An empirical investigation of advanced manufacturing technology investment patterns: Evidence from a developing country

J. Eng. Technol. Manage. 30 (2013) 136–156

Contents lists available at SciVerse ScienceDirect

Journal of Engineering andTechnology Management

journal homepage: www.elsevier.com/locate/jengtecman

An empirical investigation of advanced manufacturingtechnology investment patterns: Evidence from adeveloping country

Hasan Bulbul a, Nuri Omurbek b, Turan Paksoy c, Tolga Bektas d,*a Department of Management, Nigde University, Main Campus, Nigde 51240, Turkeyb Department of Management, Suleyman Demirel University, East Campus, Isparta 32260, Turkeyc Department of Industrial Engineering, Selcuk University, Keykubad Campus, Konya 42031, Turkeyd School of Management, Faculty of Business & Law, University of Southampton, Southampton, SO17 1BJ, UK

A R T I C L E I N F O

Keywords:

Advanced manufacturing technology

Taxonomy

Performance

Ownership

Turkish automotive industry

A B S T R A C T

Advanced manufacturing technology (AMT) investment patterns in

developing countries is in need of further investigation, particularly

in the light of the conflicting evidence from the literature. This

paper provides new evidence on AMT investment patterns from the

Turkish automotive industry and develops a taxonomy by exploring

the relationships between AMT investment patterns, ownership

structure, firm size and performance. Analysis of industry survey

data suggests the existence of three groups with different AMT

investment strategies. Results suggest that AMT investment

patterns are not only significantly correlated with firm performance

or ownership, but also reveals significant differences in manufac-

turing performance across investment patterns.

� 2013 Elsevier B.V. All rights reserved.

Introduction

Changing and increasing customer expectations and the inefficiency of conventional manufactur-ing have led manufacturers to consider new manufacturing approaches such as advancedmanufacturing technology (AMT), which allows higher quality and flexibility at lower cost. AMTapplies a range of technologies that utilize computers to control or monitor the manufacturing process(Boyer et al., 1996; Jonsson, 2000). It involves manufacturing techniques and machines combined with

* Corresponding author. Tel.: +44 0 23 8059 8969; fax: +44 0 23 8059 3844.

E-mail address: [email protected] (T. Bektas ).

0923-4748/$ – see front matter � 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.jengtecman.2013.01.002


mailto:[email protected]

http://www.sciencedirect.com/science/journal/09234748


H. Bulbul et al. / Journal of Engineering and Technology Management 30 (2013) 136–156 137

information technology, microelectronics and new practices in the manufacturing process (Boyeret al., 1996; Small and Chen, 1997; Burgess and Gules, 1998; Beaumont et al., 2002).

Diaz et al. (2003, p. 579) state that the ‘‘relevant literature dealing with an analysis of AMTinvestment patterns is fairly recent and still relatively scarce’’. In their pioneering study, Boyer et al.(1996) proposed a taxonomy of AMTs used in the North American metal-working industries based onrelative investment in design, manufacturing and administrative technologies. Following the work ofBoyer et al. (1996), Jonsson (2000) and Diaz et al. (2003) present similar taxonomies based on theSwedish metal industry and the Spanish aeronautical industry, respectively. A common characteristicof the studies by Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003) is their focus on developedcountries as far as the analysis of investment patterns in AMT is concerned. However, macro- andmicro-economical factors are just important in AMT adoption (Alcorta, 1999), and developingcountries face more challenges than developed countries in this respect. For example, firms indeveloping countries have a less educated workforce, more limited capital and resources, and a lessorganized economic system relative to firms in developed countries (Prasad et al., 2005). Furthermore,each industry has its own process structure while the type of industry affects the associatedmanufacturing activities (Hayes and Wheelwright, 1979; Swamidass and Newell, 1987). For thisreason, AMT investments may change from one country and one industry to another, and thetechnological strategies identified in the previous studies may not be relevant to firms in developingcountries.

There are three main motivations for this research: (i) the contradictory findings of previousstudies empirically exploring the impact of AMT investment strategies on firm performancenecessitates further evidence, (ii) the lack of studies exploring AMT investment patterns in developingcountries, and (iii) the suggestion repeatedly echoed in the literature to study taxonomies over timeand in different contexts (Miller and Roth, 1994; Kathuria, 2000; Frohlich and Dixon, 2001), which istaken up in this research.

The main aim of this study is to identify whether firms can be differentiated in their investments inAMT and to explore manufacturing and firm performances in the automotive industry of an emergingand developing economy, in this case Turkey1. The study aims to contribute to the operationsmanagement literature by (i) providing new evidence from the automotive sector of a developingcountry in which the use of AMT is highly relevant, which, to our knowledge is a first, and (ii) toprovide a taxonomy of AMT investments in this sector and to identify differences and similarities fromprevious studies. The results are also of use for managers to understand the characteristics of eachpattern, and the relationship between the patterns, performance, size, and ownership structure in thiscontext.

The rest of the paper is structured as follows. In Literature review and working hypotheses, wepresent a review of the relevant literature including the existing taxonomies proposed on AMTinvestments, and describe the working hypotheses. The methodological approach taken in this paperis presented in Methodology, followed by an analysis of the data and presentation of the results inAnalysis and results. Conclusions are given in section conclusions.

Literature review and working hypotheses

There are numerous studies which show the potential benefits of using AMT in aidingmanufacturers to gain competitive advantage by improving their technological prowess and abilityto manufacture a wide range of products at low volumes without a significant increase in costs orpenalties (Adler, 1988; Gerwin and Kolodny, 1992; Dean and Snell, 1996; Kotha and Swamidass,2000), increasing productivity (Swamidass and Kotha, 1998), reducing direct labor costs, rework costs,and work-in-progress inventories (Zammuto and O’Connor, 1992; Zairi, 1993; Ghani and Jayabalan,2000; Lewis and Boyer, 2002), and establishing closer and more responsive links to markets withoutincreasing costs (Gupta et al., 1997). Swink and Nair (2007), however, stated that the existing

1 See the lists by The World Bank http://data.worldbank.org/about/country-classifications/country-and-lending-groups and

International Monetary Fund (IMF) http://www.imf.org/external/pubs/ft/weo/2012/01/pdf/text.pdf (both accessed 12 January

2013) for a list of developing countries, which Turkey appears as one in the Central and Eastern Europe region.

http://data.worldbank.org/about/country-classifications/country-and-lending-groups

http://www.imf.org/external/pubs/ft/weo/2012/01/pdf/text.pdf

H. Bulbul et al. / Journal of Engineering and Technology Management 30 (2013) 136–156138

empirical evidence showing these benefits is surprisingly limited and highly diverse. For example,while some studies find a significant relationship between AMT and manufacturing or firmperformance (Tracey et al., 1999; Jonsson, 2000; Fawcett and Myers, 2001; Gordon and Sohal, 2001),some others do not report any (Boyer et al., 1996; Beaumont and Schroeder, 1997; Swamidass andKotha, 1998; Cagliano and Spina, 2000; Diaz et al., 2003). Despite these contradictory results, Heineaet al. (2003) MacDougall and Pike (2003) and Choe (2004) note that a considerable number ofmanufacturers still tend to invest in AMT as a competitive tool. Furthermore, Sun (2000) and Sun et al.(2001) posit that more firms will invest in AMT in the future.

Dean et al. (1990) describes four major factors necessary to play a role in successful AMTimplementations, which all revolve around technical, economic and political objectives, althoughChung (1996) suggests that this can also be attributed to inappropriate attention being paid to humanaspects, including the availability of a technological champion and worker involvement in planning.However, the literature also points out that many AMT implementations fail to be successfullyimplemented and thus fail to meet the expectations of their users (Chung, 1996; Hottenstein et al.,1999). In the event that the implementation is successful in one organization, the transfer to otherrelevant sites depends on factors such as previous implementations, advanced manufacturing centers,AMT workshops, informal networks, simultaneous R&D of products and processes, and the impact onoperating performance (Hottenstein et al., 1999). Even so, justifying investments is not easy andrequires that attempts are made to quantify the implications that the chosen technology will have forcost and performance (Small and Chen, 1995).

AMT taxonomies

The purpose of developing taxonomies is to classify different items into named groups that sharecommon characteristics (de Jong and Marsili, 2006). A taxonomy helps practitioners to understand,evaluate and analyze phenomena as it reduces a collection of complex, empirical evidence to few,easy-to-remember categories (de Jong and Marsili, 2006; Zhao et al., 2006), provides a parsimoniousdescription of strategic groups which is practical in discussion, research and pedagogy, and offerssome insight of the underlying structures of competition from the viewpoint of the manufacturingfunction (Miller and Roth, 1994). Taxonomies also serve as an aid to theory building and as a tool tomeaningfully capture the complexities of organizational reality (Ketchen and Shook, 1996; Frohlichand Dixon, 2001).

