P., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Ayala, P.Y., Morokuma, K., Voth, G.A., Salvador, P., Dannenberg, J.J., Zakrzewski, V.G., Dapprich, S., Daniels, A.D., Strain, M.C., Farkas, O., Malick, D.K., Rabuck, A.D., Raghavachari, K., Foresman, J.B., Ortiz, J.V., Cui, Q., Baboul, A.G., Clifford, S., Cioslowski, J., Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A., Peng, C.Y., Nanayakkara, A., Challacombe, M., Gill, P.M.W., Johnson, B., Chen, W., Wong, M.W., Gonzalez, C. & Pople, J.A. (2008) Gaussian 03, Revision E.01. Gaussian, Inc, Wallingford.

Gallos, J.K., Sarli, V.C., Varvogli, A.C., Papadoyanni, C.Z., Papaspyrou, S.D. & Argyropoulos, N.G. (2003). The hetero-Diels-Alder addition of ethyl 2-nitrosoacrylate to electron rich alkenes as a route to unnatural α-amino acids. Tetrahedron Letters. 44, 3905-3909.

Kitamura, H., Sekido, M., Takeuchi, H. & Ohno, M. (2011). The method for surface functionalization of single-walled carbon nanotubes with fuming nitric acid. Carbon. 49, 3851-3856.

Lawson, D.B. & Walker, A. (2012). Cycloaddition of ethene on a series of single-walled carbon nanotubes. Computational and Theoretical Chemistry. 981, 31-37.

Lu, X., Tian, F., Wang, N. & Zhang, Q. (2002). Organic functionalization of the sidewalls of carbon nanotubes by Diels−Alder reactions: A theoretical prediction. Organic Letters. 4, 4313-4315.

Ménard-Moyon, C., Dumas, F., Doris, E. & Mioskowski, C. (2006). Functionalization of


155

single-wall carbon nanotubes by tandem high-pressure/Cr(CO)6 activation of Diels-Alder cycloaddition. Journal of the American Chemical Society. 128, 14764-14765.

Meyyappan, M. (2005). Carbon nanotubes science and applications. CRC Press LLC, New York.

Nakahodo, T., Takahashi, K., Ishitsuka, M.O., Tsuchiya, T., Maeda, Y., Fujihara, H., Nagase, S. & Akasaka, T. (2008). Synthesis of selenylfullerene with selenium-containing dibenzo[b,g]cyclooctane moiety. Tetrahedron Letters. 49, 2302-2305.

Ohno, M., Azuma, T. & Eguchi, S. (1993). Buckminsterfullerene C60-o-quinone methide cycloadduct. Chemistry Letters. 22, 1833-1834.

Ohno, M., Kojima, S., Shirakawa, Y. & Eguchi S (1995). Hetero-diels-alder reaction of fullerene: Synthesis of thiochroman-fused C60 with o-thioquinone methide and oxidation to its S-oxides. Tetrahedron Letters. 36, 6899-6902.

Rajarajeswari, M., Iyakutti, K. & Kawazoe, Y. (2012). Noncovalent and covalent functionalization of a (5,0) single-walled carbon nanotube with alanine and alanine radicals. Journal of Molecular Modeling. 18, 771-781.

Reed, M.G. & Zhang, D.Y. (2001). Thionitroso or thiazyl? Density functional studies of relative stabilities between the two structural isomers. Journal of Molecular Structure: THEOCHEM. 548, 107-112.

Saikia, N. & Deka, R.C. (2010). Theoretical study on pyrazinamide adsorption onto covalently functionalized (5,5) metallic single-walled carbon nanotube. Chemical Physics Letters. 500, 65-70.

Tahdi, A., Titouani, S.L., Soufiaoui, M., Komiha, N., Kabbaj, O.K., Hegazi, S., Mazzah, A. & Eddaif, A. (2002). Réaction hétéro-Diels–Alder nitrosoalcènes avec le 2H-pyrrole approche expérimentale et théorique. Tetrahedron. 58, 1507-1512.

