Conference PaperPDF Available

Measuring Innovation Capability in German ICTcompanies by using DEA-Models

Authors:
  • University of Applied Sciences Neu-Ulm

Abstract and Figures

In this paper an approach to measure the innovation capability of companies is developed. Companies often have difficulties in measuring the efficiency of the generation of innovations due to the complexity of the innovation process itself and the variety of potential input and output indicators. Based on a preliminary study among 21 German companies from the ICT-sector an Input-Output-Model has been developed and tested. As a result the main input and output indicators with significant correlation to the innovation efficiency have been identified. The iTOP Innovation Capability Assessment (iTOP ICA) consists of a quantitative measurement based on Data Envelopment Analysis (DEA) and enables companies to measure their efficiency and effectiveness in generating innovations in an impartial way. This is primarily important for companies, operating in technology intensive business fields, to benchmark their Return on Innovation with further peer companies.
Content may be subject to copyright.
0
The International Society for Professional Innovation
Management (ISPIM)
Proceedings of the 5th ISPIM Innovation Symposium:
"Stimulating Innovation: Challenges for Management,
Science & Technology
09-12- December; 2012
Seoul, Korea.
Measuring Innovation Capability in German ICT-
companies by using DEA-Models
Julian V. Kauffeldt
Leo Brecht
Daniel R. A. Schallmo
Kirill Welz
Measuring Innovation Capability in German ICT-
companies by using DEA-Models
Julian V. Kauffeldt*
University of Ulm, Institute of Technology and Process Management,
Helmholtzstrasse 22, 89081 Ulm, Germany.
E-mail: Julian.Kauffeldt@uni-ulm.de
Leo Brecht
University of Ulm, Institute of Technology and Process Management,
Helmholtzstrasse 22, 89081 Ulm, Germany.
E-mail: Leo.Brecht@uni-ulm.de
Daniel R.A. Schallmo
University of Ulm, Institute of Technology and Process Management,
Helmholtzstrasse 22, 89081 Ulm, Germany.
E-mail: Daniel.Schallmo@uni-ulm.de
Kirill Welz
University of Ulm, Institute of Technology and Process Management,
Helmholtzstrasse 22, 89081 Ulm, Germany.
E-mail: Kirill.Welz@uni-ulm.de
* Corresponding author
Abstract: In this paper an approach to measure the innovation capability of
companies is developed. Companies often have difficulties in measuring the
efficiency of the generation of innovations due to the complexity of the
innovation process itself and the variety of potential input and output
indicators.
Based on a preliminary study among 21 German companies from the ICT-
sector an Input-Output-Model has been developed and tested. As a result the
main input and output indicators with significant correlation to the innovation
efficiency have been identified.
The iTOP Innovation Capability Assessment (iTOP ICA) consists of a
quantitative measurement based on Data Envelopment Analysis (DEA) and
enables companies to measure their efficiency and effectiveness in generating
innovations in an impartial way. This is primarily important for companies,
operating in technology intensive business fields, to benchmark their Return on
Innovation with further peer companies.
Keywords: Innovation Index; Innovation Measurement; Innovation Process;
Data Envelopment Analysis; Return on Innovation.
This paper was presented at The 5th ISPIM Innovation Symposium - Stimulating Innovation:
Challenges for Management, Science & Technology, Seoul, Korea on 9-12 December 2012. The
publication is available to ISPIM members at www.ispim.org.
1 Introduction
The capability to generate innovations is likely to be one of the most important indicators
for the future success of companies (Bürgin, 2007; Hausschildt and Salomo, 2007).
A significant positive correlation between the capability of companies to generate
innovations and the resulting economical success of the company has been shown by a
large number of empirical studies (Gerpott, 2005; Hauschildt and Salomo, 2007; Bürgin,
2007). Therefore more and more companies have identified the generation of innovations
as a key element of the corporate strategy. In view of an intensifying competition on
global markets and growing business challenges the focus on the reduction of innovation
cycles and improvement of the innovation capability is not a luxury but rather
indispensible requirement for economical success.
In order to distinguish between a well structured innovation strategy and simple paper
exercises a systematic measurement of the innovation capability is essential. The
evaluation of this innovation capability, which is measured as the efficiency of
innovation generation, is crucial to define an appropriate innovation strategy, allocate
budgets and coordinate technology and innovation projects.