The identification of strategic groups using taxonomies is a significant research theme in thegeneral strategy and organization literature (Bozarth and McDermott, 1998; Miller and Roth, 1994).Although AMT strategy taxonomies have attracted attention, however, particularly after Boyer et al.(1996), research dealing with taxonomies of AMT investment patterns remains scarce (Diaz et al.,2003) and, to the best of our knowledge, amounts to only three papers. The first of these is Boyer et al.(1996) who, using the data of 202 US firms in the metal-working industries, propose a taxonomy ofmanufacturers based on a cluster analysis of their relative investment in design, manufacturing, andadministrative technologies, resulting in four groups; Traditionalists, Designers, Generalists, and HighInvestors. Traditionalists represent the lowest level of investment across all technology types.Designers invest heavily in design-related AMTs, but with low investment in both manufacturing andadministrative-based technologies. Generalists have relatively larger investment in most technol-ogies. High Investors have the highest rates of investment in all technology types. The second study isthat of Jonsson (2000) who, based on a survey of 324 Swedish metal working firms, identifies threegroups as Traditionalists, Hard Integrators and High Investors. Traditionalists are relatively small-sizefirms with the lowest mean values in AMT investment and integration. Hard Integrators have thesecond highest investment rates and higher integration levels than Traditionalists. Finally, HighInvestors have the highest means in both investment and integration. The taxonomy of Diaz et al.(2003) is based on a survey of 20 small- and medium-sized firms in the aeronautical industry in Spain.The authors also identify three groups; Traditionalists, Designers and Investors. Traditionalists havethe lowest investment in design, manufacturing and planning technologies, Designers have thehighest investment in design-related technologies, and Investors have the greatest amount ofinvestment in all AMT types.

Table 1A summary of previous research on AMT taxonomies.

1.Traditionalists 2.Designers 3.Generalists 4.High Investors

Boyer et al. (1996) Design [2,3,4] [1,3,4] [1,2,4] [1,2,3]

Manufacturing [3,4] [3,4] [1,2,4] [1,2,3]

Administrative [3,4] [3,4] [1,2,4] [1,2,3]

Size (no. employees) [4] [1]

Sales growth No differences No differences No differences No differences

Return on sales No differences No differences No differences No differences

Earnings growth No differences No differences No differences No differences

1.Traditionalists 2.Hard integrators 3.High Investors

Jonsson (2000) Design [2,3] [1,3] [1,2]

Manufacturing [2,3] [1,3] [1,2]

Administrative [2,3] [1,3] [1,2]

Size (no. employees) [3] [3] [1,2]

Profit [3] [3] [1,2]

Growth [3] [1]

Flexibility (variety) No differences No differences No differences

1.Traditionalists 2.Designers 3.Investors

Diaz et al. (2003) Design [2,3] [1] [1]

Manufacturing [3] [3] [1,2]

Planning [3] [3] [1,2]

Size (annual sales) [3] [3] [1,2]

Growth No differences No differences No differences

Profit No differences No differences No differences

Note: Numbers in brackets indicate other groups with significant differences based on the pairwise comparison tests.


We present a summary and an overview of the taxonomies proposed by Boyer et al. (1996), Jonsson(2000) and Diaz et al. (2003) in Table 1.

The above exposition shows that AMT investment strategies in developed countries are relativelywell analyzed and understood. This is not the case for developing countries, however. Althoughtechnology is considered one of the most important tools in global competition, firms in developingcountries in general do not have the capability to compete with firms that have implemented state-of-the-art AMT (Putranto et al., 2003). Moreover, developed countries are in a better financial position ingeneral, as well as a more qualified and experienced workforce and a broader knowledge oftechnologies than developing countries (Chamarbagwala et al., 2000). The required technologies indeveloping countries are most likely to come from developed countries through technology transfer(Putranto et al., 2003). For this reason, it is expected that AMT investment strategies will differ indeveloping countries from those of developed countries, but the question still remains open – whichwe address in this paper.

The above discussions suggest that AMT investments are likely to be associated with differenttypes of industries and countries. Swamidass and Newell (1987) point out that process type andindustry are two variables, which influence operations. Safizadeh et al. (1996) find a strong correlationbetween process structure and production competencies. It is commonly acknowledged in theliterature that discrete product-manufacturing industries extensively use AMTs (Dean and Snell,1991; Swamidass, 2003). The taxonomies offered in the operations management literature (Miller andRoth, 1994; Kathuria, 2000; Frohlich and Dixon, 2001) suggest that future studies are undertaken overtime and in different settings.

In our study, the question of whether the AMT investment behaviors in the Turkish automotiveindustry show different patterns gives rise to the first hypothesis that will be tested in this paper:

H1. The firms in the Turkish automotive industry can be classified into distinct groups based on theirrelative investments in AMT.


Using H1, this paper will investigate the issue of whether different investment behaviors (if any)exist in the Turkish automotive industry or whether they are actually the AMT investment patterns ofdeveloped countries.

Size

Firm size has long been considered an important indicator of AMT use since larger companies are ina healthier position financially and are better able to take advantage of economies of scale (Meredith,1987; Swamidass and Kotha, 1998; Sohal et al., 2001). Previous empirical studies indicate that firmsize plays a critical role in terms of the level of adoption and use of manufacturing technologies andpractices (Thong and Yap, 1995; Lefebvre et al., 1996; Sohal et al., 2001; Buonanno et al., 2005;Salaheldin, 2007). In the AMT context, Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003) findsignificant differences between AMT investment patterns and firm size and report that largercompanies tend to invest more in AMTs than smaller ones do. Swamidass and Kotha (1998) study asample of 160 US manufacturing companies and find that there exists a positive correlation betweenfirm size and three AMT factors that they considered; information exchange and planning technology,high volume automation technology and low volume process automation technology. In a studyconducted within Australian manufacturing companies, Sohal et al. (2006) again report a positiverelationship between firm size and AMT adoption.

In contrast, some recent studies show that the use of AMT and firm size are not necessarilycorrelated. For example, using data from 224 North American plants, Swink and Nair’s (2007) findingssuggest there is no correlation between these two parameters. Similarly, Khazanchi et al. (2007) find asimilar result in a study of 110 firms of which 76.3% are small-sized companies. It has also beensuggested that there is increasing usage of technology in small- and medium-sized companies(Mabert et al., 2000), and that small manufacturers are recognizing that investment in AMTs is vital forsurvival in a world of growing competition and increasing customer needs (Ordoobadi and Mulvaney,2001).

Given the contrasting views in the literature, this study aims to generate at providing new evidencefor the relationship between AMT investment and firm size in the automotive sector. In particular, thegoal is to see whether there are any differences, with respect to investment in AMT, across firms ofdifferent sizes, where firm size is measured by the number of employees and annual sales. This givesrise to the second hypothesis to be tested in this study.

H2. Large firms invest more heavily in AMTs than smaller firms, where firm size is defined with respect tothe number of employees and annual sales.

Ownership

Firms in developed countries share their financial and technological competencies with the firmsthat they invest in or collaborate with in developing countries (Markusen, 2002; Douma et al., 2006).Schroder and Sohal (1999) observed that principal ownership of firms by country reflects thedifferences in management styles, strategies and practices based on a country or national culture. Thefindings of Kotha and Swamidass (1998) suggest that the nationality of the firm is an important factorin AMT use. The study by Beaumont et al. (2002) reveals that AMT investments of foreign-owned firmsare higher than those of local firms. Salaheldin (2007) examines manufacturers in eight differentindustries in Egypt and concludes that ownership has a significant impact on AMT adoption.

In the context of the automotive industry, Laosirihongthong et al. (2003) study the use of differenttechnologies in Thailand based on the ownership structure. The findings of the study are that whileThai-owned firms adopt relatively cheaper technologies such as CAD, foreign investors in the countrymostly adopt expensive technologies. Laosirihongthong et al. (2003) do not, however, shed any lighton AMT investment models used in the Thai automotive industry.

Firms in developing countries operate in a complex and uncertain environment with growingglobal competition. Specifically, Turkish firms have been facing serious competition since thebeginning of the economical reforms in the 1980s, particularly since joining the Customs Union of the


EU in 1996. As a result of global competition and the EU membership process, there have beenincreased collaborative arrangements and strategic partnerships between Turkish and foreignmanufacturers in the automotive industry (Wasti et al., 2005). There are currently about 200 foreignpartnerships in the industry (TAYSAD, 2008; OSD, 2010).

Turkish automotive firms have improved their manufacturing capabilities in the areas of quality,delivery and flexibility in order to respond to the requirements of foreign businesses (Burgess andGules, 1998). Foreign investments as well as strong competition in the industry are likely to haveincreased the AMT investments over recent years. The question of whether the AMT investmentpatterns of automotive firms differ with their ownership structure will be investigated throughhypothesis three, below:

H3. Foreign-owned firms have higher levels of AMT investment than domestic firms.

Performance

The potential benefits of AMT are widely reported in the literature (Ghani and Jayabalan, 2000;Lewis and Boyer, 2002; Zhang, 2005). However, empirical findings about the relationship betweenAMT and performance have been contradictory and inconsistent.

Boyer et al. (1996), Swamidass and Kotha (1998) and Diaz et al. (2003) show there is no significantrelationship between AMT use and firm performance. Cagliano and Spina (2000) suggest that there isno significant effect of AMT adoption on manufacturing performance. In contrast, Beaumont andSchroeder (1997) find a positive, albeit non-significant, impact of AMT on firm and manufacturingperformance. Jonsson (2000) found that there is a relationship between AMT patterns and growth,profitability, and flexibility. Gordon and Sohal (2001) report that firms with higher performance useAMT more than firms with lower performance do. Tracey et al. (1999) and Fawcett and Myers (2001)showed that there is a significant relationship between AMT use and performance. Raymond and St-Pierre (2005) find that AMT has a significant impact on both the operational and the organizationalperformance.