Raksaparm, U., Morakot, N. & Wanno, B. (2012). Functionalization of Diazomethyl Aromatic Compounds onto Single Wall Carbon Nanotubes: A DFT Study. International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. 3(4), 467-476.

Wanno, B. & Ruangpornvisuti, V. (2006). DFT investigation of structures of nitrosamine isomers and their transformations in gas phase. Journal of Molecular Structure: THEOCHEM. 766, 159- 164.

Wanno, B. & Ruangpornvisuti, V. (2006). Structures of gas-phase nitrosamine-dimer isomers, their interconversions and energetics: A DFT study. Journal of Molecular Structure: THEOCHEM. 775, 113- 120.

Yang, H., Ruan, X., Miao, C., Xi, H., Jiang, Y., Meng, Q. & Sun, X. (2009). Hetero-Diels–Alder reaction of [60]fullerene with nitrosoalkene. Tetrahedron Letters.


50, 7337-7339.

Yang, H.T., Wang, G.W., Xu, Y. & Huang, J.C. (2006). Heterocycloaddition of thermally generated 1,2-diaza-1,3-butadienes to [60]fullerene. Tetrahedron Letters. 47, 4129-4131.

Yim, W.L. & Johnson, J.K. (2009). Ozone oxidation of single walled carbon nanotubes from density functional theory. The Journal of Physical Chemistry C. 113, 17636-17642.

Yurovskaya, M.A. & Trushkov, I.V. (2002). Cycloaddition to buckminsterfullerene C60: advancements and future prospects. Russian Chemical Bulletin, International Edition. 51, 367-443.

Zhang, C., Li, R., Liang, Y., Shang, Z., Wang, G., Xing, Y., Pan, Y., Cai, Z., Zhao, X. & Liu, C. (2006). The nitrene cycloaddition on the sidewall of armchair single-walled carbon nanotubes. Journal of Molecular Structure: THEOCHEM. 764, 33-40.

Zhou, Z., Steigerwald, M., Hybertsen, M., Brus, L. & Friesner, R.A. (2004). Electronic Structure of Tubular Aromatic Molecules derived from the Metallic (5,5) Armchair Single Wall Carbon Nanotube. Journal of the American Chemical Society. 126, 3597-3607.

Danai Pankhao is an M.Sc. student at the Department of Chemistry, Faculty of Science, Mahasarakham University, THAILAND. He received a B.Sc. in Chemistry from Ubon Ratchathani Rajabhat University, THAILAND.

Dr.Nongnit Morakot is an Associate Professor of Department of Chemistry at Mahasarakham University. She received a B.Sc. and M.Sc. from Chiang Mai University, THAILAND. She holds a Ph.D. from Chulalongkorn University, THAILAND. Associate Professor Dr. Morakot is interested in Supramolecular Chemistry.

Dr.Somchai Keawwangchai is working in the Department of Chemistry at Mahasarakham University. He received a B.Sc. in Chemistry from Mahasarakham University, THAILAND, He holds his Ph.D. from Chulalongkorn University, THAILAND. Dr. Keawwangchai’s research fields are supramolecular investigations and host–guest investigations, reaction mechanism investigations under non-catalytic and catalytic reactions of olefins on zeolite, organometallic catalysts.

Dr.Banchob Wanno is working in the Department of Chemistry at Mahasarakham University. He received a B.Sc. in Chemistry from Mahasarakham University, THAILAND, and M.Sc. in Physical Chemistry from Mahidol University, THAILAND. He holds his Ph.D. from Chulalongkorn University, THAILAND. Dr. Wanno’s research fields are nanomaterials and nanosensors, reaction mechanism investigations and host–guest complex investigations.


