Lean production, market share and value creation in ...

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1741-038X.htm

Lean production, market share and value creation in the agricultural machinery sector in Brazil Paul L. Forrester Birmingham Business School, The University of Birmingham, Birmingham, UK

Lean production in Brazil

853 Received July 2009 Revised February 2010 Accepted March 2010

Ullisses Kazumi Shimizu Mackenzie Presbyterian University, Saõ Paulo, Brazil

Horacio Soriano-Meier Northampton Business School, The University of Northampton, Northampton, UK

Jose Arturo Garza-Reyes School of Technology, The University of Derby, Derby, UK, and

Leonardo Fernando Cruz Basso Business School-Graduate Department, Mackenzie Presbyterian University, Saõ Paulo, Brazil Abstract Purpose – The “resource-based view” (RBV) of firms considers that major operational and organisational advantages are created in the internal environment of a firm. The implementation of lean manufacturing represents the potential for strategic advantage over competitors, especially in craft-based industries in developing regions of the world. The purpose of this paper is to investigate the relationship between the adoption of lean manufacturing and market share and value creation of companies in the agricultural machinery and implements sector in Brazil. Design/methodology/approach – The paper is based on data collected in a survey conducted across 37 firms in the agricultural machinery and implements industry in Brazil. The data were used within a model for assessing the degree of leanness to test three hypotheses using correlation, regression, analysis of variance and cluster statistical methods. Findings – Brazilian firms and managers in this sector that have supported a transition towards the adoption (and adaptation) of lean manufacturing practices have shown a significant improvement in their business performance. Originality/value – The paper presents an empirical study where lean manufacturing is investigated and tested from a “RBV” perspective. It demonstrates the application of an emergent model for measuring the degree of leanness and the extent of business improvement. The study and the model are applied to smaller, craft-based industries and so is applicable in developing countries and regions, in comparison with most literature on lean production in advanced economies. It provides a useful perspective for firms to corroborate and understand the potential benefits that lean manufacturing can bring if adopted. Keywords Lean production, Craft production, Brazil, Resource management, Market share Paper type Research paper

Journal of Manufacturing Technology Management Vol. 21 No. 7, 2010 pp. 853-871 q Emerald Group Publishing Limited 1741-038X DOI 10.1108/17410381011077955

JMTM 21,7

854

1. Introduction One of the main objectives in strategy formulation and implementation is the creation of sustainable advantages for firms. De Oliveira and Fensterseifer (2003) argue the need for understanding why some firms perform better than others which operate in similar market and competitive situations. These differences in performance may be attributed to a differentiation in internal factors such as knowledge and other strategic assets, which have an impact on the firm’s overall performance. This thinking is embodied in the “resource-based view” (RBV) approach which considers firms as different amalgams of productive and strategic resources and capabilities that lead them to different performance potentials. The “RBV” has gained prominence in the strategy literature by emphasizing the firm’s internal resources as the main determinants for improved performance. However, despite the global appeal and attractiveness of this view, empirical studies to prove its worth are still in short supply in many regions of the world, including Brazil. This paper is based on data collected from an in-depth survey and interviews in Brazil but has application beyond. It adds to our knowledge on lean manufacturing and performance by demonstrating that adaptations to the basic lean model are necessary to match the context of the organisation and industry sector. Furthermore, the paper demonstrates the application and applicability of the Soriano-Meier and Forrester (2002) model for measuring the degree of leanness and lean performance to a further context, beyond the original UK study. It is shown that this model offers a generic tool for measuring degree of leanness, degree of commitment of managers and linking these to business performance. The economic success attained by companies engaged in lean production programmes has boosted interest in the understanding and adoption of “lean” by Brazilian companies. Lean is defined by NIST (2000) as a systematic approach to identify and eliminate waste through continuous improvement (CI), a demand-pull flow of materials and the pursuit of error-free processes. Although the origins of lean thinking can be found on the shop floor of Japanese manufactures and, in particular, innovations at the Toyota Motor Corporation (Hines et al., 2004), it was the publication of the book The Machine that Changed the World (Womack et al., 1990) that popularised the method and the lean terminology. The application of lean manufacturing concepts and practices, as well as their application in key business processes and industry sectors, has grown and evolved since 1990 and is now accepted generally as “best practice” for manufacturing in developed industries of the world. But how generic is the application of lean? Can it be readily applied to advantage in smaller, craft-based industries in more remote and developing regions of the world? This paper argues that it can. It tests the relationship between the adoption of lean production programmes and the market share and value creation of companies in the agricultural machinery and implements manufacturing sector in Brazil. This sector was chosen for its growing strategic importance in the country. According to the Metalworking Insider Report (2004), the agricultural machinery and implements sector of Brazil ranked 15th in the world in 2004, with production levels similar to those of Austria and The Netherlands. Brazil is fast becoming a major world grain and animal products provider, which has attracted multinational companies in the food sector and therefore demonstrates growth in the Brazilian agriculture machinery sector. The paper argues that the capacity to perform lean production is a key resource for the creation of a competitive advantage. The constructs relating to the degree of leanness

