- Email: [email protected]
Strategic project portfolio selection for national research institutes
JBR-08432; No of Pages 7 Journal of Business Research xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Business Research
Strategic project portfolio selection for national research institutes Don Jyh-Fu Jeng a,⁎, Kuo-Hsin Huang b,1 a b
Graduate Institute of Technology, Innovation & Intellectual Property Management, National Chengchi University, No. 64, Sec. 2, ZhiNan Rd., Wenshan District, Taipei 11605, Taiwan Industrial Technology Research Institute, Hsinchu 31040, Taiwan
a r t i c l e
i n f o
Article history: Received 1 December 2014 Received in revised form 1 March 2015 Accepted 1 April 2015 Available online xxxx Keywords: Multiple-criteria decision-making (MCDM) National research institute Project management Portfolio selection Research and development management
a b s t r a c t As global competition intensifies, many national research institutes (NRIs) are investing substantial resources in research and development (R&D) to gain competitive advantage and develop the national economy. However, R&D investment involves a high degree of market and technological uncertainty. The literature on project portfolio selection focuses on either quantitative economic benefits or complex criteria to assess project(s). By emphasizing the features of NRIs, the present study proposes a decision model for evaluating a project portfolio at the early initiation stage. This decision model rests on a strategy for differentiating products and services. The decision model provides a solution to a market need to maximize benefits through differentiation. A systematic hybrid multiple-criteria decision-making (MCDM) method comprising a modified Delphi method (MDM), a decision-making trial and evaluation laboratory (DEMATEL) method, and an analytic network process (ANP) offers a systematic approach to the project portfolio-selection problem. The present empirical study on the selection of alternative R&D projects in NRIs investigates the flexible electronics industry, using the hybrid MCDM method to test the decision model's effectiveness. The present study also discusses cognitive differences between NRIs and for-profit organization in terms of their R&D portfolio selection. © 2015 Elsevier Inc. All rights reserved.
1. Introduction National research institutes (NRIs) contribute substantially to a nation's economic development. In Taiwan, the government commissions NRIs to undertake research and development (R&D) projects. By adopting the outcomes of such projects, firms extend such developments to industry. NRIs play a major role in fulfilling a nation's plans for economic development by enhancing national industrial developments toward diversification and sustainability. Consequently, given limits on time and resources, NRI portfolio management is currently a topic of interest among academics. Specifically, selecting which R&D projects to undertake is crucial to technological trends and future industrial developments. Allocating funding to R&D projects does not guarantee their success. The innovation process involves a high degree of technological and market uncertainty, which can result in R&D project failure (Doctor, Newton, & Pearson, 2001; Lee, Veloso, Hounshell, & Rubin, 2010; Raz, Shenhar, & Dvir, 2002; Wang, Lin, & Huang, 2010). Company survival relies on continuous investment in developing new products or services. Selecting valuable R&D projects that satisfy future market demands is a challenge for organizations.
⁎ Corresponding author. Tel.: +886 2 2939 3091x81266; fax: +886 2 2936 3765. E-mail addresses: [email protected] (D.J.-F. Jeng), [email protected] (K.-H. Huang). 1 Tel.: +886 3 591 3741; fax: +886 3 582 0093.
Previous studies show that R&D project selection involves three major considerations: (1) the association of the project with corporate strategies (Jiang & Klein, 1999; Liberatore, 1988; Lin & Hsieh, 2004); (2) qualitative benefits and risks of undertaking candidate projects (Coffin & Taylor, 1996; Fox, Baker, & Bryant, 1984; Souder, 1986; Stewart, 1991; Wang et al., 2010); and (3) reconciliation and integration of the stakeholders' needs and desires (Carlsson, Fullér, Heikkilä, & Majlender, 2007; Dey, 2006; Hsu, Tzeng, & Shy, 2003; Huang, Chu, & Chiang, 2008; Lawson, Longhurst, & Ivey, 2006; Meade & Presley, 2002). Despite the numerous project selection criteria appearing in the aforementioned studies, such projects originate to benefit private enterprises. Furthermore, those studies propose methods without considering the interrelationships among criteria or applying a decision model to simplify the evaluation process. Focusing on NRIs, the present study proposes a decision model for selecting viable alternatives at the early stage of the R&D process. The study contributes by considering project selection separately from the benefits of national technology development and economic value creation. Because product and service differentiation is the key to allowing an enterprise to position itself successfully in a competitive market (Kotler et al., 2009, p. 503), the present study proposes a project selection decision-making framework for NRIs. In this framework, R&D projects aim to provide a solution (S) to a market need (N) and maximize an enterprise's benefit (B) through product, service, or technological differentiation (D). This need, solution, differentiation, benefit (NSDB) framework not only serves to select projects, but also to assist R&D
http://dx.doi.org/10.1016/j.jbusres.2015.06.016 0148-2963/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
2
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
practitioners to propose high-potential research projects at the initiation stage. Accordingly, the present study applies a novel hybrid multiplecriteria decision-making (MCDM) method (Jeng & Bailey, 2012) comprising a modified Delphi method (MDM) (Custer, Scarcella, & Stewart, 1999), a decision-making trial and evaluation laboratory (DEMATEL) method (Fontela & Gabus, 1976; Gabus & Fontela, 1973; Jeng, 2015), and an analytic network process (ANP) (Saaty, 1996). The MDM identifies the criteria of the NSDB framework to establish an investment decision model. Through a survey of experts, the DEMATEL method then analyzes the causal relationships between complex factors to build an impact relation map among the portfolio evaluation dimensions and criteria. This study then applies the ANP to derive weights for each factor of the MCDM problem and thereby select the optimal portfolio. Finally, the process yields the key factors by ranking the data. This research uses a case study to evaluate how NRIs prioritize projects in R&D project selection. Depicting and testing the present decision model may serve as a reference for NRI authorities. Section 2 reviews the literature to establish the foundations of the NSDB framework. Section 3 introduces the hybrid MCDM method. Section 4 describes the process of the empirical study to demonstrate the model's effectiveness. Section 5 reports results, discusses findings, and presents managerial implications. Finally, Section 6 offers concluding remarks. 2. Literature review R&D project selection plays a key role in many organizations. Hall and Nauda (1990) indicate that R&D project selection requires a strategic perspective and formal interactive process that aim to blend R&D planning with corporate business planning. The aspects of such a perspective include technology forecasting, competitor analysis, and strategic business unit planning. Liberatore (1988) describes an expert support system for industrial R&D project selection. The modeling framework links the mission, objectives, and strategy of a business unit with the criteria for selecting R&D projects. Meade and Presley (2002) assess the essential factors for project selection to formulate a decision within an enterprise's strategic objectives framework and organizational structure while considering and integrating each project's financial and strategic benefits. Lawson et al. (2006) recommend using six industry-wide categories (i.e., technical, corporate and strategy, regulatory, market, financial, and application) as filters for selecting R&D projects for profit-oriented small and medium-sized enterprises. Huang et al. (2008) list the decision criteria for early-stage project portfolio management, particularly in the selection of projects for Taiwan's Industrial Technology Development Program. These criteria could be suitable for adoption by NRIs. By adopting the analytic hierarchy process for the analysis, however, Huang et al. (2008) ignore the interdependency among the dimensions and criteria. Appendix 1 presents a thorough review of the relevant criteria. Porter (1980) recommends that firms should focus on creating a highly differentiated product or service line and marketing program to project an image as an industry leader. Matsubayashi (2007) reports a result that contradicts Bertrand's price competition model. In addition to invariably increasing firm profit, differentiation also raises consumer welfare in quality-sensitive markets. Dutta, Lach, and Rustichini (1990) demonstrates that innovation leading to product differentiation and quality improvement is a key dimension of firm competition. The Stanford Research Institute is the creator of the need, approach, benefits, and competition model (Fenwick, Daim, & Gerdsri, 2009) for developing, assessing, and presenting ideas. This model provides a systematic approach to understanding the value proposition of an original concept. The model enables innovators to present their ideas and simultaneously assesses the value of those ideas by using central parameters. The present study proposes the early-stage project selection NSDB