*Corresponding author (A. Siswanto). Tel. +60173167312 Email address: [email protected] 2013. International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies. Volume 4 No.2 ISSN 2228-9860 eISSN 1906-9642. Online Available at http://TuEngr.com/V04/157-170.pdf

157


http://TuEngr.com

The Phenomenology of Lamban Tuha: The Local Wisdom of South Sumatra Traditional Architecture Ari Siswantoa,b*, Azizah Salim Binti Syed Salimb, Nur Dalilah Dahlanb,

Ahmad Harizac a Faculty of Engineering, University of Sriwijaya, INDONESIA b Faculty of Design and Architecture, Universiti Putra Malaysia, MALAYSIA c Faculty of Human Ecology, Universiti Putra Malaysia, MALAYSIA A R T I C L E I N F O

A B S T RA C T

Article history: Received 20 July 2012 Received in revised form 23 January 2013 Accepted 08 February 2013 Available online 14 February 2013 Keywords: Phenomenology; Local wisdom; Traditional architecture; Lamban tuha, Earthquake resistant structures.

Local wisdom of traditional architecture is towards extinction along with the existence an increasingly neglected traditional house, including the one who understands it reduced drastically. Lamban Tuha in South Sumatra has demonstrated the ability to adapt to its environment and able to withstand natural catastrophes. The study used phenomenological method to reveal information from the first person who is considered experts on the local wisdom of Lamban Tuha. This study shows the construction of kalindang provide an excellence effect of providing high flexibility in case of earthquakes. The separation structure between lower, middle and upper parts is done to give building more flexible. Local wisdom is reflections of valuable experience which can be utilized as the concept of a sustainable housing development in the context of anticipate natural disasters. The existence of Lamban Tuha is an interesting experience that can be used as thoughts on designing earthquake resistant buildings.


1. Introduction South Sumatra has a rich history of diverse culture that is very stunning in architectural

treasures. Culture is an expression of society in adapting to an environment adapted to the


158 Ari Siswanto, Azizah Salim Binti Syed Salim, Nur Dalilah Dahlan and Ahmad Hariza

necessities of life. One of the cultural heritages in architecture is traditional ulu house type

called lamban tuha. Lamban tuha which means ancient house reflects the traditional house

which resist to earthquake. Typical houses in Surabaya village are ulu house and gudang

house. Ulu house is the common term for a traditional house outside the city of Palembang

while gudang house types can be found in all areas in South Sumatra (Barendregt, 2003;

Siregar and Abu, 1985). Ulu house recognized by the local community and are classified as

lamban tuha, currently amounted two houses. The existence of lamban tuha 1 very impressive

considering the house has been occupied for 11 generations. According to the heirs, lamban

tuha 1 was founded first in the hamlet of Canti (now a forest), then it moved to Surabaya Talang

village and finally lamban tuha moved to Surabaya village in Banding Agung sub district

which closes to Lake Ranau.

One exceptional of lamban tuhas are the elastic ability of those traditional buildings

against earthquake that happen in Liwa, Lampung province in 1933. Both lamban tuhas are

the only building that remained standing despite the devastating earthquake in 1933, while the

other buildings in the village of Surabaya collapsed and mostly flattened to the ground. Typical

system of traditional construction similar to lamban tuha is only about four houses including

the new ones.

Lack of attention from the public and local government and the financial inability of

lamban tuha’s owners will caused the loss of assets in term of local wisdom (Oliver, 2006).

Traditional houses in South Sumatra have demonstrated exceptional indigenous knowledge of

our ancestors in shaping the quality of their lives. This indigenous knowledge will regain its

meaning and value in the society, we should aware of the glory of the inherited tradition. The

bearers of indigenous knowledge might be developed in recent and future for sustainable

housing development.

2. Methodology Phenomenological approach is an attempt to reveal a phenomenology of the experience

from a person in everyday life in the context of the time, place and consciousness (Creswell,

1998). Context of time has to do with history, important events, technology and character.