Over recent years, the Turkish automotive industry has strived to attain international standards, asa result of which it became Europe’s sixth and the world’s sixteenth biggest automotive industry(TAYSAD, 2008; OSD, 2010). In order to study the relationships between AMT investment patterns andfirm performance in the Turkish automotive industry, the following hypotheses will be examined.

H4a. Firms with a high level of AMT investment have higher manufacturing performance compared withfirms with lower levels of AMT investment.

H4b. Firms with a high level of AMT investment have higher overall performance compared with firmswith lower levels of AMT investment.

Methodology

Our study is conducted in the spirit of Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003), butdiffers from these studies with respect to the industry under focus and the developing nature of thecountry. This study also examines the impact of ownership structure on AMT investment patterns, anissue which has not been fully investigated in the prior studies despite that fact that a firm’s AMT usemay differ according to their ownership structure (Beaumont et al., 2002; Pyke et al., 2002), as well asthe impact of firm. The theoretical model used in this study is given in Fig. 1.

Data collection

There are several motivations behind the focus of this study on the automotive industry. First,automotive industry operations involve the manufacturing of discrete products based on metal and

Fig. 1. The theoretical model linking the contextual factors and the outcome variables.


non-metal fabrication (Kotha and Swamidass, 2000) and therefore considered to be an industry wherethere is wide use of AMTs (Dean and Snell, 1991; Swamidass, 2003). Second, firms in automotiveindustry compete in a complex and uncertain environment, an environment where global competitionis intense and markets change rapidly and in which the use of AMTs spreads exponentially. One otherreason is that the automotive industry is of strategic importance for all countries as it adds value to theeconomy, and contributes to employment as well as to other industries (Segarra, 1999; Koste andMalhotra, 2000; Laosirihongthong et al., 2003).

In this study, a survey has been conducted among firms that are listed in the International StandardIndustrial Classification (ISIC 384) as transport equipment manufacturers listed in Turkey’s Leading1000 Industrial Enterprises (ISO, 2006) database. Production/operations or plant managers of thefirms identified are key informants for the necessary data collection and were contacted to ensure thattheir data would be valid and reliable (Beaumont et al., 2002; Laosirihongthong et al., 2003).

There are 74 firms operating in the Turkish automotive industry as listed in the ISIC 384 category ofthe 2007 ISO database. In the first stage of the data collection process, the names of plants orproduction managers were extracted from corporate websites. A questionnaire was then sent out tothe managers, along with a personalized cover letter explaining the objective of the study and a pre-addressed and stamped return envelope. Twenty four fully completed and usable questionnaires werereceived within six weeks of their distribution. Reminder emails were sent out to those who had notresponded. Seven more questionnaires were then received in the ensuing four weeks. In summary, thedata collection process resulted in 31 usable responses, yielding a response rate of 42%. This ratecompares favorably to those of similar studies in the operations management literature based on asingle industry; in particular, the 41.1% of Boyer et al. (1996) the 38% of Jonsson (2000), and the 33.8%of Dangayach and Deshmukh (2006).

The responding firms include nine (29% of the respondents’ sample) automobile assemblers and 22(71% of the respondents’ sample) parts and component manufacturers. Respondents had an average of1289 employees, with the smallest and the largest firms employing 233 and 8000 workers,respectively. Annual sales for 2005 ranged from $20 million to $3.8 billion, with an average annualsales figure of $435 million. Approximately half (48.4%) of the firms are Turkish-owned, 12 of them(38.7%) are joint ventures and the rest (12.9%) are foreign-owned.

A test for non-respondent bias was conducted using t-statistics. Two approaches were used toassess non-response bias. The first approach involved comparing early respondents with late ones(Lambert and Harrington, 1990). No significant differences were found between early and laterespondents in all variables, including sales figures, number of employees and performance items. Thesecond approach involved comparing the annual sales and number of employees of the respondingand non-responding firms, the results of which are shown in Table 2. Firms’ sales figures and numbersof employees were again obtained from the ISO 1000 database.

The t-statistics shown in Table 2 do not indicate any significant differences (p>0.05) between thetwo sample groups, indicating that non-response bias is not a factor. Hence, the respondents’ samplewas considered representative of the automotive industry in Turkey.

Measurement and reliability

For this study, a survey instrument in the form of a questionnaire was developed based on theliterature and was refined as follows:

Table 2Comparisons between respondents and non-respondents.

Characteristics Respondents Non-respondents t-test

Number of firms 31 43

Annual sales ($)

Mean 434,972,889 118,626,886 t=1999

Std. dev. 853,680,102 256,238,647 p>0.05

No. of employees

Mean 1289 755 t=1795

Std. dev. 1611 934 p>0.05


(i) a
ll items in the questionnaire were adapted from previously published studies, (ii) a fter several internal revisions, the questionnaire was reviewed by two scholars, familiar with
psychometric measurements, who examined the logic of the questionnaire design. Thequestionnaire was then discussed with academic scholars, who had experience in operations,to assess the content validity prior to pilot testing,

(iii) a
pilot test was conducted with four managers, whose feedback was used to improve the clarity,comprehensiveness and relevance of the research instrument. Comments from the scholars andthe managers indicated that the questionnaire appeared to have adequate face validity sincealmost all of the items were clear to interpret and answer, and respondent comments did notindicate any confusion or ambiguity. Hence, no major revisions to the questionnaire wererequired.
The questionnaire consists of three parts. The questions in Part I serve to obtain background informationfrom the responding firms. Part II of the questionnaire is intended to measure the scale of AMT investment,using a five-point Likert-type scale ranging from 1 (no investment) to 3 (moderate investment) and 5(heavy investment). Our variables encapsulate broad conceptualizations of AMT that include soft and hardtechnologies. Three categories of AMT were identified, which correspond to their function or type ofactivity (design, manufacturing and administrative) as proposed by Adler (1988) and Boyer et al. (1996),and which Table 3 presents in detail. In Part III, the respondents were asked to assess and compare theirfirm and manufacturing performance with those of their competitors. Swamidass and Newell (1987)highlighted the difficulty of obtaining objective financial measures of performance such as profit growth,profit margin, sales increase, and market share. Since firms can be reluctant to answer financial questions(Vickery et al., 1993; Boyer et al., 1996; Ward and Duray, 2000), firm and manufacturing performanceshave been measured using non-financial indicators.

Firm performance includes four common financial or marketing indicators, which are market sharegrowth, sales growth, return on investment (ROI) and return on sales (ROS). While the first two itemsreflect growth, the latter two items reflect profitability. These items are not new and have been used inprior operations management studies (see, e.g., Swamidass and Newell, 1987; Venkatraman andRamanujam, 1987; Vickery et al., 1993; Curkovic et al., 2000).

Manufacturing performance measures the fundamental objectives of cost, quality, flexibility, anddelivery, as in previous studies (e.g., Hayes and Wheelwright, 1984; Krajewski and Ritzman, 2001).Innovation is also included in our study, as in Vickery et al. (1997) and Narasimhan and Das (2001).Successful firms tend to simultaneously pursue multiple performance objectives, rather than purelyfocusing on one objective (Roth and Miller, 1990) and studies exist that take multiple performanceobjectives into account (e.g., Narasimhan and Das, 2001; Ahmad and Schroeder, 2003; Das andJayaram, 2003). In line with such work, we treat manufacturing performance as a composite constructcomposed of multiple measures; namely cost, quality, flexibility, delivery, and innovationperformance. To measure each, eight factors have been considered: (1) production cost, (2) laborproductivity, (3) product quality and performance, (4) volume flexibility, (5) variety flexibility, (6)delivery speed and reliability, (7) level of innovativeness, and (8) new product development time.

All performance items were based on the respondents’ assessment of the firm and manufacturingperformances for the last five years. Previous studies indicate that managerial assessments rely onobjective data (Venkatraman and Ramanujam, 1987; Powell and Dent-Micallef, 1997) and managers,

Table 3AMT, firm and manufacturing performance items – descriptive statistics.

Dimensions and items of AMT and performance Mean Std. dev. a

Manufacturing AMTs 3.22 1.21 0.87

Computer-aided manufacturing (CAM) 3.48 1.23

Robotics 2.64 1.22

Group technology (GT) 3.25 1.21

Flexible manufacturing systems (FMS) 3.22 1.33

Comp. numerical control machines (CNC) 3.90 1.13

Automated material handling systems 2.90 1.01

Bar coding/automatic identification 3.12 1.31

Administrative AMTs 3.80 1.10 0.84

Electronic data interchange (EDI) 3.96 0.87

Decision support systems (DSS) 3.45 1.26

Material requirements planning (MRP) 4.06 0.89

Manufacturing resource planning (MRPII) 3.77 1.23

Just in time (JIT) 3.74 1.21

Design AMTs 3.75 0.40 0.83

Computer-aided design (CAD) 4.12 1.02

Computer-aided engineering (CAE) 3.80 0.98

Computer-aided process planning (CAPP) 3.32 1.27

Manufacturing performancea 3.71 0.56 0.88

Labor productivity 3.58 0.95

Production costs 2.77 1.02

Product quality 4.12 0.67

Volume flexibility 4.06 0.81

Variety (product line) flexibility 4.25 0.81

Delivery speed 3.51 0.92

Delivery reliability 4.29 0.78

Innovativeness 3.09 1.22

Firm performance 3.67 0.87 0.77

Market share growth 3.44 0.82

Sales growth 3.68 0.89

Return on investment (ROI) 3.65 0.87

Return on sales (ROS) 3.89 0.81a Two items (product performance and new product development time) were removed because of low item – total correlations.


well-acquainted with firm performance measures, can provide an accurate subjective assessment(Choi and Eboch, 1998). Using a method described by Vickery et al. (1993), we also examined thecorrelation between perceptual and objective measures. The results show a highly positive correlation(r=0.77; p<0.001) between subjective measure of sales growth and annual sales as obtained from theISO 1000 database.