were extrapolated from a study by Soriano-Meier and Forrester (2002). Four value creation measures of profitability (value creation rates) and measures of sum (sundry accounting measurements for profit) were used. In addition, the paper features a theoretical benchmark based on the RBV which emphasizes the attributes a resource requires in order to create a sustainable competitive advantage. This constitutes the theoretical grounds to measure the degree of leanness and to test the research hypotheses from the RBV. 2. Theoretical benchmark: RBV and the degree of leanness The lean manufacturing model associates advantage in productive performance with adherence to three key principles: (1) Improvement in the flow of materials and information through operations. (2) The emphasis on demand (or usage) pull of materials and products, in place of pushing inventories through the system (triggering production by kanban). (3) The commitment to CI facilitated by people development. All these grounds indicate that lean manufacturing is a production process that cannot be easily or immediately imitated. Interestingly, the original study of International Motor Vehicle Programme (IMVP) was largely influenced by Toyota and the work of Ohno (1988). It took Ohno ten years after his retirement to write the book that effectively described the productive process of Toyota in a coherent and replicable way. Social scientists, engineers and consultants sought a systematic explanation of the success of Toyota (Womack et al., 1990; Monden, 1983; Goddard, 1986; Harrison, 1992; Cheng and Podolsky, 1996). This resulted in a number of “deconstructions” of the lean system. It was recognised that any system or form of production organisation must take into account the history and context of the region in which it was designed. Lean manufacturing, born in Japan fathered by Taiichi Ohno, is to a degree culturally bound and needed some adaptation, but its principles have generic application and were even partially in existence before Ohno’s system. For example, innovations in operations acclaimed as the work of Toyota (Ohno, 1988) were already being used by Ford in 1920 (see Williams et al., 1994, for a review of some myths of lean manufacturing). Lean evolved in a practical manner, and for some years, academics “chased” the work of the practitioners in a quest to theorise and model lean principles. The RBV has been a major vehicle in accomplishing this. Barney (1991) identifies that resources are sources of sustainable competitive advantage if they are as follows: . Valuable. Valuable meaning the resource that enables the company to maintain or implement a strategy that increases its effectiveness or efficiency. . Scant or rare. It is intuitive that if a company possesses a resource that is absolutely rare or scanty among its competitors, this resource will generate competitive advantage in relation to its rivals and will also have the possibility of becoming a sustainable competitive advantage for the company. . Imperfectly imitable. Meaning other competitors cannot replicate as they lack relevant and key resources that comprise the strategy. Resources are imperfectly imitable when the company’s ability to obtain the resource depends solely on historic conditions, the connection between the ownership of the