Context of place has to do with users, objects, physical space, the atmosphere and the


159

environment of human life. While the context of feeling have to do with experience, awareness

and knowledge visible and invisible. Based on the objectives, this research used

phenomenological method. In addition to in-depth interviews were carried out against the

respondents, this research will also see the relevance of the information provided with the

environmental conditions around it, the existence of traditional houses and history of the

houses. The first information obtained by previous research, community leaders, the owner or

tenant on the basis of their advices, and then traced the people who have a relationship with

traditional houses such as local builders, carpenters, local leaders and experts (Satori and

Komariah, 2009). Data collection will be primary data which consists of in-depth interview,

physical traditional houses and secondary data which consists of literature, journal and

research.

3. Analysis Three analyses are used in this research: a description, a comparison, and an evaluation.

The description is about architecture style, system structure and detail of structures which are

related to the environment, philosophy and their indigenous techniques. Interpretation of local

wisdom of traditional architecture would be conducted as a part of description with sources of

the owners / users, local community leaders, experts and local carpenters. The comparison is

between the people experiences in applying local wisdom.

4. Discussion

4.1 Physical Characteristic of Lamban Tuha Traditional knowledge, indigenous knowledge, and local knowledge refer to the

long-standing traditions and practices of certain regional, indigenous, or local communities.

Therefore, traditional knowledge also encompasses the wisdom, knowledge, and teachings of

these communities. Traditional knowledge has been orally accepted for generations from

person to person. The wisdom in creating natural system of thermal comfort is often found in

traditional architecture (Hardiman, 2000). Slightly different, lamban tuha has shown evidence

of local wisdom in the traditional house in anticipation of natural disasters such as earthquake.

Meanwhile, result from collective local wisdom of the contextual been able to adjust over time

and was attuned with nature and local lifestyle (Limthongsakul et al, 2005).


Vernacular that related to the process of designed and built it usually close relation

between the form and the culture. Vernacular architecture has limitation in delivering a variety

of expression however, at the same time in accordance with the characteristics of different

situations can create their respective places (Rapoport, 1969). Similar to the traditional

buildings in most parts of Indonesia, the South Sumatra traditional house shows characteristics

of timber buildings on stilts in different system structure based on the geography while others is

a kind of raft house. Due to different environment and culture, indigenous knowledge creates

traditional architecture which is adaptive with their environments. South Sumatra traditional

houses could be dismantled and rebuild in another location with mostly reusing of origin

housing materials. The typical construction of traditional house is with flexible nail-less joints,

and non-load bearing walls.

Lamban tuha has saddle-shaped roof that rise high and put the tiber angin (gable end) on

front and rear parts of the roof (Figure 1). The high rise roof has rake cross at the top as other ulu

house type. Distinctive roof form, relatively high and in accordance with the dimensions of the

house can create the beauty that is easily recognizable from a distance (Zumthor, 1998). The

construction of roof related to large span, wind and rain in specific areas. In different

geography, dwellings including roof, reflect the local knowledge, local technology and

environment (Ohno and Xihui, 2008). It explained very detail about roof structure, roof layer

construction and support systems for pitched roof. Conventional construction systems of

pitched roof in many countries always related to environmental conditions, cultural aspects and

local knowledge, it is typically seen in traditional houses such as lamban tuha. In general, the

roof truss structure of lamban tuha is very simple. Minimized the weight of steep roof of

traditional houses is an important issue for smart construction (Gruber and Herbig, 2007).

Expenses due to own weight, the wind and earthquake can reduce the risk of severe damage to

the roof.

Lower construction part of lamban tuha is a series of pillars that have stone footings

combined with a pile of round logs in rectangular shaped without finishing. Stacks of logs with

a square form support the building load known as the kalindang. Kalindang which has 7 – 11

layers of logs uses the notch on each layer as connection (Figure 5). However, not all parts of

the house supported by kalindang. Lamban tuha has a stair for entrance on the front side to

toward the garang/porch and the other in the rear for services. Porch is a transition space before


161

entering the house and serves for guests or a place to sit on an informal basis. The floor surface

in different rooms in lamban tuha has no height difference.