Content validity of all scales has been tested through the adoption of items used in prior studies(Flynn et al., 1990) and the pilot test conducted with on practitioners and scholars. Cronbach’s(coefficient) alpha was calculated to assess the internal consistency of each scale. As shown in Table 3,the respective Cronbach’s alpha (shown under the column labeled a) values for Manufacturing AMTs,Administrative AMTs, Design AMTs, and Firm Performance are 0.88, 0.84, 0.83, and 0.77, respectively,all of which are greater than 0.70 (Nunnally, 1978). As for manufacturing performance, two items(product performance and new product development time) were omitted because they attenuated theinternal consistency of this scale. The resulting eight-item scale showed good internal consistency,with an alpha coefficient of 0.88. All scales used in the analysis therefore have good reliability.

Analysis and results

This section describes the analysis carried out in order to test Hypotheses and presents the results.The analysis is performed in two stages: (i) to identify distinctive groups of Turkish automotive firmswith respect to their AMT investments (i.e., H1), for which cluster analysis is used, and (ii) to examinerelationships between context (ownership and size) and outcome factors (performance) by testing H2,H3, H4a and H4b using analysis of variance.


An empirical taxonomy of AMT investment in the automotive industry

Cluster analysis is often employed in the literature to identify classes or clusters of objects (Ketchenand Shook, 1996). It is a multivariate statistical technique that groups sets of objects based on thecharacteristics that they possess, so that clusters exhibit high internal homogeneity and high externalheterogeneity (Hair et al., 1998). We resort to this technique to develop taxonomy of AMT investmentsin the Turkish automotive sector.

One of the key questions in cluster analysis is the number of clusters to be used. Lehmann (1979)argues that the number of clusters should be between n/30 and n/60, where n represents the samplesize. Due to the small number of firms surveyed in this study, Lehmann’s rule is inapplicable. Analternative is to use a hierarchical dendrogram and an agglomeration coefficient, a measure ofdistance between two clusters being combined at each stage in the cluster analysis. A large increase ora large percentage change in the agglomeration coefficient when performing a hierarchical clusteranalysis indicates a fairly good cutoff point (Ketchen and Shook, 1996; Hair et al., 1998). Thehierarchical cluster analysis used in this study is based on Ward’s method (Ward, 1963), wheresquared Euclidean distance is used as a measure of distance between clusters. The method minimizesintra-cluster differences and maximizes inter-cluster differences among the variables used forclustering (Frohlich and Dixon, 2001; de Jong and Marsili, 2006). The agglomeration coefficients andpercentage changes calculated for the data set of this study are shown in Table 4.

The findings in Table 4 highlight that the most important changes of the agglomeration coefficientoccur in the second and the third clusters. On the other hand, the highest difference amongpercentages of change occurs in three clusters. This suggests that the appropriate number of clustersto use for this study is three. Fig. 2 presents the hierarchical dendrogram showing the three clusters

Table 4Analysis of agglomeration coefficients.

Number of

cluster

Agglomeration

coefficients

Percentage change

in the coefficient

Differences between

percentage change

10 175.86 9.19 0.79

9 192.02 9.98 �0.53

8 211.19 9.45 0.24

7 231.16 9.69 1.17

6 253.56 10.8 2.12

5 281.11 12.9 1.04

4 317.61 14.0 7.60

3 362.15 21.6 15.0

2 440.48 36.6

1 602.06

Fig. 2. Hierarchical dendrogram using Ward’s method.

Table 5AMT investments by cluster: results of ANOVA and Scheffe test.

Investors

(Cluster 1, n=7)

Followers

(Cluster 2, n=18)

Traditionalists

(Cluster 3, n=6)

ANOVA

Design [3**] [3***] [1**, 2***]

Cluster mean 4.42 3.85 2.66 F=8.90

Std. dev. 0.49 0.80 0.89 P<0.01

Manufacturing [2*, 3*] [1*, 3**] [1*, 2**]

Cluster mean 4.38 3.15 2.07 F=31.68

Std. dev. 0.13 0.55 0.68 P<0.001

Administrative [2***, 3*] [1***, 3*] [1*, 2*]

Cluster mean 4.57 3.91 2.56 F=22.08

Std. dev. 0.33 0.62 0.48 P<0.001

Note: Numbers in brackets indicate the group numbers from which this group was significantly different.* p<0.001.** p<0.01.*** p<0.05.


formed, implying that three different groups of firms exist in the Turkish automotive industry,differentiated by their level investments in AMT.

A one-way ANOVA test was used to check whether the clusters significantly differed from oneanother with respect to the three variables, i.e., Design, Manufacturing and Administrative. A Scheffepairwise comparison test was employed for a more detailed analysis of the differences betweenindividual pairs of groups on each of the three individual variables. The results of these tests arepresented in Table 5; these show significantly different means between all groups and pairs of groupson all three variables. This finding provides support for H1, i.e., Turkish automotive firms can beclassified into different strategic groups based on their investments in AMT.

The clusters were named based on their AMT investments as follows: Investors (Cluster 1),Followers (Cluster 2) and Traditionalists (Cluster 3). The naming of the three AMT investment patternsrepresented by the clusters were based on: (i) level of investments in AMT, (ii) differences betweenclusters with respect to AMT investments, and (iii) a comparison of the findings with those of the priorstudies (Boyer et al., 1996; Jonsson, 2000; Diaz et al., 2003). We now provide details for each of theseclusters below.

Cluster 1, Investors: The cluster includes seven automotive firms (22.6% of the entire sample) andhas the highest mean investment in all types of AMT variables. More specifically, the means for Design,Manufacturing and Administrative AMT variables are 4.42, 4.38 and 4.57, respectively, which areconsiderable given the highest possible rating of 5. This cluster is named as Investors as it is the sameas those described in Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003).

Investors differ significantly from the other clusters with regards to Manufacturing andAdministrative AMTs. There is, however, no significant difference in Design AMTs between investorsand followers. This result suggests that investors heavily make use of modern manufacturingtechnologies and approaches.

Cluster 2, Followers: This is the largest group with 18 firms (58% of the entire sample) and have thesecond highest group means in all three types of AMT clustering variables. This cluster only partiallyresembles the Hard Integrators cluster of Jonsson (2000) and the Designers cluster of Diaz et al. (2003)insofar as the mean and the level of AMT investments are concerned. However, it differs from theformer with the type of typology used. As for the latter, the Designer cluster has no statisticallysignificant differences with the Traditionalists except for design technologies. For this reason, we havenot used either of these terms to avoid confusion. Finally, this cluster is the same as the Generalistscluster of Boyer et al. (1996) as far as the level and ranking of AMT investments are concerned, butdiffers with regard to two aspects: (i) followers in our study have the largest membership whereas inBoyer et al. (1996) the cluster with the highest number of members is Traditionalists, and (ii) there areno statistically significant differences between Followers and Investors in our study in all types ofAMT, whereas with Boyer et al. (1996), there are statistically significant differences between Investorsand Generalists in all types of AMT. It is for this reason that we have chosen a different naming for this


cluster, one which reflects the relatively high investment made on all three AMT variables. Inparticular, the investment levels of Followers in Manufacturing and Administrative AMTs, althoughstatistically different, compare very favorably with those of Investors, suggesting that the formerclosely ‘‘follows’’ the latter.

Cluster 3, Traditionalists: The third AMT cluster consists of six firms and they represent the smallestpercentage (19.3%) of the entire sample. This cluster has the smallest investments means in all AMTvariables, and differs significantly from the other two clusters. For this reason, it has been namedTraditionalists in the same way as in Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003).

Traditionalists have less than moderate levels of investment in Design, Manufacturing andAdministrative AMTs, with 2.66, 2.07 and 2.56 as cluster means, respectively. These results suggestthat traditionalists are less innovative manufacturers, and tend to employ conventional technologies.

Taxonomy results: differences and similarities with existing work

There are some similarities and differences between our findings and those of Boyer et al. (1996),Jonsson (2000) and Diaz et al. (2003). The first difference is the high levels of AMT investment found inall clusters in this study when compared to the lower levels seen in previous works. This is likely to beattributed to the difference in the industrial sectors; in particular, the increasing reliance of theinnovative and global structure of the automotive industry on advanced technologies. Anotherdifference is, whereas the highest number of firms in prior studies is usually seen in the traditionalists(particularly in Boyer et al., 1996), this study finds that the same cluster is smallest in size. Findings ofprior studies suggest that one of the AMT dimensions in a cluster is often rated higher than others. Forexample, designers in Boyer et al. (1996) and Diaz et al. (2003) rank the design-oriented characteristicsof the firms highly, whereas the hard integrators group of Jonsson (2000) is formed on the basis offirms having higher integration means. In contrast, we find no significant differences across the AMTdimensions in any given cluster. A balanced distribution of AMT dimensions in each cluster shows thatnone of the dimensions is dominant. One possible explanation of this finding is the importance of allAMTs’ dimensions in the automotive industry, due to the competitive and innovative nature of theindustry. The only exception is that the mean investment level in manufacturing AMTs is slightlylower than the remaining two dimensions. This is an expected outcome given the relatively high costsrequired to invest in manufacturing AMTs.