In this research work we measure the efficiency of the generation of innovations by using
Data Envelopment Analysis and combine this with an analysis of the Technology and
Innovation Management processes. We define the resulting key figure Return on
Innovation as follows,
Return on Innovation=



=Outputs
Inputs
where  describes the amount of inputs deployed in the innovation process and 
describes the amount of realized outputs for one single company
i
. The aggregation of
inputs and outputs is done by using the weighting factors  (outputs) and  (inputs)
(Cantner et al., 2007; Drake et al., 2006). We use the realized company specific Return of
Innovation to determine the innovation capability of the analyzed companies.
2 Problem
The evaluation of the efficiency of innovation generation becomes difficult when there
exist multiple inputs and outputs, where the main difficulty is the determination of
benchmarks (Zhu, 2003). The knowledge of a specific production function with certain
assumptions about the error terms and other restrictions, as it is needed for most of the
existing parametric approaches, is normally not given in innovation management.
Therefore the most used approaches remain on a qualitative level consisting of innovation
audits or innovation scorecards.
The majority of the semi quantitative approaches uses the amount of R&D-expenditures
and the number of R&D employees as the main input indicators of the innovation process
due to their high level of availability (Gerpott, 2005; Santarelli and Piergiovanni, 1996;
Kotzbauer, 1992). However, this ignores the considerable importance of further items
such as innovation culture and process structure, which have been identified to be
important by several publications (Jaruzelski et al., 2011). Furthermore, the isolated
usage of R&D expenditures and number of R&D-employees leads to an underestimation
of small- and medium-sized enterprises (SMEs), where the research and development
activities are often combined with other activities and carried out without the existence of
a formal R&D budget (Flor and Oltra, 2004). Despite of the technical difficulties in
measuring the input of the generation of innovations by using quantitative indicators, the
measurement of the innovation output becomes even more complicated. Several
innovation output indicators have been developed, whereas the number of patents is the
most common one (Lanjouw and Schankermann, 2002; Griliches, 1990; Patel and Pavitt,
1992). The reason for the high usage of patents as indicators for the innovation process is
the ease of access to patent data. Nevertheless, using patent data as an indicator for
innovation output has several limitations. In this context it should be stressed that patents
are a reflection of inventions rather than innovations (Flor and Oltra, 2004). A single
patent provides no information about the commercial success of the associated product or
service. Furthermore patents are characterized by a huge heterogeneity, which means that
the technological level of the patents as well as the economic value show great variety.
Even the companies’ strategy concerning patent application varies between different
branches and countries. Some companies tend to apply for every single invention,
whereas other companies compile several inventions in one patent application or do not
apply for a patent at all.
It can be seen that the isolated usage of just one indicator to evaluate the innovation
capability tends to remain problematic. A more comprehensive Input-Output-Model is
needed to evaluate the innovation capability.
3 Current understanding
Several studies have shown a connection between the economic success of a company
and its capability to generate innovations, e.g. new products or services (Gunday, 2011;
Stefik, 2004; Wuyts, 2004; Grulke, 2002).
The literature provides several approaches to measure certain elements of the innovation
process in companies by using patent data, R&D expenditures and qualitative methods in
general e.g. innovation audits or innovation scorecards (Rosenbusch, 2011; Flor, 2004;
Lanjouw and Schankermann, 2002; Sterlacchini, 1998; Santarelli and Piergiovanni,
1996).
It is recognized that the measurement of the innovation process is affected by a high
complexity and several problems when trying to apply traditional methods and
techniques of evaluation e.g. using patent data for evaluation of the success of innovation
generation disregarding that a patent does not always lead to an innovative product.
Furthermore, this does not say anything about the commercial success of a related
product and disregards the return on innovation. Most of the existing approaches depend
on an innovation audit consisting of multiple questions or require the knowledge of a
specific production function e.g. traditional cost-benefit-analysis (Li, 2002). This shows
that the data used for the evaluation is often characterized by a high level of subjectivity.