model, which focuses on product and service differentiation. The definitions of the NSDB are as follows. Need (N) A need should exploit a market opportunity and fulfill customer requirements relating to market size. The market should be large enough to warrant investment in R&D. Market opportunity and market size are the two key criteria for this dimension. Solution (S) A solution must meet a client's specific needs. Development of the approach to solving a problem is incremental, and the solution changes iteratively until a full proposal or business plan emerges. This business plan may include market segmentation, customer targeting, market positioning, intellectual property (IP) protection, analysis of relevant costs, deliverables, and timescales. For a new product, the solution must contain information about problems relating to product specifications, manufacturing processes, distribution, and sales. IP protection, proposal quality, and value chain information are the key criteria for this dimension. Benefits (B) Each solution for each need generates unique client benefits, such as lower costs, better performance, or faster responses. To ensure business success, benefits should be quantifiable and should yield substantial improvements rather than simply differing from the benefits that competitors provide. Differentiation (D) The innovative elements of an idea can generate differentiated solutions that represent optimal value. Access to critical IP is generally a benchmark for attracting commercial customers. Clearly stating why a solution is far superior to competitors' solutions is key to business success. 3. Method: A hybrid MCDM method The present study proposes a hybrid MCDM method comprising the MDM, DEMATEL method, and ANP. The MDM (Custer et al., 1999) refines and validates the criteria. Because criteria may affect each other, the DEMATEL method (Fontela & Gabus, 1976; Gabus & Fontela, 1973; Jeng, 2015) identifies the structure of interrelations between criteria. Finally, the ANP (Saaty, 1996) yields weights for each criterion. Jeng and Bailey (2012) show that applying the DEMATEL-based ANP for determining criteria weights and evaluating performance yields a stable converged weighted supermatrix and determines overall priorities. Through a simplified data collection process, combining the DEMATEL and ANP resolves the issue of interdependence among
Table 1 Description of experts. Organization
Code
Title
NRIs
NRI-1 NRI-2 NRI-3 NRI-4 NRI-5 NRI-6 NRI-7 NRI-8 NRI-9 NRI-10 PO-1 PO-2 PO-3 AC-1 AC-2
Deputy Center Director Deputy Division Director Deputy Division Director Department Manager Project Manager Principle Investigator Principle Investigator Principle Investigator Principle Investigator Principle Investigator Chairman CEO General Manager Professor Deputy Professor
For-profit organizations
Academia
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
3
Table 2 Finalized NSDB framework after the MDM survey. Dimension
Criteria
Description
Need (N)
New market potential (NMP) Potential size of market (PSM) Time to market (TTM) Urgent customer requirement (UCR) Ability to meet likely future regulations (MFR) Technical resource availability (TRA) Opportunity for technical success (OTS) Content of a technical plan (CTP)
Potential to meet the requirements of a new market. The potential size or growth of a market for products based on the proposed technology. The time from product conception to commercial sale. The urgency of customer demands. The extent to which a proposed technology coincides with science and technology policy. The degree to which a project has access to technical resources. The opportunity for success of a proposed technology. The project must be described in detail to answer questions on clear and concise planning, clear identification of the core technology, feasibility of the technical approach, and the major technical constraints. How advanced is the proposed technology compared with existing technology? How innovative is the proposed technology? The critical characteristics of a technology for product or industry development. The proprietary technology position through the collection of patents owned. The maximum possible return on an investment. The extent to which a proposed technology may further improve technological developments and competiveness based on the project outcomes. Benefits to society achieved through the improvement of national standards of quality, environmental protection, industrial safety, national image, and industrial standards (QESIS). The opportunity for the market success of a product based on a proposed technology.
Solution (S)
Differentiation (D)
Benefit (B)
Advancements in technology (AOT) Innovation of technology (IOT) Key of technology (KOT) Patent portfolios (PP) Potential return on investment (PROI) Aid an organization in competing in the market (AOC) Improvement on the QESIS (QESIS) Opportunity for market success (OMS)
dimensions and criteria in a complex hierarchical system. In practice, using this hybrid MCDM method is more rational than using standalone methods such as the analytic hierarchy process.
4. Empirical study: Case of R&D project portfolio selection by an NRI Fifteen experts (Table 1) completed an MDM survey and DEMATELbased ANP. Validating the unique strategic goal of NRIs involves assigning experts to either an experimental group (i.e., NRIs; n = 10) or a control group (i.e., for-profit organizations; n = 3), and testing the cognitive differences between these two groups of experts and a third group of academics (n = 2). A case study of R&D projects for equipment development in the flexible electronic industry involves evaluating and weighting the proposals in the project portfolio. The five R&D proposals are (1) a roll-to-roll touch-panel lamination system (R2R); (2) slot die wet-coating equipment (SD); (3) a plasma-enhanced chemical vapor deposit (PECVD); (4) a measuring instrument (MI); and (5) a laser-patterning machine (LPM). Evaluation of these alternatives took place using the hybrid MCDM method, and comparison took place using the NSDB model.
The research team described the issue at hand and provided experts with a summary of relevant research. Experts received the questionnaires in advance via email and responded in face-to-face interviews. Responses converged after five rounds of MDM. During the process, recommendation and evaluation of some criteria missing from the preliminary NSDB framework took place by ranking the criteria in order of importance. Table 2 summarizes the final criteria. Applying the DEMATEL method clarifies the interrelationships among the dimensions and criteria arising from the expert survey. Applying this process generates the total-influence matrix T (Table 3) and identifies the factors of concern (Table 4). This analysis yields a comprehensive impact relation map (Fig. 1). The DEMATEL analysis results enable the formulation of the unweighted supermatrix W, which is the principal eigenvector of the criteria pairwise comparison matrix relative to the NSDB dimensions. Obtaining the normalized total-influence matrix Ts enables the retrieval of the weighted supermatrix Ww. The last stage of the process yields the converged weighted supermatrix by raising the matrix to a largeenough power such that the weights converge and the matrix becomes a long-term stable supermatrix (Jeng & Bailey, 2012). Table 5 shows the weighted dimensions and criteria of the NSDB model.
Table 3 DEMATEL total-influence matrix T. Need (N)
N
S
D
B
NMP PSM TTM UCR MFR TRA OTS CTP AOT IOT KOT PP PROI AOC QESIS OMS
Solution (S)
Differentiation (D)
Benefit (B)
NMP
PSM
TTM
UCR
MFR
TRA
OTS
CTP
AOT
IOT
KOT
PP
PROI
AOC
QESIS
OMS
0.083 0.168 0.117 0.132 0.116 0.096 0.092 0.058 0.114 0.129 0.114 0.125 0.107 0.118 0.083 0.116
0.168 0.104 0.119 0.142 0.125 0.12 0.099 0.075 0.122 0.148 0.133 0.143 0.13 0.133 0.083 0.137
0.174 0.187 0.095 0.2 0.125 0.139 0.144 0.1 0.135 0.157 0.159 0.144 0.152 0.149 0.092 0.171
0.129 0.142 0.113 0.083 0.097 0.134 0.121 0.083 0.09 0.103 0.118 0.113 0.099 0.113 0.07 0.121
0.132 0.155 0.105 0.121 0.09 0.146 0.135 0.11 0.162 0.162 0.13 0.147 0.147 0.147 0.12 0.153
0.169 0.201 0.145 0.189 0.158 0.138 0.186 0.119 0.174 0.205 0.2 0.187 0.192 0.184 0.11 0.207
0.172 0.194 0.175 0.202 0.161 0.244 0.135 0.165 0.183 0.212 0.217 0.215 0.204 0.199 0.11 0.224
0.093 0.107 0.088 0.124 0.105 0.146 0.097 0.052 0.109 0.107 0.12 0.11 0.103 0.111 0.066 0.122
0.141 0.154 0.116 0.127 0.126 0.173 0.143 0.106 0.118 0.188 0.186 0.179 0.132 0.163 0.111 0.169
0.192 0.197 0.16 0.192 0.168 0.21 0.188 0.122 0.228 0.17 0.263 0.235 0.191 0.213 0.141 0.218
0.184 0.201 0.154 0.193 0.165 0.214 0.181 0.131 0.235 0.252 0.165 0.248 0.187 0.197 0.115 0.208
0.208 0.239 0.179 0.187 0.188 0.209 0.194 0.146 0.263 0.275 0.267 0.176 0.214 0.232 0.122 0.233
0.2 0.245 0.178 0.197 0.181 0.202 0.203 0.12 0.201 0.227 0.225 0.23 0.148 0.223 0.133 0.237
0.172 0.201 0.186 0.193 0.19 0.229 0.216 0.127 0.229 0.258 0.266 0.247 0.219 0.166 0.144 0.25
0.12 0.139 0.108 0.126 0.128 0.128 0.134 0.092 0.143 0.157 0.139 0.136 0.153 0.159 0.066 0.178
0.207 0.238 0.219 0.219 0.214 0.254 0.226 0.153 0.239 0.267 0.271 0.255 0.244 0.273 0.158 0.191
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
4
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
Table 4 Influence of the factors of concern.
Table 5 Weighted dimensions and criteria of the NSDB model.