One of the similarities between the findings of this study and those of Jonsson (2000) and Diaz et al.(2003) is that the traditionalists were found to have the lowest investment levels. The cluster meanvalues of traditionalists in this study are high on average in absolute terms, as in Boyer et al. (1996).Findings of both studies therefore suggest that traditionalists invest slightly higher in design andlower in manufacturing AMTs. Another similarity between the followers of this study and thegeneralists of Boyer et al. (1996) is that both clusters have significant AMT investments and theyfollow a higher investment group. They also follow similar patterns of investment. A further similaritywith the previous studies concerns the investors, for whom AMT investments in all dimensions are attheir highest levels. Furthermore, as with the investors in Boyer et al. (1996) and Jonsson (2000), ourinvestors make their lowest level of investments in manufacturing AMTs. One last similarity is relatedto the level of investments in manufacturing AMTs. All clusters identified in Boyer et al. (1996),Jonsson (2000) and in the present study have their lowest level of investments in manufacturingAMTs. This finding thus re-emphasizes the importance of the cost of investments in manufacturingAMTs, which has a significant effect on investment decisions.

Patterns of AMT investment and firm size

The number of employees in the firms in the sample used in this study ranges from 233 to 8000.Annual sales of the same are in the interval $20 million to $3.8 billion. These two indicators have beenused as representative of firm size in comparing the three different clusters. One reason is to remainconsistent with previous studies that have primarily considered the number of employees asreflecting the firm size (Boyer et al., 1996; Jonsson, 2000). A second reason is to increase reliability ofthe results by using multiple, instead of single, indicators of firm size. Finally, a third reason is to be

Table 6Firm size by cluster: results of ANOVA and Scheffe test.

Measure Investors

(Cluster 1, n=7)

Followers

(Cluster 2, n=18)

Traditionalists

(Cluster 3, n=6)

ANOVA

No. of employees

Cluster mean 2474 970 865 F=2.740

Std. dev. 2648 2042 958 P>0.05

Annual sales ($)

Cluster mean 930,653,148 339,076,536 144,368,312 F=1.720

Std. dev. 1,321,600,959 723,082,941 143,357,197 P>0.05


able to see whether firms will differ in cases where financial status is the only determinant of theirsize.

To find out whether there was a significant difference between the three groups in terms of size, aone-way ANOVA was run with the corresponding multiple comparisons Scheffe test that allows thedifferences to be observed by pairs of clusters. Table 6 shows the results of the tests with number ofemployees and annual sales as the dependent variables and the three groups as levels of theindependent variable.

As Table 6 shows, investors clearly have substantially more employees than the other two groups.It is also noteworthy that mean annual sales are substantially higher in investors compared tofollowers and traditionalists. Despite these differences, however, the results of the one-way ANOVAand Scheffe pairwise test indicate that there are no significant differences across AMT groups (p>0.05)with regard to the size of the companies. For this reason, H2 is rejected.

It might be surprising at first glance to find that the three groups are not statistically different fromone another as far as the firm size is concerned. However, one should bear the following points in mindwhen interpreting the results. First, the sample used in the testing is composed of medium- and large-scale firms only. The findings of Boyer et al. (1996) and Jonsson (2000) suggest that the difference theyfind in their studies is primarily caused by the cluster with the smallest number of employees. In theformer study, for instance, this is a group with a mean of 328.25 employees, which is significantlysmaller than the traditionalist group in our study which has an average of 865 employees. Our resultsare more in line with those of Swink and Nair (2007), who did not consider small-scale firms with lessthan 100 employees and found no particular correlation between AMT investment patterns and firmsize. The fact that our sample does not include any such small-size firms has led us to arrive at similarconclusions.

The second point is the availability of financial resources to invest in process innovation. This is animportant assumption behind the claims made in the literature as to the likelihood of AMTinvestments increasing with firm size. Our findings concerning annual sales (with p=0.197) clearlyshow that when firms are in a financially healthy enough position to make AMT investments, they willproceed regardless of their size. This is in contrast to findings described in prior studies suggesting apositive correlation between AMT investment and firm size, but this is again due to the size of thefirms in their samples.

Patterns of AMT investment and ownership

A significant number of the firms in the Turkish automotive industry are foreign establishments.Table 6 presents a distribution of the groups by ownership and shows that 51.6% of the respondentsample is either wholly or partially owned by foreign investors. A chi-square (x2) test was run to findout whether there was a significant difference between firms’ ownership with respect to AMTinvestment. The test indicated no difference at a 0.05 significance level. In other words, the resultssuggest that AMT investment patterns and firm ownership are not interrelated. H3 is thereforerejected on these grounds (Table 7).

The findings of prior studies commonly draw attention to the fact that foreign firms tend to makemore AMT investments than domestic firms (Schroder and Sohal, 1999; Robb and Xie, 2001;

Table 7Distribution of AMT groups by ownership.a

Ownership Investors

(Cluster 1, n=7)

Followers

(Cluster 2, n=18)

Traditionalists

(Cluster 3, n=6)

Total

Turkish ownership 3 8 4 15

Foreign ownership 1 3 0 4

Joint venture 3 7 2 12

Total 7 18 6 31

Note: A chi-square test of the sample distribution against the expected distribution based on a random distribution does not

indicate any significant difference (x2 =0.81; df=4; p>0.05).a Firms are named as foreign firms if 100% of capital belongs to the foreign firms, they are named as Turkish owned firms if 100% of

capital belongs to the domestic firms. If between 1% and 99% of the firm’s capital belongs to the foreign partner, it is named as joint

venture firm.


Beaumont et al., 2002; Pyke et al., 2002; Salaheldin, 2007). However, most of these rely on multi-industry surveys and do not consider individual industries. An exception is the study byLaosirihongthong et al. (2003), which looked at statistical differences among ownership of the firmsin the use of 15 different AMTs in the automotive industry in Thailand. Their findings suggest thatthere is no significant impact of foreign ownership on the level of AMT use. While, in this study, noneof the traditionalist firms is owned by foreign investors, only three of the foreign-owned firms are thefollowers. A closer look at joint venture and domestic firms shows that the numbers are distributed ina similar way across all three clusters. Our findings therefore parallel those of Laosirihongthong et al.(2003).

Contrary to the common conclusion in the literature, the findings of this study indicate thatforeign-owned firms do not make higher AMT investments in the Turkish automotive industrycompared with others. This is due to the unique characteristics of the automotive industry, whereglobal dynamics are more important than national ones. Domestic firms in the industry feel morepressure to make investments in AMTs for this reason.

Patterns of AMT investment and performance

Performance as used and defined in this study relates to manufacturing as well as overall firmperformance. Table 3 presented the detailed factors used to measure both indicators. To find outwhether there is any significant difference among the three AMT clusters in terms of manufacturingand overall firm performance, a one-way ANOVA and a multiple comparisons Scheffe tests wereconducted. Table 8 shows the results of the ANOVA and Scheffe tests with manufacturing and overallperformance as the dependent variables and three AMT groups as the independent variables.

Table 8 shows that investors with the highest AMT investment levels have the highestmanufacturing performance, and traditionalists with the lowest AMT investments have the lowestmanufacturing performance. These findings are found to be statistically significant using an ANOVAtest (p<0.001).

The results of the Scheffe test indicate that the manufacturing performance of the traditionalists issignificantly different from those of investors and followers, with the level of significance being 0.001and 0.01, respectively. These results also show that a relationship exists between manufacturingperformance and AMT investment patterns, hence the acceptance of hypothesis H4a. However, thereis no significant difference in manufacturing performance between the investors and the followers.

Hypothesis H4b tests whether there is a significant difference across AMT investment patterns interms of overall performance. Although Table 8 suggests that the overall performance of the followersand the traditionalists are similar and the performance of the investors is better than that of the othertwo, the results of the ANOVA test indicate that differences across the three groups are not statisticallysignificant. The Scheffe test also confirms that there are no significant differences between the threeclusters. Hypothesis 4b is therefore rejected.

In order to investigate the issue further, one-way ANOVA and Scheffe tests were repeated on eachof the growth and profit performances. The results of the analysis reveal that the differences across the

Table 8Manufacturing and firm performance by cluster: results of ANOVA and Scheffe test.

Measure Investors

(Cluster 1, n=7)

Followers

(Cluster 2, n=18)

Traditionalists

(Cluster 3, n=6)

ANOVA

Manufacturing performance [3*] [3**] [1*, 2**]

Cluster mean 4.14 3.84 2.83 F=12.46

Std. dev. 0.34 0.42 0.79 P<0.001

Overall performance

Cluster mean 4.03 3.50 3.50 F=1.75

Std. dev. 0.48 0.77 0.41 P>0.05

Growth

Cluster mean 4.14 3.47 3.75 F=2.31

Std. dev. 0.37 0.84 0.42 P>0.05

Profit

Cluster mean 3.92 3.50 3.16 F=1.24

Std. dev. 0.67 0.90 0.98 P>0.05

Note: Numbers in brackets indicate the group numbers from which this group was significantly different.* p<0.001.** p<0.01.


groups are not significant in terms of growth and profit. As a result, no relationship is found betweenAMT investment patterns and overall performance. This finding parallels the results of Boyer et al.(1996), Cagliano and Spina (2000), Kotha and Swamidass (2000) and Diaz et al. (2003). Some possibleexplanations of this result as given in prior studies are; the quality of the implementation process,eventualities related to customers and firm products, the principle of equifinality, and delay betweenthe investments made and the resulting improvement in the performance (Boyer et al., 1996; Diazet al., 2003). The latter is particularly relevant in our case, as the Turkish automotive firms have onlyrecently started to make substantial investments in AMT (TAYSAD, 2008).