The Data Envelopment Analysis (DEA) relies on a purely quantitative data basis and has
been deployed successfully to answer a multitude of scientific questions, e.g. efficiency
of healthcare institutions (Nunamaker, 1985) or analysis of commercial banks (Berg et
This paper was presented at The 5th ISPIM Innovation Symposium - Stimulating Innovation:
Challenges for Management, Science & Technology, Seoul, Korea on 9-12 December 2012. The
publication is available to ISPIM members at www.ispim.org.
al., 1993). This includes the study of innovation systems at the national level in order to
revise country-specific policies for creating a positive national innovation climate
(Abbasi et al., 2011).
As mentioned above the measurement of innovation capability has gained large attention
in literature. Nevertheless an extensive approach to combine a purely quantitative
measurement of innovation capability on company level with a qualitative analysis of the
innovation process is missing.
4 Research questions
Based on the problems described and current understanding we answer the following
main research questions:
How does a generic Input-Output-Model on innovation generation look like?
How can the Return on Innovation be measured in an objective way?
How can the most important levers be identified, which can help to optimize the
innovation capability of a company?
5 Methodology and Approach
We address these questions by developing an index of innovative companies, based on
DEA-Models, which we refer to as “Innovation Capability Index (ICI)”. This ICI
measures the Return on Innovation based on a comprehensive Input-Output-Model. We
use the Data Envelopment Analysis as analytical basis to evaluate the efficiency of the
company specific generation of innovations. The Data Envelopment Analysis allows the
measurement of the efficiency without specifying the production function (Cooper et al.,
2007; Cantner et al., 2007). Therefore this approach enables us to find a best practice
benchmark for each analysed company on the basis of an extensive Input-Output-Model.
Moreover, this measurement of the Return on Innovation helps to identify levers
consisting of input and output slacks in order to improve the innovation capability in a
company.
The Data Envelopment Analysis (DEA) which has been developed by Charnes, Cooper
and Rhodes (1978) identifies, on the basis of empirical data, the specific maximum
achievable efficiency score for each considered decision making unit (DMU). By doing
this best practice companies that serve as peers for the remaining inefficient companies
are identified. The DMUs can be defined as a company, a business unit or a whole
country.
Figure 1: Procedure of Model Construction
In the framework of the DEA-approach the first step is to measure the efficiency of each
company considered as the ratio of the aggregated inputs and outputs of the innovation
process. The relevant input and output indicators of the innovation process have been
identified using an extensive literature review. The procedure of constructing the Input-
Output-Model is visualized in figure 1.
The input- and output- model as it has been used in the research work is presented in
figure 2 listed below.
The DEA maximizes the efficiency measure of each company by solving a maximization
problem using common methods from Operations Research (basically Charnes-Cooper-
Transformation, Simplex Algorithm). In doing so for each company weighting factors are
determined for every single input and output indicator. These input and output indicators
remain purely quantitative. Therefore the calculated efficiency score for each company in
Figure 2: Input-Output-Model of Innovation Generation
This paper was presented at The 5th ISPIM Innovation Symposium - Stimulating Innovation:
Challenges for Management, Science & Technology, Seoul, Korea on 9-12 December 2012. The
publication is available to ISPIM members at www.ispim.org.
the sample is characterized by a maximum of objectivity. One main advantage of the
DEA approach is that a target is identified for each inefficient company. Based on the
calculated efficiency scores in the next step the DEA-approach constructs an efficient
frontier which represents the best-practice observations. We use these efficiency scores to
build the iTOP Innovation Capability Index, which ranks the analysed companies by their
calculated Return on Innovation.
Furthermore the DEA-approach enables us to identify company specific input and output
slacks. These slacks are company specific inefficiencies of single inputs or outputs. In
other words: If a company shows slacks in regard to one input indicator this means that
the company is indicated by an overflow of this input indicator compared to the other
companies of the sample.
Therefore the Innovation Capability Index can be used as a benchmarking tool to
compare the Return on Innovation of certain companies with regard to their ability to
create innovative products and services.
In our research we applied literature review, focus group interviews and carried out a
preliminary study. The development and foundation of the approach is made by literature
review and analysis (Kromrey, 2007). The examination of practice is done via data
analysis of the preliminary study and focus group interviews. Focus group interviews
were embedded in three workshops from March to July 2012, including eight to twelve
practitioners from nine different companies with different fields of expertise. The
duration of a workshop was one day and at least two academics were permanently
involved.