Dimensions/Criteria
di
ri
di + ri
di − ri
Dimensions/Criteria
Need (N) NMP PSM TTM UCR Solution (S) MFR TRA OTS CTP Differentiation (D) AOT IOT KOT PP Benefit (B) PROI AOC QESIS OMS
0.644 2.544 2.872 2.257 2.627 0.586 2.337 2.782 2.494 1.759 0.727 2.745 3.017 2.973 2.890 0.629 2.622 2.78 1.724 2.935
0.488 1.768 1.981 2.323 1.729 0.600 2.162 2.764 3.012 1.660 0.736 2.332 3.088 3.030 3.332 0.761 3.15 3.293 2.106 3.628
1.131 4.312 4.853 4.580 4.356 1.186 4.499 5.546 5.506 3.419 1.463 5.077 6.105 6.003 6.222 1.390 5.772 6.073 3.830 6.563
0.156 0.776 0.891 −0.066 0.898 −0.014 0.175 0.018 −0.518 0.099 −0.010 0.413 −0.071 −0.057 −0.442 −0.132 −0.528 −0.513 −0.382 −0.693
Need (N) NMP PSM TTM UCR Solution (S) MFR TRA OTS CTP Differentiation (D) AOT IOT KOT PP Benefit (B) PROI AOC QESIS OMS
To verify the cognitive differences between NRIs and for-profit organizations, Table 6 shows the DEMATEL-based ANP results of the weighting and ranking. To evaluate the R&D projects, NRI managers evaluated the project alternatives by assigning them a score between 1 and 10. Table 7 shows the score regarding five alternatives with respect to the criteria in terms of the dimension of NSDB. The five project alternatives receive scores according to the NSDB model weights (Table 5).
Local weight (ranking) 0.1871 (4) 0.2265 (3) 0.2544 (2) 0.2956 (1) 0.2235 (4) 0.2325 (3) 0.2262 (3) 0.2884 (2) 0.3127 (1) 0.1727 (4) 0.2855 (2) 0.1996 (4) 0.2625 (2) 0.2567 (3) 0.2813 (1) 0.2949 (1) 0.2571 (3) 0.2716 (2) 0.1736 (4) 0.2976 (1)
Global weight (ranking) 0.0424 (14) 0.0476 (13) 0.0553 (10) 0.0418 (15) 0.0526 (11) 0.0670 (8) 0.0727 (7) 0.0401 (16) 0.0570 (9) 0.0749 (5) 0.0733 (6) 0.0803 (2) 0.0758 (4) 0.0801 (3) 0.0512 (12) 0.0878 (1)
5. Discussion 5.1. Results Table 4 shows the strength of influences among the dimensions and criteria according to the DEMATEL analysis results. Fig. 1 shows that among the NSDB model dimensions, the benefit dimension is the most important, whereas the need dimension is the least important.
Fig. 1. Comprehensive impact relation map for NSDB model.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
5
Table 6 Cognitive difference in the weighted criteria of the NSDB model between the experimental and control groups. Dimension/Criteria
Need (N)
Solution (S)
Differentiation (D)
Benefit (B)
NRI
New market potential (NMP) The potential size of market (PSM) Time to market (TTM) Urgent customer requirement(UCR) Ability to meet likely future regulations (MFR) Technical resource availability (TRA) Opportunity of technical success (OTS) Contents of technical plan (CTP) Advancement of technology (AOT) Innovation of technology (IOT) Key of technology (KOT) Patent portfolios (PP) Potential return on investment (PROI) Aid the organization in competing in the market (AOC) Improvement on QESIS (QESIS) Opportunity of market success (OMS)
In addition, results show that need is a causal factor of the other dimensions. Two of the top four degrees of importance are opportunity for market success (OMS) (6.563) and aiding the organization to compete in the market (AOC) (6.073), both of which belong to the benefit dimension. The other two criteria are patent portfolios (PP) (6.222) and innovation of technology (IOT) (6.105), both of which belong to the differentiation dimension. Two criteria with the lowest degrees of importance are contents of a technical plan (CTP) (3.419) and improvements in benefits to society through improving national standards of quality, environmental protection, industrial safety, national image, and industrial standards (QESIS) (3.830). Negative values of (di–ri) represent the role of receiver (Jeng & Bailey, 2012). In addition to advances in technology (AOT) (0.413), seven of the nine net receivers belong to the dimensions of differentiation and benefit. The other receivers are opportunity of technical success (OTS) (− 0.518) and time to market (TTM) (− 0.066), which belong to the dimensions of solution and need, respectively. Criteria belonging to the need dimension are all strong causal factors, except TTM (−0.066), which is a weak receiver. Although criteria belonging to the need dimension are the least important, they are highly influential on the other criteria.
For-profit organization
Weight
Ranking
Weight
Ranking
0.025 0.026 0.023 0.031 0.092 0.070 0.085 0.051 0.052 0.061 0.062 0.106 0.074 0.093 0.053 0.096
15 14 16 13 4 7 5 12 11 9 8 1 6 3 10 2
0.019 0.018 0.038 0.031 0.078 0.061 0.083 0.076 0.052 0.073 0.071 0.086 0.085 0.094 0.042 0.093
15 16 13 14 6 10 5 7 11 8 9 3 4 1 12 2
The weighted NSDB model results show that OMS plays the most crucial role (weight = 0.088), followed by PP (weight = 0.080), AOC (weight = 0.080), and potential return on investment (PROI) (weight = 0.076). Three of the four highest priorities belong to the benefit dimension. This result implies that the benefit dimension is the most important when organizations decide to invest in an R&D project. Moreover, the weights of the NSDB model show that the lowest four criteria are CTP, urgent customer requirement (UCR), new market potential (NMP), and the potential size of market (PSM) (weights = 0.040, 0.042, 0.042, and 0.048, respectively). Three of these four criteria belong to the need dimension, indicating that this dimension has the least influence when NRIs consider whether to invest in an R&D project. In addition, CTP has the lowest weight, implying that the content of an R&D proposal is less relevant than other factors are as long as the proposal's concept is innovative and can provide benefits. According the result of DEMATEL-based ANP (Table 7), the assessment committee prioritizes the five R&D projects as follows: PECVD ≻ LPM ≻ R2R ≻ MI ≻ SD. Comparing the criteria-weighting results between the NRIs and forprofit organizations reveals differences between these two groups. The for-profit organizations rank TTM as the most important criterion
Table 7 Priority and comparison of the R&D project alternatives. Dimension/ Criteria
N
NMP PSM TTM UCR S MFR TRA OTS CTP D AOT IOT KOT PP B PROI AOC QESIS OMS Total Score Priority
Alternatives LPM
PECVD
R2R
SD
MI
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
6.5 5.5 8 6.5 5.75 7.5 7 7 7.5 7.25 7 6.75 5.75 6.75 5.75 6.25
0.276 0.262 0.442 0.272 0.302 0.503 0.509 0.281 0.427 0.543 0.513 0.542 0.436 0.541 0.294 0.548 6.692 2
7.75 8.25 8.25 7.75 5.75 6.75 7.75 7 7.75 7.25 8 7.25 8.5 8.25 6 7.75
0.328 0.393 0.456 0.324 0.302 0.453 0.563 0.281 0.442 0.543 0.586 0.582 0.644 0.661 0.307 0.680 7.547 1
7.5 6.75 7 8.5 5.75 6.75 7.75 5.25 5.25 5.75 5.75 5.5 6.25 7.25 6 8.5
0.318 0.321 0.387 0.355 0.302 0.453 0.563 0.211 0.299 0.431 0.421 0.442 0.474 0.581 0.307 0.746 6.612 3
6 6.75 7.5 6.25 5.75 5.75 5.75 5.25 6.75 6.75 5.5 5.5 7.5 7.75 6 6.25
0.254 0.321 0.415 0.261 0.302 0.386 0.418 0.211 0.385 0.506 0.403 0.442 0.569 0.621 0.307 0.548 6.349 5
6.25 5.5 7.75 6.75 6.25 7.5 6.75 6.5 6.5 6 6.25 6.5 5.75 7 6 7.25
0.265 0.262 0.429 0.282 0.329 0.503 0.491 0.261 0.370 0.450 0.458 0.522 0.436 0.561 0.307 0.636 6.561 4
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
6
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
industries to compete, thereby improving the national economy. Given this consideration, the portfolio selection criteria and its importance differ from those of for-profit enterprises. This study shows that the portfolio selection process of NRIs tends to involve selecting unique and differentiated R&D alternatives. Examining the government-assigned key performance indicators of NRIs (e.g., number of technical reports, number of patents filed, scale of technology transfer, industrial investment, capability to support other projects or derive new projects, and promoting people's skill or employment) shows that such indicators relate to the benefit dimension. This research reveals that QESIS is less important. NRIs, which assist governments with economic or social development, care less about social welfare than about benefit, implying that NRIs should enhance their social responsibility when considering technological developments.