Figure 1: Lamban tuha 1 (left) and lamban tuha 2 (right).

The composition of rooms in lamban tuha is very simple and tends to be symmetrical.

Arrangement of rooms on lamban tuha is as follows:

a. Garang (porch), a transition space.

b. Lapang unggak (living room)

c. Lapang doh (dining room)

d. Lapang tengah (bedroom)

e. Kebik (front porch)

f. Parogan (side porch), storage for goods or old coconuts.

g. Dapo (kitchen)

h. Pagu hantu (attic), a place to store the heirlooms and spears sacred objects.


Figure 2: Floor plan of lamban tuha 1(left) and changing orientation of lamban tuha 2 (right).

Lamban tuha 1 is still the original shape as before; there has been no change in orientation

and buildings addition (Figure 2). In contrast to the lamban tuha 1, lamban tuha 2 has changed

the orientation of the building and built new stair due to consider the access road (Figure 2). At

first, lamban tuha 2 facing Qibla (west), then converted facing east because the road

consideration. As a result, the main entrance at the western is cut and moved to the east by

making entrance door facing the north.

4.2 Local Wisdom of Lamban Tuha Acting as Earthquake Resistant Based on the identification of floor plan, indigenous building materials and timber

construction system, lamban tuha has local wisdom that can be proved by experience that is

quite convincing during occupied by 11 generations (lamban tuha 1) and 6 generations (lamban

tuha 2) until today. During that time, lamban tuha 1 has moved for three times and hit by a

severe earthquake in 1933 which had destroyed all the buildings except lamban tuhas in the

village of Surabaya. The site selection to establish lamban tuha based on the land that has good

carrying capacity, far from the possibility of landslides or flooding. While the orientation of


163

lamban tuhas face the Qibla, or in this case is the west. Until know, lamban tuhas never hit by

floods and landslides struck.

According to the informants, people who sleep in lamban tuha should put the head and feet

directed west to the east, may not sleep in addition to that direction. They believe there are

certain rules of superstitious that apply to lamban tuhas (Fishwick and Vinning, 1982). Attic is

an important part of the house is believed to be a sacred place, so this place is respected and

used to store the heirlooms of their ancestral heritage. Believe in a supernatural or who has the

power associated with the presence of the house is something that is common in the past.

Lamban tuha has a simple floor plan without a rigid division of rooms and tends to

symmetrical. This indicates if the relationship between family members is very close, open and

has the nature of togetherness. Social life and communication between family members are

very close and communal. The simplicity of the floor plan and symmetrical shape is very

precise from that anticipates the influence of the earthquake. Symmetrical shapes can create a

balance of construction in every corner of the house when rocked by an earthquake.

The effect of earthquake was the collapse buildings because of bad reinforcement

structures, unreinforced masonry walls and brick walls. On the other hand, timber houses

performed relatively well compare to brick house during the earthquakes (Maidiawati. and

Sanada, 2008). Traditional houses still stand usually because of using timber structure, lighter

building material, and applying flexibility of structure. Furthermore, materials and structures

that are used in traditional houses have been made to reduce the effects that occur in the event of

earthquake (Audefroy, 2011). Lamban tuha, a typical traditional house in South Sumatra has

identifiable timber structure which resist to earthquake. Most of the sufferers of the

earthquake are the victims of collapsed concrete structures (Gruber, 2007). Building of

traditional architecture has a symmetrical shape and express in the form of floor plan and

facade. The concept of the design through the axis of symmetry generally implies a balance of

organization and function space with macro cosmos. Emphasizes a balance by referring to the

axis is the most elementary concepts of earthquake resistant buildings. Floor plan of lamban

tuha is a simple open plan design while the physical form of the house tends to be symmetrical


and proportional. The horizontal load displacement characteristics of a traditional timber house

can be simulated fairly well by adapting a mud wall and hanging wall models. This model is

embodiment of Japanese culture that has so concern about their familiar natural disaster such as

earthquake and typhoon (Fujita et al, 2004.). Most traditional houses in South Sumatra are

timber houses, only a small part of the house with bamboo or a combination of both.