Although integrating focus group interviews and literature review the generalization of
the research to ICT-companies in further countries remains a challenge. Therefore an
international study named iTOP Innovation Capability Assessment 2013is planned in
order to apply the developed approach.
6 Findings
In the framework of our research work we developed an approach to measure the Return
on Innovation by applying DEA-models with the objective of constructing an Innovation
Capability Index. We have been able to show how the model is developed and the
underlying input and output indicators of the innovation process. The approach has been
tested based on a preliminary study under participation of 21 German ICT-companies. By
usage of a correlation analysis we have identified the main input- and output indicators
with a significant positive correlation to the calculated Return on Innovation which
represents the position in the Innovation Capability Index.
In the table 1 listed above eight relevant input indicators and two relevant output
indicators are presented.
As expected against the background of former research studies, financial expenditures for
Research & Development and in particular for the Development of New Products
(leaving the established technological paths, completely new product- and service
categories) show a significant positive correlation to the calculated Return on Innovation.
This emphasizes the importance of R&D-Expenditures as one relevant input indicator of
the generation of innovations.
Furthermore the input classes Knowledge Creation and Cooperations show a high
relevance as input indicators due to the significant positive correlation of expenditures for
education and training, the degree of the employees’ professional experience and the
companies’ cooperations in general to the calculated Return on Innovation.
As relevant output indicators we have identified the profit generated by new products and
the realized savings on development costs.
It can be seen that especially indicators that are related to Intellectual Property
Management do not show a significant positive correlation to the calculated Return on
Innovation.
7 Contribution
The contribution of this research is an Input-Output-Model for innovation measurement
which leads, in combination with Data Envelopment Analysis, to an index of innovative
companies. The Input-Output-Model consists of four input categories (R&D-Investments,
Knowledge Creation, Innovation Culture, Cooperations) and three output-categories
(Commercialisation, Intellectual Property and Process Changes).
We have tested several DEA-models in order to measure the Return on Innovation and
benchmark companies on the basis of these results.
Moreover we seek to provide an explanation for the efficiency differences by analysing
the innovation processes of the companies considered.
This closes an existing gap of the current research on methods to measure the innovation
capability of companies.
Input- and Output-Indicators
R&D Expenditures significance level (two sided)
R&D Expenditures for New Products
significance level (two sided) 0,562**
0,008
Expenditures for Education and Training
significance level (two sided)
0,005
EmployeesProfessional Experience
significance level (two sided) 0,612**
0,003
Acquisition of External Knowledge
significance level (two sided)
0,048
Cooperations, upstream Value Chain
significance level (two sided)
Cooperations, same level of Value Chain
significance level (two sided)
0,041
Cooperations, downstream Value Chain
significance level (two sided)
Profit with New Products
significance level (two sided)
0,009
Savings on Development Costs
significance level (two sided)
Table 1: Relevant Input- and Output-Indicators
*= significant on the 0.05 level, **= significant on the 0,01 level
This paper was presented at The 5th ISPIM Innovation Symposium - Stimulating Innovation:
Challenges for Management, Science & Technology, Seoul, Korea on 9-12 December 2012. The
publication is available to ISPIM members at www.ispim.org.
8 Practical implications
Managers and business developers will gain advantage from the findings by having a
specific benchmarking tool that enables companies to identify key levers to improve the
company’s capability to generate innovations. These levers consist of input and output
slacks which are identified within the Data Envelopment Analysis. By increasing the
innovative outcome companies can be helped to differentiate stronger from competition
and enhance the outcome e.g. EBIT or market share.
With some minor adjustments concerning the Input-Output-Model it is possible to carry
out internal benchmarks based on the developed approach e.g. to compare internally the
Return on Innovation of business units or branch offices.
9 Limitations
Regarding the validity of the research work some limitations are existing. As the
underlying data for the calculations we have used submissions of 21 German companies
from the ICT-sector. Therefore the research work can only be seen as a preliminary study
in order to prepare the next international iTOP Innovation Capability Assessment in
2013. Nevertheless the data basis is sufficient in order to identify relevant input- and
output- indicators and to set up an appropriate DEA model.