(Rank 13) in the need dimension, whereas the NRIs rank TTM as the least important criterion (Rank 16). This difference implies NRIs and for-profit organizations have different priorities for meeting market opportunities. Regarding the solution dimension, the for-profit organizations and NRIs rank the priority of technical resource availability (TRA) as 10th and 7th, respectively. In contrast, the for-profit organizations consider CTP to be more crucial (Rank 7) than the NRIs do (Rank 12), implying that for-profit organizations care more about technical planning from the perspective of internal support, whereas the NRIs care more about resource availability for innovative development. For the differentiation dimension, both types of organization assign relatively equal importance to AOT (Rank 11), IOT (Rank 9 and 8), key of technology (KOT) (Rank 8 and 9). NRIs and for-profit organizations rank patent portfolios (PP) 1st and 3rd, respectively, indicating differences between the two types of organization. NRIs pursue national technology development, whereas for-profit organizations focus on competing in the market. Regarding the criteria in the benefit dimension, managers of NRIs and mangers of for-profit organizations rank PROI 6th and 4th, respectively. NRIs and for-profit organizations rank AOC 3rd and 1st, respectively, indicating that competitive strength is the most important criterion for market survival according to for-profit organizations. Both groups rank the criterion of OMS 2nd, implying that market success remains the major goal of R&D activities in both types of organization. The NRIs, however, rank improvement in QESIS higher than for-profit organizations do because NRIs undertake government tasks relating to social welfare.
6. Conclusions The present study employs a hybrid MCDM method to overcome complexity when exploring interdependence and feedback between criteria and alternatives. The proposed NSDB model is a systematic approach for R&D project portfolio selection. This empirical study demonstrates the effectiveness of applying the hybrid MCDM method within the NSDB framework. The issues of business, competitiveness, and technological development act as factors to evaluate R&D alternatives at the early stage of a project. This study is an in-depth exploration of the characteristics of NRIs. Unlike previous studies, the present study focuses primarily on forprofit enterprises. The findings show that private enterprises conduct R&D for the sake of long-term sustainability, whereas NRIs promote industrial development through innovative technology R&D. This research discusses the methods that NRIs employ when they select R&D projects and weight the importance of decision-making criteria. Future research should consider investigating effective mechanisms for project collaboration between NRIs and for-profit organizations to maximize benefits for both parties.
5.2. Implications Innovation and IP drive most R&D-oriented NRI business models. These drivers enable the transfer of technologies to industry, helping firms improve their products or services and compete in the market. The goal of NRIs is to develop new technologies that assist local Appendix 1 Literature review of the relevant criteria for project selection
Criteria Benefit/cost ratio Cost Capital investment Rate of return Contribution of profitability Growth rate Payback period Direct cash flow Effect on existing market share Effect on existing market outlook New market potential Potential return on investment Net present value Contribution to organizational goals/objectives Aid the organization in competing in the market Improvements on research capability Internal political decisions Strategic fit Importance to the organization for the future success Contribution to knowledge Importance to the functioning of the organization Public relation effect Importance to organization's critical success factors Fit with company business plan Ability to meet likely future regulations External regulations Staff training, development and job satisfaction
Huang et al. (2008)
Lawson et al. (2006)
Meade and Presley (2002)
Jiang and Klein (1999)
Stewart (1991)
Liberatore (1988)
v v v v v v v v v v v v
v v v v v
v
v v v v v v v v v v v v
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
7
Appendix 1 (continued) Criteria Required by regulations Coincidence with S&T policy Response to competition Required by customers/suppliers QESIS (1)Improvements on QESIS (2)Workplace safety (3)Environmental considerations (4)New industry standards (5)Environmental and safety consideration lawsuit requires information The potential size of market Technology spillover effects Benefits for human life Technical resource availability Technical support Equipment support Timing for project Product range growth potential Product life cycle Synergy with other products/ processes Ability to implement production/process Patentability/design protection Existence of project champion Existence of required competence Number and strength of competitors Isolated, simple, and modular project High visibility of project Basic subsystem to system Basic module for operations Availability of skilled IS personnel Availability of needed technology Advancement of technology Innovation of technology Key of technology Proprietary technology Generics of technology Technological connections Technological extendibility Specification of technology Appropriateness for research period Appropriateness for research cost Evidence of scientific feasibility
Huang et al. (2008)
Lawson et al. (2006)
Meade and Presley (2002)
Jiang and Klein (1999)
Stewart (1991)
v
v
v v
v v
v(4)
v
Liberatore (1988)
v
v(1)(5)
v(2)(3)
v v v v v v v
v
v
v
v v v
v
v v v v v
v v v v
v v v v v v v v v v v v v v v
References Carlsson, C., Fullér, R., Heikkilä, M., & Majlender, P. (2007). A fuzzy approach to R&D project portfolio selection. International Journal of Approximate Reasoning, 44(2), 93–105. Coffin, M. A., & Taylor, B. W. (1996). Multiple criteria R&D project selection and scheduling using fuzzy logic. Computers & Operations Research, 23(3), 207–220. Custer, R. L., Scarcella, J. A., & Stewart, B. R. (1999). The modified Delphi technique—A rotational modification. Journal of Vocational and Technical Education, 15(2), 50–58. Dey, P. K. (2006). Integrated project evaluation and selection using multiple attribute decision making technique. International Journal of Production Economics, 103(1), 90–103. Doctor, R. N., Newton, D. P., & Pearson, A. (2001). Managing uncertainty in research and development. Technovation, 21(2), 79–90. Dutta, P., Lach, S., & Rustichini, A. (1990). Innovation and product differentiation. Unpublished Discussion Paper 894. Northwestern University, Center for Mathematical Studies in Economics and Management Science. Fenwick, D., Daim, T. U., & Gerdsri, N. (2009). Value Driven Technology Road Mapping (VTRM) process integrating decision making and marketing tools: Case of Internet security technologies. Technological Forecasting and Social Change, 76(8), 1055–1077. Fontela, E., & Gabus, A. (1976). The DEMATEL observer, DEMATEL 1976 Report. Geneva, Switzerland: Battelle Geneva Research Center. Fox, E., Baker, N., & Bryant, J. (1984). Economic models for R&D project selection in the presence of project interactions. Management Science, 30(7), 890–902. Gabus, A., & Fontela, E. (1973). Perceptions of the world problematique: Communication procedure, communicating with those bearing collective responsibility. DEMATEL Report No. 1. Geneva, Switzerland: Battelle Geneva Research Center. Hall, D., & Nauda, A. (1990). An interactive approach for selecting IR&D projects. IEEE Transactions on Engineering Management, 37(2), 126–133. Hsu, Y. -G., Tzeng, G. -H., & Shy, J. Z. (2003). Fuzzy multiple criteria selection of government-sponsored frontier technology R&D projects. R&D Management, 33(5), 539–551. Huang, C. -C., Chu, P. -Y., & Chiang, Y. -H. (2008). A fuzzy AHP application in governmentsponsored R&D project selection. OMEGA: The International Journal of Management Science, 36(6), 1038–1052.
v v
Jeng, D. J. -F. (2015). Generating a causal model of supply chain collaboration using the fuzzy DEMATEL technique. Computers & Industrial Engineering, 87, 283–295. Jeng, D. J. -F., & Bailey, T. (2012). Assessing customer retention strategies in mobile telecommunications: A hybrid MCDM approach. Management Decision, 50(9), 1570–1595. Jiang, J. J., & Klein, G. (1999). Project selection criteria by strategic orientation. Information & Management, 36(2), 63–75. Kotler, P., Armstrong, G., Ang, S. H., Leong, S. M., Tan, C. T., & Hon-Ming, O. Y. (2009). Principles of marketing: A global perspective. Cincinnati, Ohio: Pearson Education Inc. Lawson, C. P., Longhurst, P. J., & Ivey, P. C. (2006). The application of a new research and development project selection model in SMEs. Technovation, 26, 242–250. Lee, J., Veloso, F. M., Hounshell, D. A., & Rubin, E. S. (2010). Forcing technological change: A case of automobile emissions control technology development in the US. Technovation, 30(4), 249–264. Liberatore, M. J. (1988). Applications of the AHP. Paper presented at the Proc. 6th Int. Conf. on Mathematical Modelling. Lin, C., & Hsieh, P. -J. (2004). A fuzzy decision support system for strategic portfolio management. Decision Support Systems, 38(3), 383–398. Matsubayashi, N. (2007). Price and quality competition: The effect of differentiation and vertical integration. European Journal of Operational Research, 180(2), 907–921. Meade, L., & Presley, A. (2002). R&D project selection using the analytic network process. IEEE Transactions on Engineering Management, 49(1), 59–66. Porter, M. E. (1980). Competitive strategy: Techniques for analyzing industries and competitors, Vol. Chapter 2, New York: The Free Press. Raz, T., Shenhar, A. J., & Dvir, D. (2002). Risk management, project success, and technological uncertainty. R&D Management, 32(2), 101–109. Saaty, T. L. (Ed.). (1996). Decision making with dependence and feedback: Analytic network process. Pittsburgh: RWS Publications. Souder, W. E. (1986). Achieving organizational consensus with respect to R&D project selection. Management Science, 21(6), 669–681. Stewart, T. J. (1991). A multi-criteria decision support system for R&D project selection. Journal of the Operational Research Society, 42(1), 17–26. Wang, J., Lin, W., & Huang, Y. -H. (2010). A performance-oriented risk management framework for innovative R&D projects. Technovation, 30(11–12), 601–611.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
Contents lists available at ScienceDirect
Journal of Business Research
Strategic project portfolio selection for national research institutes Don Jyh-Fu Jeng a,⁎, Kuo-Hsin Huang b,1 a b
Graduate Institute of Technology, Innovation & Intellectual Property Management, National Chengchi University, No. 64, Sec. 2, ZhiNan Rd., Wenshan District, Taipei 11605, Taiwan Industrial Technology Research Institute, Hsinchu 31040, Taiwan
a r t i c l e
i n f o
Article history: Received 1 December 2014 Received in revised form 1 March 2015 Accepted 1 April 2015 Available online xxxx Keywords: Multiple-criteria decision-making (MCDM) National research institute Project management Portfolio selection Research and development management
a b s t r a c t As global competition intensifies, many national research institutes (NRIs) are investing substantial resources in research and development (R&D) to gain competitive advantage and develop the national economy. However, R&D investment involves a high degree of market and technological uncertainty. The literature on project portfolio selection focuses on either quantitative economic benefits or complex criteria to assess project(s). By emphasizing the features of NRIs, the present study proposes a decision model for evaluating a project portfolio at the early initiation stage. This decision model rests on a strategy for differentiating products and services. The decision model provides a solution to a market need to maximize benefits through differentiation. A systematic hybrid multiple-criteria decision-making (MCDM) method comprising a modified Delphi method (MDM), a decision-making trial and evaluation laboratory (DEMATEL) method, and an analytic network process (ANP) offers a systematic approach to the project portfolio-selection problem. The present empirical study on the selection of alternative R&D projects in NRIs investigates the flexible electronics industry, using the hybrid MCDM method to test the decision model's effectiveness. The present study also discusses cognitive differences between NRIs and for-profit organization in terms of their R&D portfolio selection. © 2015 Elsevier Inc. All rights reserved.