Figure 3: Fundamental timber construction of press, pivot, pinch and pull.

When lamban tuha 1 was built, indigenous building materials were collected in advance by

soaking in Lake Ranau. After the perceived amount sufficient, then the house was built with no

nails, just using timber connection that is fundamental of press, pivot, pinch, and pull (Figure

3). The construction of lamban tuha could be dismantled because it has nail-less timber

construction (Figure 4). This typical construction can provide excellent flexibility in case of

oscillation due to earthquakes.

Figure 4: Nail-less timber construction of lamban tuha.


165

In addition to separating the construction of the house with lower construction, the

separation is also reinforced by the provision of ijuk (palm fibres) on the stone footing between

stilts with beams, kalindang with beams, kalindang with stone footing and stack wood blocks

on kalindang (Figure 7). Interestingly, the fibres can be seen by its presence at the bottom

construction of lamban tuha 1 (Figure 7). The informant strongly believes that the palm fibres

can serve as a sort of bearing on the structure in anticipation of earthquake. Lamban tuha 2 does

not use palm fibres in separation between structures. This difference indicate if an

understanding of fibres function of bearing structures have not understood more as a local

wisdom.

Figure 5: Kalindang of lamban tuha 1 (9 layers) and kalindang lamban tuha 2 (11 layers).

Lamban tuha has three important parts of construction that is the bottom, the middle and

the upper (Figure 6). Construction of the bottom part is the poles and kalindangs, construction

of the middle part is the framework of the house while construction of the upper part is the roof

truss. Further information mentioned that the owner of lamban tuha 1 had planned a strong and

sturdy timber construction system but also can be flexible during an earthquake. The area

around Lake Ranau is prone to earthquake disaster. Therefore, the construction lamban tuha 1’s

body is separated by lower structure, the construction of the house just rested on the structure;

this gives the effect of high flexibility. In the context of house construction, there are interesting

things, piles on the outer wall are not in a straight line with stilts or in other words, the outer

wall is cantilever, about 30 cm from the composition of stilts.


Earthquakes give bigger impact to reinforced concrete building than traditional building

(Dogangun et al, 2006). Relatively, most type of traditional building performed well during

earthquakes. Some tribes in Sumatera have local wisdom about timber construction which

resist to earthquake. Some type of traditional timber house construction which resist to

earthquake is not found in other areas such as traditional nias houses in North Sumatra, gadang

house in West Sumatra and lamban tuha in South Sumatra. Typical wood construction shows in

understanding the specific geographical conditions to adapt and survive. Lamban tuha 1 has

kalindang at four points while the lamban tuha 2 only has kalindang at two points. The number

of kalindang adapted to the dimensions of the house, the more spacious houses more kalindang

required.

Figure 6: The separation of structure into three parts (lower, middle and upper).

Lower construction of lamban tuha consists of stilts and wooden blocks shaped square

called kalindang (Figure 6). Poles and kalindangs as a whole bear lamban tuha. Poles and

kalindangs rested on stone footing, it also provides high flexibility in case of shaking during

earthquakes (Rautela and Joshi, 2008). This condition also can keep the timber from moisture

and termites influence.


167

Based on the information, the main strength of the lamban tuha 1 is four pillars as main

structure inside the house that bear the beams, these beams become the basis for pillars in the

attic. Construction of lamban tuha 1 does not have a constant pillar intact from the bottom up to

the roof.