A further limitation is the consideration of just one single year 2011 during the collection
of data. Yet this is a common approach in the literature to reduce complexity.
Besides, it is likely that manager of companies that show poor results in the framework of
the Innovation Capability Index may argue that owing to the external circumstances a
better result for the specific company would have been impossible. Indeed, the analysis
of external drivers has shown a significant negative correlation between the imitation of
products by competitors and the Return on Innovation of the specific company. Further
external drivers like fast changing of technologies, markets and customer behaviour as
well as legal restrictions have not shown a significant correlation to the Return on
Innovation.
10 Further Research
It makes sense to overcome the limitations of the German ICT-sector by generalizing the
approach to further sectors and countries within the framework of further studies.
Furthermore, it is important to determine the time lag between the allocation of inputs
and the realisation of the associated outputs. A first attempt to measure this time lag
carried out by the Institute of technology and Process Management at the University of
Ulm based on the analysis of 17 technology companies listed in the German Tech DAX
index indicates that the time lag ranges between three and four years. As soon as further
studies provide a resilient data basis collected over a period of several years it will be
possible to analyse the changes in innovation efficiency over time and to identify
associated reasons.
Furthermore, the influence of the economic situation should be analysed in the context of
the calculation of the Return on Innovation. The impact of structural and organisational
changes, e.g. in the wake of mergers and acquisitions, on the innovation efficiency has
not yet been taken into account. This should be done in further research projects.
The level of uncertainty of the underlying data and its impact on the calculated Return on
Innovation should be quantified and analysed. This could be done using of fuzzy-logic
methods.
Moreover, trends in the Return on Innovation over time are to be forecasted based on
developments in the past and the current situation.
11 Areas for Feedback and Development
Feedback is appreciated in the field of the generic Input-Output-Model which shall be
used in the iTOP Innovation Capability Assessment 2013. In this context the identified
main input and output indicators should be complemented and validated by the research
community.
This paper was presented at The 5th ISPIM Innovation Symposium - Stimulating Innovation:
Challenges for Management, Science & Technology, Seoul, Korea on 9-12 December 2012. The
publication is available to ISPIM members at www.ispim.org.
12 References and Notes
Abbasi, Farhad/ Hajihoseini, Hojatolah and Haukka, Sandra (2011), Use of virtual index
for measuring efficiency of innovation systems: A cross-country study, International
Journal of Technology Management & Sustainable Development, Volume 9, Number 3,
pp. 195-212.
Berg, Sigbjorn Atle/ Forsund, Finn R./ Hjalmarsson, Lennart and Suominen, Matti
(1993), Banking Efficiency in the Nordic Countries, Journal of Banking and Finance,
Volume 17, pp.74-84.
Bürgin, Christian (2007), Reifegradmodell zur Kontrolle des Innovationssystems von
Unternehmen, Dissertation, Zürich.
Cantner, Uwe/ Krüger, Jens and Hanusch, Horst (2007), Produktivitäts- und
Effizienzanalyse: Der nichtparametrische Ansatz, Springer-Lehrbuch, Springer-Verlag,
Berlin Heidelberg.
Cooper, William W./ Seiford, Lawrence M. and Tone, Kaoru (2007), Data envelopment
analysis: A comprehensive text with models, applications, references and DEA-solver
software, 2nd edition, Springer, New York.
Drake, Miles P./ Sakkab, Nabil/ Jonash, Ronald (2006), Maximizing Return on
Innovation Investment, Research Technology Management, Volume 49, Number 6, pp.
32-41.
Flor, Marisa L. and Oltra, Maria (2004), Identification of innovating firms through
technological innovation indicators. An application to the Spanish ceramic tile industry,
Research policy, Volume 33, Number 2, pp. 323–336.
Gerpott, Torsten J. (2005), Strategisches Technologie- und Innovationsmanagement,
Sammlung Poeschel, 2nd edition, Schäffer-Poeschel, Stuttgart.
Griliches, Zvi (1990), Patent statistics as economic indicators: A survey, Journal of
Economic Literature 28, pp. 1661-1707.
Grulke, Wolfgang and Silber, Gus (2002), Lessons in radical innovation: Out of the box -
straight to the bottom line, International ed., Financial Times Prentice Hall, London.