1. Introduction National research institutes (NRIs) contribute substantially to a nation's economic development. In Taiwan, the government commissions NRIs to undertake research and development (R&D) projects. By adopting the outcomes of such projects, firms extend such developments to industry. NRIs play a major role in fulfilling a nation's plans for economic development by enhancing national industrial developments toward diversification and sustainability. Consequently, given limits on time and resources, NRI portfolio management is currently a topic of interest among academics. Specifically, selecting which R&D projects to undertake is crucial to technological trends and future industrial developments. Allocating funding to R&D projects does not guarantee their success. The innovation process involves a high degree of technological and market uncertainty, which can result in R&D project failure (Doctor, Newton, & Pearson, 2001; Lee, Veloso, Hounshell, & Rubin, 2010; Raz, Shenhar, & Dvir, 2002; Wang, Lin, & Huang, 2010). Company survival relies on continuous investment in developing new products or services. Selecting valuable R&D projects that satisfy future market demands is a challenge for organizations.
⁎ Corresponding author. Tel.: +886 2 2939 3091x81266; fax: +886 2 2936 3765. E-mail addresses: [email protected] (D.J.-F. Jeng), [email protected] (K.-H. Huang). 1 Tel.: +886 3 591 3741; fax: +886 3 582 0093.
Previous studies show that R&D project selection involves three major considerations: (1) the association of the project with corporate strategies (Jiang & Klein, 1999; Liberatore, 1988; Lin & Hsieh, 2004); (2) qualitative benefits and risks of undertaking candidate projects (Coffin & Taylor, 1996; Fox, Baker, & Bryant, 1984; Souder, 1986; Stewart, 1991; Wang et al., 2010); and (3) reconciliation and integration of the stakeholders' needs and desires (Carlsson, Fullér, Heikkilä, & Majlender, 2007; Dey, 2006; Hsu, Tzeng, & Shy, 2003; Huang, Chu, & Chiang, 2008; Lawson, Longhurst, & Ivey, 2006; Meade & Presley, 2002). Despite the numerous project selection criteria appearing in the aforementioned studies, such projects originate to benefit private enterprises. Furthermore, those studies propose methods without considering the interrelationships among criteria or applying a decision model to simplify the evaluation process. Focusing on NRIs, the present study proposes a decision model for selecting viable alternatives at the early stage of the R&D process. The study contributes by considering project selection separately from the benefits of national technology development and economic value creation. Because product and service differentiation is the key to allowing an enterprise to position itself successfully in a competitive market (Kotler et al., 2009, p. 503), the present study proposes a project selection decision-making framework for NRIs. In this framework, R&D projects aim to provide a solution (S) to a market need (N) and maximize an enterprise's benefit (B) through product, service, or technological differentiation (D). This need, solution, differentiation, benefit (NSDB) framework not only serves to select projects, but also to assist R&D
http://dx.doi.org/10.1016/j.jbusres.2015.06.016 0148-2963/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
2
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
practitioners to propose high-potential research projects at the initiation stage. Accordingly, the present study applies a novel hybrid multiplecriteria decision-making (MCDM) method (Jeng & Bailey, 2012) comprising a modified Delphi method (MDM) (Custer, Scarcella, & Stewart, 1999), a decision-making trial and evaluation laboratory (DEMATEL) method (Fontela & Gabus, 1976; Gabus & Fontela, 1973; Jeng, 2015), and an analytic network process (ANP) (Saaty, 1996). The MDM identifies the criteria of the NSDB framework to establish an investment decision model. Through a survey of experts, the DEMATEL method then analyzes the causal relationships between complex factors to build an impact relation map among the portfolio evaluation dimensions and criteria. This study then applies the ANP to derive weights for each factor of the MCDM problem and thereby select the optimal portfolio. Finally, the process yields the key factors by ranking the data. This research uses a case study to evaluate how NRIs prioritize projects in R&D project selection. Depicting and testing the present decision model may serve as a reference for NRI authorities. Section 2 reviews the literature to establish the foundations of the NSDB framework. Section 3 introduces the hybrid MCDM method. Section 4 describes the process of the empirical study to demonstrate the model's effectiveness. Section 5 reports results, discusses findings, and presents managerial implications. Finally, Section 6 offers concluding remarks. 2. Literature review R&D project selection plays a key role in many organizations. Hall and Nauda (1990) indicate that R&D project selection requires a strategic perspective and formal interactive process that aim to blend R&D planning with corporate business planning. The aspects of such a perspective include technology forecasting, competitor analysis, and strategic business unit planning. Liberatore (1988) describes an expert support system for industrial R&D project selection. The modeling framework links the mission, objectives, and strategy of a business unit with the criteria for selecting R&D projects. Meade and Presley (2002) assess the essential factors for project selection to formulate a decision within an enterprise's strategic objectives framework and organizational structure while considering and integrating each project's financial and strategic benefits. Lawson et al. (2006) recommend using six industry-wide categories (i.e., technical, corporate and strategy, regulatory, market, financial, and application) as filters for selecting R&D projects for profit-oriented small and medium-sized enterprises. Huang et al. (2008) list the decision criteria for early-stage project portfolio management, particularly in the selection of projects for Taiwan's Industrial Technology Development Program. These criteria could be suitable for adoption by NRIs. By adopting the analytic hierarchy process for the analysis, however, Huang et al. (2008) ignore the interdependency among the dimensions and criteria. Appendix 1 presents a thorough review of the relevant criteria. Porter (1980) recommends that firms should focus on creating a highly differentiated product or service line and marketing program to project an image as an industry leader. Matsubayashi (2007) reports a result that contradicts Bertrand's price competition model. In addition to invariably increasing firm profit, differentiation also raises consumer welfare in quality-sensitive markets. Dutta, Lach, and Rustichini (1990) demonstrates that innovation leading to product differentiation and quality improvement is a key dimension of firm competition. The Stanford Research Institute is the creator of the need, approach, benefits, and competition model (Fenwick, Daim, & Gerdsri, 2009) for developing, assessing, and presenting ideas. This model provides a systematic approach to understanding the value proposition of an original concept. The model enables innovators to present their ideas and simultaneously assesses the value of those ideas by using central parameters. The present study proposes the early-stage project selection NSDB
model, which focuses on product and service differentiation. The definitions of the NSDB are as follows. Need (N) A need should exploit a market opportunity and fulfill customer requirements relating to market size. The market should be large enough to warrant investment in R&D. Market opportunity and market size are the two key criteria for this dimension. Solution (S) A solution must meet a client's specific needs. Development of the approach to solving a problem is incremental, and the solution changes iteratively until a full proposal or business plan emerges. This business plan may include market segmentation, customer targeting, market positioning, intellectual property (IP) protection, analysis of relevant costs, deliverables, and timescales. For a new product, the solution must contain information about problems relating to product specifications, manufacturing processes, distribution, and sales. IP protection, proposal quality, and value chain information are the key criteria for this dimension. Benefits (B) Each solution for each need generates unique client benefits, such as lower costs, better performance, or faster responses. To ensure business success, benefits should be quantifiable and should yield substantial improvements rather than simply differing from the benefits that competitors provide. Differentiation (D) The innovative elements of an idea can generate differentiated solutions that represent optimal value. Access to critical IP is generally a benchmark for attracting commercial customers. Clearly stating why a solution is far superior to competitors' solutions is key to business success. 3. Method: A hybrid MCDM method The present study proposes a hybrid MCDM method comprising the MDM, DEMATEL method, and ANP. The MDM (Custer et al., 1999) refines and validates the criteria. Because criteria may affect each other, the DEMATEL method (Fontela & Gabus, 1976; Gabus & Fontela, 1973; Jeng, 2015) identifies the structure of interrelations between criteria. Finally, the ANP (Saaty, 1996) yields weights for each criterion. Jeng and Bailey (2012) show that applying the DEMATEL-based ANP for determining criteria weights and evaluating performance yields a stable converged weighted supermatrix and determines overall priorities. Through a simplified data collection process, combining the DEMATEL and ANP resolves the issue of interdependence among
Table 1 Description of experts. Organization
Code
Title
NRIs
NRI-1 NRI-2 NRI-3 NRI-4 NRI-5 NRI-6 NRI-7 NRI-8 NRI-9 NRI-10 PO-1 PO-2 PO-3 AC-1 AC-2
Deputy Center Director Deputy Division Director Deputy Division Director Department Manager Project Manager Principle Investigator Principle Investigator Principle Investigator Principle Investigator Principle Investigator Chairman CEO General Manager Professor Deputy Professor
For-profit organizations
Academia
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
3
Table 2 Finalized NSDB framework after the MDM survey. Dimension
Criteria
Description
Need (N)
New market potential (NMP) Potential size of market (PSM) Time to market (TTM) Urgent customer requirement (UCR) Ability to meet likely future regulations (MFR) Technical resource availability (TRA) Opportunity for technical success (OTS) Content of a technical plan (CTP)
Potential to meet the requirements of a new market. The potential size or growth of a market for products based on the proposed technology. The time from product conception to commercial sale. The urgency of customer demands. The extent to which a proposed technology coincides with science and technology policy. The degree to which a project has access to technical resources. The opportunity for success of a proposed technology. The project must be described in detail to answer questions on clear and concise planning, clear identification of the core technology, feasibility of the technical approach, and the major technical constraints. How advanced is the proposed technology compared with existing technology? How innovative is the proposed technology? The critical characteristics of a technology for product or industry development. The proprietary technology position through the collection of patents owned. The maximum possible return on an investment. The extent to which a proposed technology may further improve technological developments and competiveness based on the project outcomes. Benefits to society achieved through the improvement of national standards of quality, environmental protection, industrial safety, national image, and industrial standards (QESIS). The opportunity for the market success of a product based on a proposed technology.