Figure 7: The lower structure of lamban tuha (kalindang and stone footings)

Traditional houses in Nias are based on the structure of vertical and slanted posts structures

placed on a stone footing. Vertical posts and X and V are strengthens the element of this

substructure. A three-dimensional structure offers greater resistance and has the elasticity

required for not sticking in the ground (Gruber, 2007). Based on the experience of local

communities, kalindang construction on the lamban tuha has a big role in anticipating the

effects of earthquakes.

South Sumatra traditional architecture belongs to the grand tradition and requires special

skills and expertise in indigenous knowledge. Traditional architecture is not only beautiful and

elegant but also has flexible nail-less construction that has been proven to be earthquake

resistant buildings. This technique adds to the flexibility of the house. Indigenous knowledge

there is representing local wisdom that people have developed for centuries. It is based on long

experience, adapted to local culture and environment.


5. Conclusion

In principle, lamban tuha have different lower construction system with other traditional

houses in South Sumatra. A series of stilts and kalindangs worked as a system that supports

the load of the house. The building is only supported by wooden pillars and beams as a

foundation and located above stone footing and the pile of timber logs (kalindang).

Besides kalindang, lamban tuha which has connections and details of timber without nails

believed to be powerful force able to withstand earthquake shaking. The timber connections are

very appropriate considering the tensile strength and shear caused by the earthquake. Placement

of kalindang to support the weight of the house is symmetrical and synergize with stilts resting

on stone footings. Overall, the timber connection practices form of press, pivot, pinch, and

pull with reinforced by the dowel.

Local wisdom is reflections of valuable experience from the South Sumatra traditional

architecture which can be utilized as the concept of a sustainable housing development in the

context of anticipate natural disasters such as floods, landslides and earthquakes. The

existence of lamban tuha is an interesting experience of our ancestors that can be used as

thoughts on designing earthquake resistant buildings.

6. Acknowledgements The authors would like to show gratitude to the respondents who took time and patience to

share their experience of phenomenology. Also the authors deliver high appreciation to the

University of Sriwijaya, Indonesia for providing financial support for field data collection.

7. References Audefroy, J. F. (2011). Haiti: Post-earthquake lessons learned from traditional construction.

Environment and Urbanization, 23(2), 447-462.

Barendregt, B. (2003). Architecture on the move. Processes of migration and mobility in the South Sumatran highlands. In Indonesian Houses Tradition and transformation in vernacular architecture., eds. R. Schefold, J. M. P. Nas & G. Domenig. KTLV Press, Leiden.

Creswell, J. W. (1998). Qualitative Inquiry and Research Design Choosing among five


169

traditions. California, USA: Sage Publications.

Dogangun, A., O. I. Tuluk, R. Livaogle & R. Acar (2006) Traditional wooden buildings and their damages during earthquake in Turkey. Engineering Failure Analysis, 13, 981-996.

Fishwick, L. & J. Vinning (1982) Toward a phenomenology of recreation place. Environmental Psychology, 12, 57-63.

Fujita, K., T. Kondo, M. Koshihara & S. Iwata ( 2004.) On-Site Lateral Loading Test of a Traditional Timber House in Japan. Asian Architecture and Building Engineering, 3(1), 41-46.

Gruber, P. (2007). Adaptation and Earthquake Resistance of Traditional Nias Architecture. 1-12. Department of Design and Building Construction Vienna University of Technology

Gruber, P. & U. Herbig. (2007). Settlement and Housing on Nias island Adaptation and Development. Department of Design and Building Construction Vienna University of Technology.

Hardiman, G. (2000). The Wisdom of Traditional Architecture in Indonesia to Anticipate the Problem of The Thermal Comfort inside the Building. In SENVAR. ITB Bandung.

Limthongsakul, S., V. Davidongs, A. Santisan, Yamyolnham. & T. Chaisawataree. (2005). Sustainable Design and Muslim Vernacular Houses: Understanding Local Wisdom, Cultural Relationship, and Architectural Evalution. In 43rd Meeting of the authors. Kasetsart University, Thailand.