Gunday, Gurhan/ Ulusoy, Gunduz/ Kilic, Kemal and Alpkan, Lutfihak (2011), Effects of
innovation types on firm performance, Journal of Production Economics, Volume 133,
Number 2, pp. 662–676.
Hauschildt, Jürgen and Salomo, Sören (2007), Innovationsmanagement, Vahlens
Handbücher der Wirtschafts- und Sozialwissenschaften, 4th edition, Vahlen, München.
Jaruzelski, Barry/ Loehe, John/ Holman, Richard/ Baker, Edward H. and Johnson, Marc
(2011), The 2011 Global Innovation: Why culture is key, Booz & Co.
Kotzbauer, Norbert (1992), Erfolgsfaktoren neuer Produkte, Univ, Frankfurt am Main,
Regensburg, 1992.
Kromrey, Helmut (2007), Empirische Sozialforschung: Modelle und Methoden der
standardisierten Datenerhebung und Datenausweitung: Modelle und Methoden der
Datenerhebung und Datenauswertung, UTB, Stuttgart.
Lanjouw, Jean O. and Schankerman, Mark (2002), Research productivity and patent
quality: Measurement with multiple indicators, Centre for Economic Policy Research,
London.
Li, Ling X. (2002), An analysis of sources of competitiveness and performance of
Chinese manufacturer, International Journal of Operations & Production Management,
Volume 20, Number 3, pp. 299-315.
Nunamaker, Thomas (1985), Using Data Envelopment Analysis to Measure the
Efficiency of Non-profit Organizations: A Critical Evaluation, Managerial and Decision
Economics, Volume 6, pp. 50-58.
Patel, Pari and Pavitt, Keith (1992), The innovative performance of the world’s largest
firms: Some new evidence, Economics of Innovation and New Technology 2, pp. 91-102.
Rosenbusch, Nina/ Brinckmann, Jan and Bausch, Andreas (2011), Is innovation always
beneficial? A meta-analysis of the relationship between innovation and performance in
SMEs, Journal of business venturing, Volume 26, Number 4, pp. 441–457.
Santarelli, Enrico and Piergiovanni, Roberta (1996), “Analyzing literature-based
innovation output indicatorpp. The Italian experience”, Research policy, Volume 25,
Number 5, pp. 689–711.
Stefik, Mark and Stefik, Barbara (2004), Breakthrough: Stories and strategies of radical
innovation, MIT Press, Cambridge.
Sterlacchini, Alessandro (1998), Inputs and outputs of innovative activities in Italian
manufacturing, Economics of innovation and new technology, Volume 7, Number 4, pp.
323344.
Wuyts, Stefan H./ Dutta, Shantanu and Stremersch, Stefan (2004), Portfolios of interfirm
agreements in technology-intensive markets: Consequences for innovation and
profitability, Journal of Marketing Research, Volume 68, Number 2, pp. 88–100.
Zhu, Joe (2003), Quantitative Models for Performance Evaluation and Benchmarking:
Data Envelopment Analysis with Spreadsheets and DEA Excel Solver, Kluwer Academic
Publishers, Boston.
... EBIT generated by new products etc.) is difficult to determine by using the existing approaches. A comprehensive multivariate input-output-model for the innovation process is presented in Kauffeldt et al. (2012). ...
... In order to determine the time lag between innovation inputs and associated innovation outputs a more comprehensive model of input and output indicators might enhance the validity of the results. A suggestion for a comprehensive input-output-model of the innovation process and an approach to determine the efficiency of innovation generation is presented in Kauffeldt et al., (2012). Nevertheless the data collection of the complete input-output-model for a longer time period remains a challenge. ...
Conference Paper
Full-text available
In this paper the results of an empirical analysis concerning the R&D expenditures of quoted European companies during the economical crisis of 2009 and their share price development in the following years 2009-2012 are presented. Companies often have difficulties in allocating R&D resources due to the uncertainty and complexity of the innovation process. One key element raising the uncertainty of the success of R&D projects is the time lag elapsing between the allocation of R&D budgets and the resulting financial success. Based on a study among 1394 European companies operating in different branches the linkage between R&D expenditures during the economical crisis of 2009 and the companies' share price development within the following years has been analyzed in detail. The empirical results show some evidence that companies with a high R&D quota during the economical crisis of 2009 have been more successful in the following years than companies expanding their austerity programs to their R&D activities. Furthermore we have been able to identify differences within the analyzed branches in regard to the specific time lag between inputs and outputs in the innovation process. This is primarily important for companies, willing to reduce uncertainty within the allocation of R&D budgets.