Solution (S)
Differentiation (D)
Benefit (B)
Advancements in technology (AOT) Innovation of technology (IOT) Key of technology (KOT) Patent portfolios (PP) Potential return on investment (PROI) Aid an organization in competing in the market (AOC) Improvement on the QESIS (QESIS) Opportunity for market success (OMS)
dimensions and criteria in a complex hierarchical system. In practice, using this hybrid MCDM method is more rational than using standalone methods such as the analytic hierarchy process.
4. Empirical study: Case of R&D project portfolio selection by an NRI Fifteen experts (Table 1) completed an MDM survey and DEMATELbased ANP. Validating the unique strategic goal of NRIs involves assigning experts to either an experimental group (i.e., NRIs; n = 10) or a control group (i.e., for-profit organizations; n = 3), and testing the cognitive differences between these two groups of experts and a third group of academics (n = 2). A case study of R&D projects for equipment development in the flexible electronic industry involves evaluating and weighting the proposals in the project portfolio. The five R&D proposals are (1) a roll-to-roll touch-panel lamination system (R2R); (2) slot die wet-coating equipment (SD); (3) a plasma-enhanced chemical vapor deposit (PECVD); (4) a measuring instrument (MI); and (5) a laser-patterning machine (LPM). Evaluation of these alternatives took place using the hybrid MCDM method, and comparison took place using the NSDB model.
The research team described the issue at hand and provided experts with a summary of relevant research. Experts received the questionnaires in advance via email and responded in face-to-face interviews. Responses converged after five rounds of MDM. During the process, recommendation and evaluation of some criteria missing from the preliminary NSDB framework took place by ranking the criteria in order of importance. Table 2 summarizes the final criteria. Applying the DEMATEL method clarifies the interrelationships among the dimensions and criteria arising from the expert survey. Applying this process generates the total-influence matrix T (Table 3) and identifies the factors of concern (Table 4). This analysis yields a comprehensive impact relation map (Fig. 1). The DEMATEL analysis results enable the formulation of the unweighted supermatrix W, which is the principal eigenvector of the criteria pairwise comparison matrix relative to the NSDB dimensions. Obtaining the normalized total-influence matrix Ts enables the retrieval of the weighted supermatrix Ww. The last stage of the process yields the converged weighted supermatrix by raising the matrix to a largeenough power such that the weights converge and the matrix becomes a long-term stable supermatrix (Jeng & Bailey, 2012). Table 5 shows the weighted dimensions and criteria of the NSDB model.
Table 3 DEMATEL total-influence matrix T. Need (N)
N
S
D
B
NMP PSM TTM UCR MFR TRA OTS CTP AOT IOT KOT PP PROI AOC QESIS OMS
Solution (S)
Differentiation (D)
Benefit (B)
NMP
PSM
TTM
UCR
MFR
TRA
OTS
CTP
AOT
IOT
KOT
PP
PROI
AOC
QESIS
OMS
0.083 0.168 0.117 0.132 0.116 0.096 0.092 0.058 0.114 0.129 0.114 0.125 0.107 0.118 0.083 0.116
0.168 0.104 0.119 0.142 0.125 0.12 0.099 0.075 0.122 0.148 0.133 0.143 0.13 0.133 0.083 0.137
0.174 0.187 0.095 0.2 0.125 0.139 0.144 0.1 0.135 0.157 0.159 0.144 0.152 0.149 0.092 0.171
0.129 0.142 0.113 0.083 0.097 0.134 0.121 0.083 0.09 0.103 0.118 0.113 0.099 0.113 0.07 0.121
0.132 0.155 0.105 0.121 0.09 0.146 0.135 0.11 0.162 0.162 0.13 0.147 0.147 0.147 0.12 0.153
0.169 0.201 0.145 0.189 0.158 0.138 0.186 0.119 0.174 0.205 0.2 0.187 0.192 0.184 0.11 0.207
0.172 0.194 0.175 0.202 0.161 0.244 0.135 0.165 0.183 0.212 0.217 0.215 0.204 0.199 0.11 0.224
0.093 0.107 0.088 0.124 0.105 0.146 0.097 0.052 0.109 0.107 0.12 0.11 0.103 0.111 0.066 0.122
0.141 0.154 0.116 0.127 0.126 0.173 0.143 0.106 0.118 0.188 0.186 0.179 0.132 0.163 0.111 0.169
0.192 0.197 0.16 0.192 0.168 0.21 0.188 0.122 0.228 0.17 0.263 0.235 0.191 0.213 0.141 0.218
0.184 0.201 0.154 0.193 0.165 0.214 0.181 0.131 0.235 0.252 0.165 0.248 0.187 0.197 0.115 0.208
0.208 0.239 0.179 0.187 0.188 0.209 0.194 0.146 0.263 0.275 0.267 0.176 0.214 0.232 0.122 0.233
0.2 0.245 0.178 0.197 0.181 0.202 0.203 0.12 0.201 0.227 0.225 0.23 0.148 0.223 0.133 0.237
0.172 0.201 0.186 0.193 0.19 0.229 0.216 0.127 0.229 0.258 0.266 0.247 0.219 0.166 0.144 0.25
0.12 0.139 0.108 0.126 0.128 0.128 0.134 0.092 0.143 0.157 0.139 0.136 0.153 0.159 0.066 0.178
0.207 0.238 0.219 0.219 0.214 0.254 0.226 0.153 0.239 0.267 0.271 0.255 0.244 0.273 0.158 0.191
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
4
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
Table 4 Influence of the factors of concern.
Table 5 Weighted dimensions and criteria of the NSDB model.
Dimensions/Criteria
di
ri
di + ri
di − ri
Dimensions/Criteria
Need (N) NMP PSM TTM UCR Solution (S) MFR TRA OTS CTP Differentiation (D) AOT IOT KOT PP Benefit (B) PROI AOC QESIS OMS
0.644 2.544 2.872 2.257 2.627 0.586 2.337 2.782 2.494 1.759 0.727 2.745 3.017 2.973 2.890 0.629 2.622 2.78 1.724 2.935
0.488 1.768 1.981 2.323 1.729 0.600 2.162 2.764 3.012 1.660 0.736 2.332 3.088 3.030 3.332 0.761 3.15 3.293 2.106 3.628
1.131 4.312 4.853 4.580 4.356 1.186 4.499 5.546 5.506 3.419 1.463 5.077 6.105 6.003 6.222 1.390 5.772 6.073 3.830 6.563
0.156 0.776 0.891 −0.066 0.898 −0.014 0.175 0.018 −0.518 0.099 −0.010 0.413 −0.071 −0.057 −0.442 −0.132 −0.528 −0.513 −0.382 −0.693
Need (N) NMP PSM TTM UCR Solution (S) MFR TRA OTS CTP Differentiation (D) AOT IOT KOT PP Benefit (B) PROI AOC QESIS OMS
To verify the cognitive differences between NRIs and for-profit organizations, Table 6 shows the DEMATEL-based ANP results of the weighting and ranking. To evaluate the R&D projects, NRI managers evaluated the project alternatives by assigning them a score between 1 and 10. Table 7 shows the score regarding five alternatives with respect to the criteria in terms of the dimension of NSDB. The five project alternatives receive scores according to the NSDB model weights (Table 5).