Maidiawati. & Y. Sanada (2008) Investigation and Analysis of Buildings Damage during the September 2007 Sumatra, Indonesia Earthquakes. Asian Architecture and Building Engineering, 7(2), 371-378.

Ohno, T. & W. Xihui (2008) Outline of Conventional Construction Systems for Pitched Roofs in Eastern Asia. Asian Architecture and Building Engineering, 7(1), 101-107.

Oliver, P. (2006). Built to Meet Needs: Cultural Issues in Vernacular Architecture. Burlington, M.A. USA: Architectural Press, Elsevier Ltd.

Rapoport, A. (1969). House Form and Culture. Eaglewood Cliffs, NJ: Prentice Hall, inc.

Rautela, P. & G. C. Joshi (2008) Earthquake-safe Koti Banal architecture of Uttarakhand, India. Current Science, 95(4), 475-481.

Satori, D. & A. Komariah. (2009). Metodologi Penelitian Kualitatif. Bandung: Penerbit Alfabeta.


Siregar, J. & R. Abu. (1985). Arsitektur Tradisional Daerah Sumatera Selatan. Palembang: Direktorat Sejarah dan Nilai Tradisional Departemen Pendidikan dan Kebudayaan.

Zumthor, P. (1998). Thinking Architecture. Basel, Switzerland: Birkhauser-Publisher for Architecture.

Ari Siswanto is a Ph.D. Student at the Department of Architecture, Faculty of Design and Architecture, University Putra Malaysia. MALAYSIA. He is a lecturer at the Department of Architecture, Faculty of Engineering, University of Sriwijaya. INDONESIA. He received a B.Sc. in Architecture from 10th November Surabaya Institute of Technology, INDONESIA and Master of City & Regional Planning from The Ohio State University, USA.

Dr. Azizah Salim is an Associate Professor of Department of Architecture at Universiti Putra Malaysia. She received a B.Sc. in Architecture from Robert Gordon’s Institute of Technology, Aberdeen, SCOTLAND and M.sc. from University College London. U.K. She holds a Ph.D. from University Newcastle-upon-Tyne in U.K. Her interest is in research related to housing and development policies

Dr. Ahmad Hariza is an Associate Professor of Department of Architecture at Universiti Putra Malaysia. MALAYSIA. He received a B.Sc. In Human Development from University Pertanian Malaysia. MALAYSIA. He received M.Sc. and holds a Ph.D. from The University of Birmingham, U.K. His research in housing involves person and environment relationship and housing studies.

Dr. Nur Dalilah Dahlan is a senior lecturer at the Department of Architecture at Universiti Putra Malaysia. She received a B.Sc. in Architecture from University of Malaysia, MALAYSIA and M.sc. In Architecture from Universiti Putra Malaysia. MALAYSIA. She holds a Ph.D. from Cardiff University. U.K. Her interest is in studies how people behave in response to their sensual perceptions when exposed to architectural ambiances developed using passive design approaches.

Peer Review: This article has been internationally peer-reviewed and accepted for publication according to the guidelines given at the journal’s website. Note: The original of this article was accepted and presented at the 2nd International Conference-Workshop on Sustainable Architecture and Urban Design (ICWSAUD) organized by School of Housing, Building & Planning, Universiti Sains Malaysia, Penang, Malaysia from March 3rd -5th, 2012.

More Research Publication Available at

http://TUENGR.COM

Submit your science research today.


http://tuengr.com/

http://tuengr.com/

http://tuengr.com/

http://tuengr.com/

http://tuengr.com/

http://tuengr.com/

itjemast-v04(2):: international transaction journal of engineering, management, & applied...

Documents

thailand professor dr

thailand associate prof

usa associate

greece associate

turkey associate

brazil associate

brasil associate

japan professor dr