Book
Full-text available
Data Envelopment Analysis: A Comprehensive Text with Models, Applications, References, And DEA-Solver Software, 2nd Edition is designed to provide a systematic introduction to DEA and its uses as a multifaceted tool for evaluating problems in a variety of contexts. Each chapter accompanies its developments with simple numerical examples and discussions of actual applications. Emphasis is placed on the use as well as an understanding of DEA and the topics in this book have been selected and treated accordingly. The first nine chapters cover the basic principles of DEA and the final seven chapters are more advanced treatment of DEA. These final chapters were completely revised into new chapters, reflecting recent developments that greatly extend the power and scope of DEA and lead to new directions for research and DEA uses. Together with the first ten chapters of the basic principles, they will provide students and researchers with a solid understanding of the methodology, its uses and its potential.
Article
Full-text available
Innovation is broadly seen as an essential component of competitiveness, embedded in the organizational structures, processes, products and services within a firm. The objective of this paper is to explore the effects of the organizational, process, product and marketing innovations on the different aspects of firm performance, including innovative, production, market and financial performances, based on an empirical study covering 184 manufacturing firms in Turkey. A theoretical framework is empirically tested identifying the relationships amid innovations and firm performance through an integrated innovation-performance analysis. The results reveal the positive effects of innovations on firm performance in manufacturing industries.
Article
An exploratory model of competence and performance is presented and tested in the study. Competence in marketing, product innovation, manufacturing, and human resource development across a list of 42 items have been analyzed. Marketing competence was ranked high by Chinese managers. Elements contributed to manufacturing competence regarding process reengineering have not been emphasized as much by Chinese managers. Human resource competence, on the other hand, has been found to be significantly correlated with performance. The results clearly show a trend moving from the state-planned economy to the market economy.
Article
"How does the level of business innovation investment really impact company growth and performance?" This question was put to an Industrial Research Institute panel for discussion at its annual meeting, May 2006. The panelists were also asked to comment on an initial hypothesis presented in strategy + business by Alexander Kandybin and Martin Kihn, of Booz Allen Hamilton, that there is no broad correlation between innovation investment and growth. In their responses, panelists from Air Products and Chemicals, Procter & Gamble and The Monitor Group suggest ways to move forward.
Article
Data on R & D expenditures and US patenting in the world's 660 largest, technologically active firms both show a linear relation between firm size and the volume of innovative activities, despite some inconsistencies between the two measures. Technological diversity increases with firm size, but relatively weakly, especially in R & D-intensive sectors. Sectoral variations in the comparative technological advantage of large firms based in Japan, the USA and W. Europe are better explained by variations in rivalry in their home market than by variations in their size.
Article
This study examines the potential effects of variable set expansion and data variations upon the efficiency scores generated using the Data Envelopment Analysis (DEA) model. It was found that variable set expansion (either through disaggregation of existing variables or addition of new factors) should produce an upward trend in efficiency scores. In addition, ample opportunity exists for ‘decision-making units’ to increase their efficiency scores through manipulation of reported data. In real-world applications of DEA, these problems must be resolved as much as possible (e.g. increased audit of data) in order to improve DEA's practical usefulness and reliability.
Article
The identification of technologically innovating firms is related to the development of indicators of their innovation activity. In this study, we carry out a review of indicators of firms’ technological innovation activity and then classify them according to whether they are based on inputs or outputs of the innovation process, and to the sources of the information they consider, whether primary or secondary. Their applicability as indicators of firm innovation is analyzed in the Spanish ceramic tile industry for the period from 1994 to 1996. From the results obtained we are able to conclude that the method based on direct information—self-assessment by managers—is more effective in identifying both product and process innovators. Within the set of methods based on secondary information, the literature-based innovation output (LBIO) method is the one which provides the best results when identifying product innovators alone.