Some authors argue that principle of equifinality can also affect the results (Boyer et al., 1996; Diazet al., 2003). The principle of equifinality is valid only to a certain extent here since various paths orstrategies implemented and pursued by a firm may lead to equal outcomes. The differences inmanufacturing performances by AMT investment patterns indicate that the principle of equifinality isnot fully applicable in this study. In this case, another relevant factor to consider here is institutionalisomorphorism (DiMaggio and Powell, 1983), in particular mimetic isomorphorism, wherebymanufacturers mimic organizations they are closely associated with or those who they perceive to besuccessful. This suggests that the creation of a minimum level of AMT investment might be required toremain competitive.

Conclusions

Summary of findings

This paper has contributed to the operations management literature concerning AMT investmentpatterns and the relationship between the patterns and manufacturing, firm performance, size, andownership structure in the context of the Turkish automotive industry. Our taxonomy reveals threedistinct AMT groups representing differing investment strategies – investors, followers andtraditionalists.

Our results indicate that different clusters of firms, namely investors, followers and traditionalists,hint at different patterns of AMT investments within the Turkish automotive industry, and also do notdiffer with respect to ownership structure and size. Investors and followers where firms have highlevels of investment in AMT perform better in manufacturing performance than the traditionalistswho have the lowest investments in all three AMT variables. However, their firm performance issimilar to that of the other two groups.

The similarities and differences between the technological strategies that emerge in our study andthose of previous studies conducted in various and developed industries clearly show that firms in the

Table 9A summary of the findings of this study in line with previous AMT taxonomies.

1. Traditionalists 2.Followers 3. Investors

This study Design [2,3] [1] [1]

Manufacturing [2,3] [1,3] [1,2]

Administrative [2,3] [1,3] [1,2]

Size (no. employees) No differences No differences No differences

Size (annual sales) No differences No differences No differences

Ownership No differences No differences No differences

Manufacturing performance [2,3] [1] [1]

Profit No differences No differences No differences

Growth No differences No differences No differences

Note: Numbers in brackets indicate other groups with significant differences based on the pairwise comparison tests.


automotive industry act according to global rather than local dynamics. Table 9 presents a tabulatedsummary of the findings of this study, extending the classification presented in Table 1 to the contextof a developing country.

Implications for theory

There are a number of differences between the clusters that emerged in the automotive industrystudied here and those of previous studies. First, the level of investments made by the three clusters isgenerally higher than those of previous studies. AMT investments are at a very high level for theinvestors, and not even low for traditionalists. Second, traditionalists form the smallest cluster in thethree identified in the present study, whereas the same group is observed to be the largest in Boyeret al. (1996) and Diaz et al. (2003). Finally, none of the three AMT variables has come to the fore interms of investment within a particular cluster, i.e., there are differences between clusters but notwithin.

In our view, these differences mainly arise from the innovative and competitive nature of theautomotive industry. A smaller number of traditionalists and a higher level of AMT investments in allclusters support the claim of Dean and Snell (1991) and Swamidass (2003) that the automotiveindustry is one of the industries using AMT most widely. Furthermore, the relatively balanced aspectof investments in design, manufacturing and administrative AMTs of the three groups identified in thisstudy suggests that firms in the automotive industry are willing to benefit from AMTs as much aspossible as regards lower cost, better quality and higher flexibility in order to gain a competitiveadvantage.

Despite the differences noted above, there are some similarities between the technology patternsidentified in the automotive industry and those in other industries. In particular, automotive firmsmake the lowest investments in manufacturing technologies among the three types of AMT variables,similar to those in Boyer et al. (1996) and Jonsson (2000). This feature is understandable; given thatthe already high investment costs of manufacturing technologies is relatively higher still than those ofdesign and administrative technologies. This is an important result in that investment costs of AMTsstill seems to be one of the primary determinants for investment decisions in developing countries aswell.

Another similarity between the generalists of Boyer et al. (1996) and our followers is that bothclusters have significant levels of AMT investments, and they follow, with regard to the investmentlevels and ranks, the higher level investment group. Moreover, the number of firms in both groups issignificantly higher than in the investor group; this similarity shows that there is a large clusterfollowing a small number of leading firms in their technological investment strategies. The lack of anystatistically significant difference in all three types of AMT variables between the followers andinvestors indicates that there is a closer relationship between the two clusters than in other industries.This finding also implies that any lack of technological investments would potentially diminish thetechnological superiority of the investors in the sector, in which case the similarity between investorsand followers would be even greater. These findings parallel those of Boyer et al. (1996).


The results obtained in this study have given way to an AMT investment model that is similar butnot exactly the same as those described in Boyer et al. (1996), Jonsson (2000) and Diaz et al. (2003). Inparticular, findings are similar with the previous studies in that there is the emergence oftraditionalist and investor groups in all taxonomies. This indicates the stability and robustness of theproposed taxonomy, and provides further support for its usefulness. However, whereas previoustaxonomies describe at least one group which comes to the fore with a particular attribute, e.g., thedesigners described in Boyer et al. (1996) and Diaz et al. (2003) and the hard integrators described inJonsson (2000), we have found this not to be the case with the automotive industry. The reasonbehind this finding could be attributed to the innovative and competitive nature of the automotiveindustry, as well as the perception of this very industry of seeing AMT as a whole, rather than as beingsegmented.

Analysis reveals that there is a significant difference in manufacturing performance across differentinvestment patterns; however, this difference is primarily attributed to traditionalist firms. Thedifference in manufacturing performance between the followers and investors is not significant. Thisfinding supports our classification by showing that the manufacturing performance of the followers issimilar to that of the investors, even though the former do not invest in AMT to the extent that thelatter do.

The results of this study suggest that AMT investment patterns are not significantly correlated withfirm performance, as reported in previous studies. This indicates that the impact of AMT use on firmperformance is minimal, and therefore confirms the findings of Boyer et al. (1996), Beaumont andSchroeder (1997) Swamidass and Kotha (1998) and Cagliano and Spina (2000) in a developing country.The weak link between AMT use and firm performance supports the claim of Swamidass and Kotha(1998) that ‘‘strategic rather than financial benefits may be the primary reason for investing in AMTs’’.

The results of this research also indicate that there are no significant differences across AMTinvestment patterns in terms of firm size. This result implies that it is not only large-size firms that areinterested in AMT use, but also those of medium-size, provided they are financially able to invest inthese technologies.

Our results showed that there is no clear relationship between investment patterns of AMT and theownership structure of the firms. Although this result does not parallel those of prior studiesconducted in different industries (Schroder and Sohal, 1999; Robb and Xie, 2001; Beaumont et al.,2002; Pyke et al., 2002), it does parallel the findings of Laosirihongthong et al. (2003) in the automotiveindustry. From these results, we can conclude that AMT investments of domestic firms in theautomotive industry are at the same level as foreign-owned or joint venture investments. It isenvisaged that the use of these technologies will become widespread among the firms in developingcountries, especially as technology costs decrease as a result of developments in technology.

Finally, we note that the literature on AMT taxonomies has recently been deemed inconsistent(Chung and Swink, 2009) with regards to the industry and country from which the data are collected,the number of clusters identified and the factors analysis components. Our study has contributed tothis literature by offering further evidence from a developing country, which we hope will reduce theabove-mentioned inconsistency to arrive at more general AMT utilization patterns.

Table 10 presents a high-level comparison between the results found in this study and those ofrelevant studies.

Implications for practice

The findings of this paper have useful implications for managers of firms in the automotive sectorof a developing country who have not yet adopted AMT but are considering doing so, which are listedbelow:

One fundamental implication is that managers who hesitate to adopt AMT in the automotive sectorof a developing country, for whatever reason, are facing the risk of falling out of competition,particularly with regards to manufacturing performance. In particular, our findings on the relationshipbetween manufacturing performance and AMT investments suggest that, investing in the latter islikely to result in: (i) an increase in manufacturing performance with benefits of having a better market

position and being a stronger competitor in the industry, and (ii) providing a unique attribute of having

Table 10A high-level comparison of previous studies and the current study.

Boyer et al. (1996) Jonsson (2000) Diaz et al. (2003) This study

Nature of the country study

is conducted in

Developed Developed Developed Developing

Type of industry Metal-working Metal-working Aeronautical Automotive

AMT investment groups Traditionalists

Designers

Generalists

High Investors

Traditionalists

Hard Integrators

Investors

Traditionalists

Designers

Investors

Traditionalists

Followers

Investors

Ownership N/A N/A N/A No differences

Size (no. employees) Differences found Differences found N/A No differences

Size (annual sales) N/A N/A Differences found No differences

Manufacturing performance N/A N/A N/A Differences found

Overall performance No differences Differences found No differences No differences

Note: N/A indicates that this particular attribute has not been analyzed in the corresponding study.


invested in a new technology, particularly in a developing country where take-up of AMT would belimited, resulting in increased reputation and customer attraction.