Local weight (ranking) 0.1871 (4) 0.2265 (3) 0.2544 (2) 0.2956 (1) 0.2235 (4) 0.2325 (3) 0.2262 (3) 0.2884 (2) 0.3127 (1) 0.1727 (4) 0.2855 (2) 0.1996 (4) 0.2625 (2) 0.2567 (3) 0.2813 (1) 0.2949 (1) 0.2571 (3) 0.2716 (2) 0.1736 (4) 0.2976 (1)
Global weight (ranking) 0.0424 (14) 0.0476 (13) 0.0553 (10) 0.0418 (15) 0.0526 (11) 0.0670 (8) 0.0727 (7) 0.0401 (16) 0.0570 (9) 0.0749 (5) 0.0733 (6) 0.0803 (2) 0.0758 (4) 0.0801 (3) 0.0512 (12) 0.0878 (1)
5. Discussion 5.1. Results Table 4 shows the strength of influences among the dimensions and criteria according to the DEMATEL analysis results. Fig. 1 shows that among the NSDB model dimensions, the benefit dimension is the most important, whereas the need dimension is the least important.
Fig. 1. Comprehensive impact relation map for NSDB model.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
5
Table 6 Cognitive difference in the weighted criteria of the NSDB model between the experimental and control groups. Dimension/Criteria
Need (N)
Solution (S)
Differentiation (D)
Benefit (B)
NRI
New market potential (NMP) The potential size of market (PSM) Time to market (TTM) Urgent customer requirement(UCR) Ability to meet likely future regulations (MFR) Technical resource availability (TRA) Opportunity of technical success (OTS) Contents of technical plan (CTP) Advancement of technology (AOT) Innovation of technology (IOT) Key of technology (KOT) Patent portfolios (PP) Potential return on investment (PROI) Aid the organization in competing in the market (AOC) Improvement on QESIS (QESIS) Opportunity of market success (OMS)
In addition, results show that need is a causal factor of the other dimensions. Two of the top four degrees of importance are opportunity for market success (OMS) (6.563) and aiding the organization to compete in the market (AOC) (6.073), both of which belong to the benefit dimension. The other two criteria are patent portfolios (PP) (6.222) and innovation of technology (IOT) (6.105), both of which belong to the differentiation dimension. Two criteria with the lowest degrees of importance are contents of a technical plan (CTP) (3.419) and improvements in benefits to society through improving national standards of quality, environmental protection, industrial safety, national image, and industrial standards (QESIS) (3.830). Negative values of (di–ri) represent the role of receiver (Jeng & Bailey, 2012). In addition to advances in technology (AOT) (0.413), seven of the nine net receivers belong to the dimensions of differentiation and benefit. The other receivers are opportunity of technical success (OTS) (− 0.518) and time to market (TTM) (− 0.066), which belong to the dimensions of solution and need, respectively. Criteria belonging to the need dimension are all strong causal factors, except TTM (−0.066), which is a weak receiver. Although criteria belonging to the need dimension are the least important, they are highly influential on the other criteria.
For-profit organization
Weight
Ranking
Weight
Ranking
0.025 0.026 0.023 0.031 0.092 0.070 0.085 0.051 0.052 0.061 0.062 0.106 0.074 0.093 0.053 0.096
15 14 16 13 4 7 5 12 11 9 8 1 6 3 10 2
0.019 0.018 0.038 0.031 0.078 0.061 0.083 0.076 0.052 0.073 0.071 0.086 0.085 0.094 0.042 0.093
15 16 13 14 6 10 5 7 11 8 9 3 4 1 12 2
The weighted NSDB model results show that OMS plays the most crucial role (weight = 0.088), followed by PP (weight = 0.080), AOC (weight = 0.080), and potential return on investment (PROI) (weight = 0.076). Three of the four highest priorities belong to the benefit dimension. This result implies that the benefit dimension is the most important when organizations decide to invest in an R&D project. Moreover, the weights of the NSDB model show that the lowest four criteria are CTP, urgent customer requirement (UCR), new market potential (NMP), and the potential size of market (PSM) (weights = 0.040, 0.042, 0.042, and 0.048, respectively). Three of these four criteria belong to the need dimension, indicating that this dimension has the least influence when NRIs consider whether to invest in an R&D project. In addition, CTP has the lowest weight, implying that the content of an R&D proposal is less relevant than other factors are as long as the proposal's concept is innovative and can provide benefits. According the result of DEMATEL-based ANP (Table 7), the assessment committee prioritizes the five R&D projects as follows: PECVD ≻ LPM ≻ R2R ≻ MI ≻ SD. Comparing the criteria-weighting results between the NRIs and forprofit organizations reveals differences between these two groups. The for-profit organizations rank TTM as the most important criterion
Table 7 Priority and comparison of the R&D project alternatives. Dimension/ Criteria
N
NMP PSM TTM UCR S MFR TRA OTS CTP D AOT IOT KOT PP B PROI AOC QESIS OMS Total Score Priority
Alternatives LPM
PECVD
R2R
SD
MI
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
Direct score
Weighted score
6.5 5.5 8 6.5 5.75 7.5 7 7 7.5 7.25 7 6.75 5.75 6.75 5.75 6.25
0.276 0.262 0.442 0.272 0.302 0.503 0.509 0.281 0.427 0.543 0.513 0.542 0.436 0.541 0.294 0.548 6.692 2
7.75 8.25 8.25 7.75 5.75 6.75 7.75 7 7.75 7.25 8 7.25 8.5 8.25 6 7.75
0.328 0.393 0.456 0.324 0.302 0.453 0.563 0.281 0.442 0.543 0.586 0.582 0.644 0.661 0.307 0.680 7.547 1
7.5 6.75 7 8.5 5.75 6.75 7.75 5.25 5.25 5.75 5.75 5.5 6.25 7.25 6 8.5
0.318 0.321 0.387 0.355 0.302 0.453 0.563 0.211 0.299 0.431 0.421 0.442 0.474 0.581 0.307 0.746 6.612 3
6 6.75 7.5 6.25 5.75 5.75 5.75 5.25 6.75 6.75 5.5 5.5 7.5 7.75 6 6.25
0.254 0.321 0.415 0.261 0.302 0.386 0.418 0.211 0.385 0.506 0.403 0.442 0.569 0.621 0.307 0.548 6.349 5
6.25 5.5 7.75 6.75 6.25 7.5 6.75 6.5 6.5 6 6.25 6.5 5.75 7 6 7.25
0.265 0.262 0.429 0.282 0.329 0.503 0.491 0.261 0.370 0.450 0.458 0.522 0.436 0.561 0.307 0.636 6.561 4
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
6
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
industries to compete, thereby improving the national economy. Given this consideration, the portfolio selection criteria and its importance differ from those of for-profit enterprises. This study shows that the portfolio selection process of NRIs tends to involve selecting unique and differentiated R&D alternatives. Examining the government-assigned key performance indicators of NRIs (e.g., number of technical reports, number of patents filed, scale of technology transfer, industrial investment, capability to support other projects or derive new projects, and promoting people's skill or employment) shows that such indicators relate to the benefit dimension. This research reveals that QESIS is less important. NRIs, which assist governments with economic or social development, care less about social welfare than about benefit, implying that NRIs should enhance their social responsibility when considering technological developments.
(Rank 13) in the need dimension, whereas the NRIs rank TTM as the least important criterion (Rank 16). This difference implies NRIs and for-profit organizations have different priorities for meeting market opportunities. Regarding the solution dimension, the for-profit organizations and NRIs rank the priority of technical resource availability (TRA) as 10th and 7th, respectively. In contrast, the for-profit organizations consider CTP to be more crucial (Rank 7) than the NRIs do (Rank 12), implying that for-profit organizations care more about technical planning from the perspective of internal support, whereas the NRIs care more about resource availability for innovative development. For the differentiation dimension, both types of organization assign relatively equal importance to AOT (Rank 11), IOT (Rank 9 and 8), key of technology (KOT) (Rank 8 and 9). NRIs and for-profit organizations rank patent portfolios (PP) 1st and 3rd, respectively, indicating differences between the two types of organization. NRIs pursue national technology development, whereas for-profit organizations focus on competing in the market. Regarding the criteria in the benefit dimension, managers of NRIs and mangers of for-profit organizations rank PROI 6th and 4th, respectively. NRIs and for-profit organizations rank AOC 3rd and 1st, respectively, indicating that competitive strength is the most important criterion for market survival according to for-profit organizations. Both groups rank the criterion of OMS 2nd, implying that market success remains the major goal of R&D activities in both types of organization. The NRIs, however, rank improvement in QESIS higher than for-profit organizations do because NRIs undertake government tasks relating to social welfare.