Our findings also suggest that a manager’s decisions to invest in AMT should be independent of the size

and the ownership of the firm. Medium, but not necessarily high, levels of investments in AMT (as in thecase of followers) would suffice to yield operational benefits to help firms remain competitive.Managers, however, should not expect to achieve such improvements in operational performance, inparticular that of manufacturing, without any AMT investment.

The theoretical finding mentioned in the previous section that none of the three AMT variables hascome to the fore in terms of investment within a particular cluster within the automotive industrystudied here indicates that AMT investments should be seen as a whole. In other words, firms in theautomotive industry in a developing country should make balanced investment in all three types ofAMT variables.

Our findings, as with Boyer et al. (1996) and Diaz et al. (2003), do not indicate any significantcorrelation between AMT investment patterns and profit and growth as performance measures. Thisfinding indicates that AMT investments alone will not suffice in increasing the overall performance of theautomotive industry in developing countries. The implication for managers is that this relationship ismore complex than it seems, and may also include other factors such as employee performance.

The effect of learning curves for some AMT technologies should not be dismissed, as the benefits ofusing these technologies will only emerge after they have been in use for some time after some time ofuse (Sun, 2000). More research on this is needed in different sectors and geographical regions to beable to arrive at general conclusions.

Like any other study, this study also has some limitations that suggest directions for futureresearch. The use of a cross-sectional research methodology in the study provides limited longitudinalevidence, and does not show precisely how the development of different types of AMT investmentpatterns and patterns affect manufacturing and overall performance. Thus, with a longitudinalresearch, the subject of how firms make progress between strategic types and how these types affectthe firm performance variables might be investigated in future studies. The extent to whichinstitutional isomorphorism, in particular mimetic but also possibly coercive and normative, isrelevant in AMT adoption in the automotive industry is another interesting question that warrantsfurther research. More specifically, the question to be addressed is whether the decisions on AMTinvestment in the automotive industry are taken rationally or influenced by coercive, mimetic ornormative isomorphism. In this regard, the level of rationality in the AMT investment process and itsrelation to firm performance is another question to be looked into.

Another limitation is the sample used in this research, which is naturally relatively smallconsidering the size of the whole population. In order to overcome these limitations and obtain furtherinsights, future research could look into using samples from multiple countries, including firms ofdifferent sizes and ownership.


Acknowledgement

Thanks are due to two reviewers for their valuable comments. In carrying out this research, thethird author has been supported by the Selcuk University Scientific Research Project Fund (BAP).

References

Adler, P.S., 1988. Managing flexible automation. California Management Review 30 (3), 34–57.Ahmad, S., Schroeder, R.G., 2003. The impact of human resource management practices on operational performance:

recognizing country and industry differences. Journal of Operations Management 21 (1), 19–43.Alcorta, L., 1999. The diffusion of advanced automation in developing countries: factors and adoption process. Technovation 19

(3), 163–175.Beaumont, N.B., Schroeder, R., 1997. Technology, manufacturing performance and business performance amongst Australian

manufacturers. Technovation 17 (6), 297–307.Beaumont, N.B., Schroder, R., Sohal, A., 2002. Do foreign-owned firms manage advanced manufacturing technology better?

International Journal of Operations & Production Management 22 (7), 759–771.Boyer, K.K., Ward, P.T., Leong, G.K., 1996. Approaches to the factory of the future: an empirical taxonomy. Journal of Operations

Management 14 (4), 297–313.Bozarth, C., McDermott, C., 1998. Configurations in manufacturing strategy: a review and directions for future research. Journal

of Operations Management 16 (4), 427–439.Buonanno, G., Faverio, P., Pigni, F., Ravarini, A., Sciuto, D., Tagliavini, M., 2005. Factors affecting ERP system adoption: a

comparative analysis between SMEs and large companies. Journal of Enterprise Information Management 18 (4), 384–426.Burgess, T.F., Gules, H.K., 1998. Buyer–supplier relationships in firms adopting advanced manufacturing technology: an

empirical analysis of the implementation of hard and soft technologies. Journal of Engineering & Technology Management15 (2–3), 127–152.

Cagliano, R., Spina, G., 2000. Advanced manufacturing technologies and strategically flexible production. Journal of OperationsManagement 18 (2), 169–190.

Chamarbagwala, R., Ramaswamy, S., Wunnava, P.V., 2000. The role of foreign capital in domestic manufacturing productivity:empirical evidence from Asian economies. Applied Economics 32 (4), 393–398.

Choe, J.M., 2004. The consideration of cultural differences in the design of information systems. Information & Management 41(5), 669–684.

Choi, T.Y., Eboch, K., 1998. The TQM paradox: relations among TQM practices, plant performance, and customer satisfaction.Journal of Operations Management 17 (1), 59–75.

Chung, C.A., 1996. Human issues influencing the successful implementation of advanced manufacturing technology. Journal ofEngineering & Technology Management 13 (3–4), 283–299.

Chung, W., Swink, M., 2009. Patterns of advanced manufacturing technology utilization and manufacturing capabilities.Production and Operations Management 18 (5), 533–545.

Curkovic, S., Vickery, S., Droge, C., 2000. Quality-related action programs: their impact on quality performance and firmperformance. Decision Sciences 31 (4), 885–905.

Dangayach, G.S., Deshmukh, S.G., 2006. An exploratory study of manufacturing strategy practices of machinery manufacturingcompanies in India. Omega: The International Journal of Management Science 34 (3), 254–273.

Das, A., Jayaram, J., 2003. Relative importance of contingency variables for advanced manufacturing technology. InternationalJournal of Production Research 41 (18), 4429–4452.

de Jong, J.P.J., Marsili, O., 2006. The fruit flies of innovations: a taxonomy of innovative small firms. Research Policy 35 (2), 213–229.

Dean Jr., J.W., Snell, S.A., 1991. Integrated manufacturing and job design: moderating effects of organizational inertia. AcademicManagement Journal 34 (4), 776–804.

Dean Jr., J.W., Snell, S.A., 1996. The strategic use of integrated manufacturing: an empirical examination. Strategic ManagementJournal 17 (6), 459–480.

Dean Jr., J.W., Susman, G.I., Porter, P.S., 1990. Technical, economic and political factors in advanced manufacturing technologyimplementation. Journal of Engineering & Technology Management 7 (2), 129–144.

Diaz, M.S., Machuca, J.A.D., Alvarez-Gil, M.J., 2003. A view of developing patterns of investment in AMT through empiricaltaxonomies: new evidence. Journal of Operations Management 21 (5), 577–606.

DiMaggio, P.J., Powell, W., 1983. The iron cage revisited: institutional isomorphism and collective rationality in organizationalfields. American Sociological Review 48 (2), 147–160.

Douma, S., George, R., Kabir, R., 2006. Foreign and domestic ownership, business groups, and firm performance: evidence from alarge emerging market. Strategic Management Journal 27 (7), 637–657.

Fawcett, S.E., Myers, M.B., 2001. Product and employee development in advanced manufacturing implementation and impact.International Journal of Production Research 39 (1), 65–79.

Flynn, B.B., Sakakibara, S., Schroeder, R.G., Bates, K.A., Flynn, E.J., 1990. Empirical research methods in operations management.Journal of Operations Management 9 (2), 250–284.

Frohlich, M.T., Dixon, J.R., 2001. A taxonomy of manufacturing strategies revisited. Journal of Operations Management 19 (5),541–558.

Gerwin, D., Kolodny, H., 1992. Management of Advanced Manufacturing Technology. John Wiley & Sons, New York.Ghani, K.A., Jayabalan, V., 2000. Advanced manufacturing technology and planned organizational change. The Journal of High

Technology Management Research 11 (1), 1–18.Gordon, J., Sohal, A.S., 2001. Assessing manufacturing plant competitiveness: an empirical field study. International Journal of

Operations & Production Management 21 (1/2), 233–253.


Gupta, A., Chen, I.J., Chiang, D., 1997. Determining organizational structure choices in advanced manufacturing technologymanagement. Omega: The International Journal of Management Science 25 (5), 511–521.

Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis. Prentice-Hall, Englewood Cliffs, NJ.Hayes, R.H., Wheelwright, S.C., 1979. Link manufacturing process and product life cycles. Harvard Business Review 133–140

January/February.Hayes, R.H., Wheelwright, S.C., 1984. Restoring Our Competitive Edge: Competing through Manufacturing. John Wiley & Sons,

New York.Heinea, M.L., Grover, V., Malhotra, M.K., 2003. The relationship between technology and performance: a meta-analysis of

technology models. Omega: The International Journal of Management Science 31 (3), 189–204.Hottenstein, M.P., Casey, M.S., Dunn, S.C., 1999. The diffusion of advanced manufacturing technology in multiplant, multi-

divisional corporations. Journal of Engineering & Technology Management 16 (2), 129–146.ISO, 2006. Turkey’s top 500 industrial enterprises 2005 and Turkey’s second 500 industrial enterprises 2005. Journal of the

Istanbul Chamber of Industry 41, 485–486.Jonsson, P., 2000. An empirical taxonomy of advanced manufacturing technology. International Journal of Operations &

Production Management 20 (12), 1446–1474.Kathuria, R., 2000. Competitive priorities and managerial performance: a taxonomy of small manufacturers. Journal of

Operations Management 18 (6), 627–641.Ketchen Jr., D.J., Shook, C.L., 1996. The application of cluster analysis in strategic management research: an analysis and critique.

Strategic Management Journal 17 (6), 441–458.Khazanchi, S., Lewis, M.W., Boyer, K.K., 2007. Innovation-supportive culture: the impact of organizational values on process

innovation. Journal of Operations Management 25 (4), 871–884.Koste, L.L., Malhotra, M.K., 2000. Trade-offs among the elements of flexibility: a comparison from the automotive industry.

Omega: The International Journal of Management Science 28 (6), 693–710.Kotha, S., Swamidass, P.M., 1998. Advanced manufacturing technology uses: exploring the effect of the nationality variable.

International Journal of Production Research 36 (11), 3135–3146.Kotha, S., Swamidass, P.M., 2000. Strategy, advanced manufacturing technology and performance: empirical evidence from U.S.

manufacturing firms. Journal of Operations Management 18 (3), 257–277.Krajewski, L.J., Ritzman, L.P., 2001. Operations Management: Strategy and Analysis. Prentice Hall, New Jersey.Lambert, D.M., Harrington, T.C., 1990. Measuring nonresponse bias in customer service mail surveys. Journal of Business

Logistics 11 (2), 5–25.Laosirihongthong, T., Paul, H., Speece, M.W., 2003. Evaluation of new manufacturing technology implementation: an empirical

study in the Thai automotive industry. Technovation 23 (4), 321–331.Lefebvre, L., Lefebvre, E., Harvey, J., 1996. Intangible assets as determinants of advanced manufacturing technology adoption in

SME’s: toward an evolutionary model. IEEE Transaction Engineering Management 43 (3), 307–322.Lehmann, D.R., 1979. Market Research and Analysis. Irwin, Homewood, IL.Lewis, M.W., Boyer, K.K., 2002. Factors impacting AMT implementation: an integrative and controlled study. Journal of

Engineering & Technology Management 19 (2), 111–130.Mabert, V.M., Soni, A.K., Venkataramanan, M.A., 2000. Enterprise resource planning survey of US manufacturing firms.

Production & Inventory Management Journal 41 (20), 52–58.MacDougall, S.L., Pike, R.H., 2003. Consider your options: changes to strategic value during implementation of advanced

manufacturing technology. Omega: The International Journal of Management Science 31 (1), 1–15.Markusen, J.R., 2002. Multinational Firms and the Theory of International Trade. MIT Press, Cambridge, MA.Meredith, J.R., 1987. The strategic advantages of new manufacturing technologies for small firms. Strategic Management Journal

8 (3), 249–258.Miller, J.G., Roth, A.V., 1994. A taxonomy of manufacturing strategies. Management Science 40 (3), 285–304.Narasimhan, R., Das, A., 2001. The impact of purchasing integration and practices on manufacturing performance. Journal of

Operations Management 19 (5), 593–609.Nunnally, J.C., 1978. Psychometric Theory. McGraw-Hill, New York.Ordoobadi, S.M., Mulvaney, N.J., 2001. Development of a justification tool for advanced manufacturing technologies: system-

wide benefits value analysis. Journal of Engineering & Technology Management 18 (2), 157–184.OSD. Official Web Site for the Turkish Automotive Manufacturers Association. http://www.osd.org.tr.Powell, T.C., Dent-Micallef, A., 1997. Information technology as competitive advantage: the role of human, business, and

technology resources. Strategic Management Journal 18 (5), 375–405.Prasad, S., Tata, J., Madan, M., 2005. Build to order supply chains in developed and developing countries. Journal of Operations

Management 23 (5), 551–568.Putranto, K., Stewart, D., Moore, G., Diatmoko, R., 2003. Implementing a technology strategy in developing countries: the

experience of the Indonesian rolling stock industry. Technological Forecasting & Social Change 70, 163–176.Pyke, D., Farley, J., Robb, D., 2002. Manufacturing technology and operations in China: a survey of state-owned enterprises,

private firms, joint Ventures and wholly-owned foreign subsidiaries. European Management Journal 20 (4), 356–375.Raymond, L., St-Pierre, J., 2005. Antecedents and performance outcomes of advanced manufacturing systems sophistication in

SMEs. International Journal of Operations & Production Management 25 (6), 514–533.Robb, D.J., Xie, B., 2001. A survey of operations strategy in China-based manufacturing enterprises. International Journal of

Production Economics 72 (2), 181–199.Roth, A., Miller, J.G., 1990. Manufacturing strategy, manufacturing strength, managerial success and economic outcomes. In:

Ettlie, J.E., Burnstein, M.C., Feigenbaum, A. (Eds.), Manufacturing Strategy: the Research Agenda for the Next Decade.Kluwer Academic Publishers, Boston, MA.

Safizadeh, M.H., Ritzman, L.P., Sharma, D., Wood, C.H., 1996. An empirical analysis of the product-process matrix. ManagementScience 42 (11), 1576–1591.

Salaheldin, I.S., 2007. The impact of organizational characteristics on AMT adoption: a study of Egyptian manufacturers. Journalof Manufacturing Technology Management 18 (4), 443–460.

http://www.osd.org.tr/


Schroder, R., Sohal, A., 1999. Organizational characteristics associated with AMT adoption. International Journal of Operations &Production Management 19 (12), 1270–1291.

Segarra, G., 1999. The advanced information technology innovation roadmap. Computers in Industry 40 (2–3), 185–195.Small, M.H., Chen, I.J., 1995. Investment justification of advanced manufacturing technology: an empirical analysis. Journal of

Engineering & Technology Management 12 (1–2), 27–55.Small, M.H., Chen, I.J., 1997. Economic and strategic justification of AMT inferences from industrial practices. International

Journal of Production Economics 49 (1), 65–75.Sohal, A.S., Schroder, R., Uliana, E.O., Maguire, W., 2001. Adoption of AMT by South African manufacturers. Integrated

Manufacturing Systems 12 (2), 15–34.Sohal, A.S., Sarros, J., Schroder, R., O’neill, P., 2006. Adoption framework for advanced manufacturing technologies. International

Journal of Production Research 44 (24), 5225–5246.Sun, H., 2000. Current and future patterns of using advanced manufacturing technologies. Technovation 20 (11), 631–641.Sun, H., Tian, Y., Cui, H., 2001. Evaluating advanced manufacturing technology in Chinese state-owned enterprises: a survey and

case studies. International Journal of Advanced Manufacturing Technology 18 (7), 528–536.Swamidass, P.M., 2003. Modeling the adoption rates of manufacturing technology innovations by small US manufacturers: a

longitudinal investigation. Research Policy 32 (3), 351–366.Swamidass, P.M., Kotha, S., 1998. Explaining manufacturing technology use, firm size and performance using a multidimen-

sional view of technology. Journal of Operations Management 17 (1), 23–37.Swamidass, P.M., Newell, W.T., 1987. Manufacturing strategy, environmental uncertainty and performance: a path analytic

model. Management Science 33 (4), 509–524.Swink, M., Nair, A., 2007. Capturing the competitive advantages of AMT: design–manufacturing integration as a complementary

asset. Journal of Operations Management 25 (3), 736–754.TAYSAD. Official Web Site for the Turkish Automotive Parts and Component Manufacturers Association. http://www.taysad.

org.tr.Thong, J., Yap, C., 1995. CEO characteristics, organizational characteristics and information technology adoption in small

businesses. Omega: The International Journal of Management Science 23 (4), 429–442.Tracey, M., Vonderembse, M.A., Lim, J.S., 1999. Manufacturing technology and strategy formulation: keys to enhancing

competitiveness and improving performance. Journal of Operations Management 17 (4), 411–428.Venkatraman, N., Ramanujam, V., 1987. Measurement of business economic performance: an examination of method

convergence. Journal of Management 13 (1), 109–122.Vickery, S.K., Droge, C., Markland, R.E., 1993. Production competence and business strategy: do they affect business perfor-

mance. Decision Sciences 24 (2), 435–455.Vickery, S.K., Droge, C., Markland, R.E., 1997. Dimensions of manufacturing strength in the furniture industry. Journal of

Operations Management 15 (4), 317–330.Ward Jr., J.H., 1963. Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association 58

(301), 236–244.Ward, P.T., Duray, R., 2000. Manufacturing strategy in context: environment, competitive strategy and manufacturing strategy.

Journal of Operations Management 18 (2), 123–138.Wasti, S.N., Kozan, M.K., Kuman, A., 2005. Buyer-supplier relationships in the Turkish automotive industry. International Journal

of Operations & Production Management 26 (9), 947–970.Zairi, M., 1993. Competitive manufacturing: combining total quality with advanced technology. Long Range Planning 26 (3),

123–132.Zammuto, R.F., O’Connor, E.J., 1992. Gaining advanced manufacturing technologies’ benefits: the roles of organization design

and culture. Academy of Management Review 17 (4), 701–728.Zhang, M.J., 2005. Information systems, strategic flexibility and firm performance: an empirical investigation. Journal of

Engineering & Technology Management 22 (3), 163–184.Zhao, X., Sum, C.C., Qi, Y., Zhang, H., Lee, T.S., 2006. A taxonomy of manufacturing strategies in China. Journal of Operations

Management 24 (5), 621–636.

http://www.taysad.org.tr/

http://www.taysad.org.tr/

Documents

An empirical investigation of advanced manufacturing technology investment patterns: Evidence from a developing country