6. Conclusions The present study employs a hybrid MCDM method to overcome complexity when exploring interdependence and feedback between criteria and alternatives. The proposed NSDB model is a systematic approach for R&D project portfolio selection. This empirical study demonstrates the effectiveness of applying the hybrid MCDM method within the NSDB framework. The issues of business, competitiveness, and technological development act as factors to evaluate R&D alternatives at the early stage of a project. This study is an in-depth exploration of the characteristics of NRIs. Unlike previous studies, the present study focuses primarily on forprofit enterprises. The findings show that private enterprises conduct R&D for the sake of long-term sustainability, whereas NRIs promote industrial development through innovative technology R&D. This research discusses the methods that NRIs employ when they select R&D projects and weight the importance of decision-making criteria. Future research should consider investigating effective mechanisms for project collaboration between NRIs and for-profit organizations to maximize benefits for both parties.
5.2. Implications Innovation and IP drive most R&D-oriented NRI business models. These drivers enable the transfer of technologies to industry, helping firms improve their products or services and compete in the market. The goal of NRIs is to develop new technologies that assist local Appendix 1 Literature review of the relevant criteria for project selection
Criteria Benefit/cost ratio Cost Capital investment Rate of return Contribution of profitability Growth rate Payback period Direct cash flow Effect on existing market share Effect on existing market outlook New market potential Potential return on investment Net present value Contribution to organizational goals/objectives Aid the organization in competing in the market Improvements on research capability Internal political decisions Strategic fit Importance to the organization for the future success Contribution to knowledge Importance to the functioning of the organization Public relation effect Importance to organization's critical success factors Fit with company business plan Ability to meet likely future regulations External regulations Staff training, development and job satisfaction
Huang et al. (2008)
Lawson et al. (2006)
Meade and Presley (2002)
Jiang and Klein (1999)
Stewart (1991)
Liberatore (1988)
v v v v v v v v v v v v
v v v v v
v
v v v v v v v v v v v v
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016
D.J.-F. Jeng, K.-H. Huang / Journal of Business Research xxx (2015) xxx–xxx
7
Appendix 1 (continued) Criteria Required by regulations Coincidence with S&T policy Response to competition Required by customers/suppliers QESIS (1)Improvements on QESIS (2)Workplace safety (3)Environmental considerations (4)New industry standards (5)Environmental and safety consideration lawsuit requires information The potential size of market Technology spillover effects Benefits for human life Technical resource availability Technical support Equipment support Timing for project Product range growth potential Product life cycle Synergy with other products/ processes Ability to implement production/process Patentability/design protection Existence of project champion Existence of required competence Number and strength of competitors Isolated, simple, and modular project High visibility of project Basic subsystem to system Basic module for operations Availability of skilled IS personnel Availability of needed technology Advancement of technology Innovation of technology Key of technology Proprietary technology Generics of technology Technological connections Technological extendibility Specification of technology Appropriateness for research period Appropriateness for research cost Evidence of scientific feasibility
Huang et al. (2008)
Lawson et al. (2006)
Meade and Presley (2002)
Jiang and Klein (1999)
Stewart (1991)
v
v
v v
v v
v(4)
v
Liberatore (1988)
v
v(1)(5)
v(2)(3)
v v v v v v v
v
v
v
v v v
v
v v v v v
v v v v
v v v v v v v v v v v v v v v
References Carlsson, C., Fullér, R., Heikkilä, M., & Majlender, P. (2007). A fuzzy approach to R&D project portfolio selection. International Journal of Approximate Reasoning, 44(2), 93–105. Coffin, M. A., & Taylor, B. W. (1996). Multiple criteria R&D project selection and scheduling using fuzzy logic. Computers & Operations Research, 23(3), 207–220. Custer, R. L., Scarcella, J. A., & Stewart, B. R. (1999). The modified Delphi technique—A rotational modification. Journal of Vocational and Technical Education, 15(2), 50–58. Dey, P. K. (2006). Integrated project evaluation and selection using multiple attribute decision making technique. International Journal of Production Economics, 103(1), 90–103. Doctor, R. N., Newton, D. P., & Pearson, A. (2001). Managing uncertainty in research and development. Technovation, 21(2), 79–90. Dutta, P., Lach, S., & Rustichini, A. (1990). Innovation and product differentiation. Unpublished Discussion Paper 894. Northwestern University, Center for Mathematical Studies in Economics and Management Science. Fenwick, D., Daim, T. U., & Gerdsri, N. (2009). Value Driven Technology Road Mapping (VTRM) process integrating decision making and marketing tools: Case of Internet security technologies. Technological Forecasting and Social Change, 76(8), 1055–1077. Fontela, E., & Gabus, A. (1976). The DEMATEL observer, DEMATEL 1976 Report. Geneva, Switzerland: Battelle Geneva Research Center. Fox, E., Baker, N., & Bryant, J. (1984). Economic models for R&D project selection in the presence of project interactions. Management Science, 30(7), 890–902. Gabus, A., & Fontela, E. (1973). Perceptions of the world problematique: Communication procedure, communicating with those bearing collective responsibility. DEMATEL Report No. 1. Geneva, Switzerland: Battelle Geneva Research Center. Hall, D., & Nauda, A. (1990). An interactive approach for selecting IR&D projects. IEEE Transactions on Engineering Management, 37(2), 126–133. Hsu, Y. -G., Tzeng, G. -H., & Shy, J. Z. (2003). Fuzzy multiple criteria selection of government-sponsored frontier technology R&D projects. R&D Management, 33(5), 539–551. Huang, C. -C., Chu, P. -Y., & Chiang, Y. -H. (2008). A fuzzy AHP application in governmentsponsored R&D project selection. OMEGA: The International Journal of Management Science, 36(6), 1038–1052.
v v
Jeng, D. J. -F. (2015). Generating a causal model of supply chain collaboration using the fuzzy DEMATEL technique. Computers & Industrial Engineering, 87, 283–295. Jeng, D. J. -F., & Bailey, T. (2012). Assessing customer retention strategies in mobile telecommunications: A hybrid MCDM approach. Management Decision, 50(9), 1570–1595. Jiang, J. J., & Klein, G. (1999). Project selection criteria by strategic orientation. Information & Management, 36(2), 63–75. Kotler, P., Armstrong, G., Ang, S. H., Leong, S. M., Tan, C. T., & Hon-Ming, O. Y. (2009). Principles of marketing: A global perspective. Cincinnati, Ohio: Pearson Education Inc. Lawson, C. P., Longhurst, P. J., & Ivey, P. C. (2006). The application of a new research and development project selection model in SMEs. Technovation, 26, 242–250. Lee, J., Veloso, F. M., Hounshell, D. A., & Rubin, E. S. (2010). Forcing technological change: A case of automobile emissions control technology development in the US. Technovation, 30(4), 249–264. Liberatore, M. J. (1988). Applications of the AHP. Paper presented at the Proc. 6th Int. Conf. on Mathematical Modelling. Lin, C., & Hsieh, P. -J. (2004). A fuzzy decision support system for strategic portfolio management. Decision Support Systems, 38(3), 383–398. Matsubayashi, N. (2007). Price and quality competition: The effect of differentiation and vertical integration. European Journal of Operational Research, 180(2), 907–921. Meade, L., & Presley, A. (2002). R&D project selection using the analytic network process. IEEE Transactions on Engineering Management, 49(1), 59–66. Porter, M. E. (1980). Competitive strategy: Techniques for analyzing industries and competitors, Vol. Chapter 2, New York: The Free Press. Raz, T., Shenhar, A. J., & Dvir, D. (2002). Risk management, project success, and technological uncertainty. R&D Management, 32(2), 101–109. Saaty, T. L. (Ed.). (1996). Decision making with dependence and feedback: Analytic network process. Pittsburgh: RWS Publications. Souder, W. E. (1986). Achieving organizational consensus with respect to R&D project selection. Management Science, 21(6), 669–681. Stewart, T. J. (1991). A multi-criteria decision support system for R&D project selection. Journal of the Operational Research Society, 42(1), 17–26. Wang, J., Lin, W., & Huang, Y. -H. (2010). A performance-oriented risk management framework for innovative R&D projects. Technovation, 30(11–12), 601–611.
Please cite this article as: Jeng, D.J.-F., & Huang, K.-H., Strategic project portfolio selection for national research institutes, Journal of Business Research (2015), http://dx.doi.org/10.1016/j.jbusres.2015.06.016