855

JMTM 21,7

856

resource and the sustainable competitive advantage is ambiguous as far as the cause of the advantage is concerned (i.e. casual ambiguity exists when the connection between the resources controlled by the company and the competitive advantage is not understood) and the resource that generates sustainable competitive advantage is socially complex and depends on the relationship between people and processes, for example, interpersonal relationships between and among managers, the company culture and reputation among suppliers and customers. Barney (1991) also identified another characteristic of a resource, imperfectly substitutable, that provides the company with a competitive advantage where there is no equivalent resource that enables the rival to implement a similar strategy. Grant (1991) observes that the internal resources and capabilities of the organisation represent a more stable and secure basis for the formulation of competitive strategies. Resources and organisational capabilities should be the bases for the definition of competitive strategies sustainable over the long term, as they are more efficient than strategies sustained by external factors. Therefore, companies should be well coordinated internally and aware that the level of resources (physical, financial and technological media and the reputation of the company) and the organisational climate (organisational skills) will define the company’s capacity to adapt to changes in the standard of competition. These are clearly key concerns in the case of lean production. In the language of traditional strategic analysis, company resources are forces that can be used to design and implement strategies (Learned et al., 1969; Porter, 1981). According to Collis and Montgomery (1997), resources can come in various shapes, ranging from the most common, widely available and easily purchasable factor to the most differentiated resource such as brand, which should be developed for many years and is very hard to imitate. The same authors propose the classification of resources in three categories: tangible assets, intangible assets and organisational capabilities. It appears that the latter two are particularly to the fore in the adoption of lean production. Intangible assets offer the basis for diversified expansion as they are difficult to replicate. Applied to production, organisational capability is what governs the efficiency of the company’s activities. If well developed, the organisational capabilities can be sources of competitive advantage, as they enable the organisation to convert inputs (whether services or products that result in greater efficiency or quality) into outputs, such as lean manufacturing, more efficiently than its rivals. The use of the resources and capabilities of the firm as the foundation for a long-term strategy is based on two assumptions. First, internal resources and capabilities provide the basic direction for the firm’s strategy. Second, resources and capabilities are the primary origin of profit for the firm (Grant, 1991). Extending this to production strategies, Lewis (2000) illustrates how the dynamics of sustainable competitive advantage works (Figure 1). Resources can be strategic if inimitable by rivals. Firms do not need to own strategic resources as they can belong to their suppliers. Likewise, value creation processes go beyond the borders of the firm, involving current and potential customers. Significant results are attained when they improve the firm’s performance and differentiate it from its competitors. Lewis (2000) proposes that the success of lean manufacturing in affording sustainable competitive advantage is contingent to the external context of the firm. Contextual factors can include type of market (competition activity, profile of different demands), dominant technology in the sector and productive chain structure.


Lean production is characterized by a reduced level of input of resources for a certain level of outputs. This is attained with the elimination of waste of the system. The elimination of waste of resources (materials, invertories) takes place first, but it also includes the transformation of other resources such as people, process technology and facilities

Inputs

Outputs The firm

Suppliers

Flow: after the definition of the value stream, lean production emphasizes the production activity as a direct flow to the consumer. This process reveals the points of waste and has significant implication on the structure of the organization

857

Customers

Pulled production: The lean production concept is when the customer pulls the product (precisely when necessary) through the operation internally, this calls for new work methods, such as production cells, karbon controls etc., and there are also profound changes in the production chain, as it requires suppliers to work with just-in-time (JTT), small production batches, quality certification, etc.

Source: Lewis (2000)

Evidence suggests the need for more extensive contingency-related studies of the lean manufacturing model. Katayama and Bennett (1996) have claimed that lean manufacturing is incapable of responding to major oscillations of added demands of volume, arguing that, at the time of the IMVP study, the Japanese economy exhibited highly specific characteristics that created conditions for a high and stable domestic demand. The sustainable competitive advantage model argues that resources (trained staff, market information, technological data, etc.) create value when they integrate and act in processes. Sometimes these processes allow the company to learn and thus to create new resources. Learning in organisations occur in countless ways (Huber, 1991), but it is useful to distinguish between the following: . the development of progressive efficiency and reliable routines; and . progressive improvement and step changes when facing new situations (Sitkin, 1991). The first (progressive efficiency and reliable routine) runs to the heart of the lean manufacturing model, with its emphasis on perfection through CI. When allied with the concepts of value stream efficiency and pull production, lean manufacturing suggests a model of information and flow of material that is highly organised and adopted and adapted over time. If the environment of the firm is stable and changes slowly over time, competitive advantage can become sustainable by means of this adaptation (Lewis, 2000).

Figure 1. The internal dynamics of sustainable competitive advantage

JMTM 21,7

858

Another finding presented by Lewis (2000) is that the greater the success of the firm in the principles of lean manufacturing, the lower the degree of engagement for innovative or transforming activities, as the emphasis of the programme is on CI (kaizen), understood to be incremental and not revolutionary or transforming. There has been much attention in the literature on the benefits lean manufacturing can offer to companies (Soriano-Meier and Forrester, 2002; Lewis, 2000; Karlsson and Ahlstrom, 1996) and how to measure these results. According to the model proposed by Karlsson and Ahlstrom (1996), in order to determine whether a firm is lean, in transition or still works according to traditional models, it is necessary to measure the progress made from an earlier point in time. It is important to distinguish between the determinants and performance of lean manufacturing. The objectives of lean manufacturing are an improvement in productivity, attain an adequate level of quality and reduce production times, cut costs and others. The determinants of lean manufacturing are the actions taken, the principles implemented and the changes made in the organisation to attain the desired performance. According to Karlsson and Ahlstrom (1996), the notion of progress is important in the lean concept. This view is supported by Soriano-Meier and Forrester (2002), who do not see lean manufacturing as a short-term technical tool but rather a long-term strategic initiative. This model can be used as a tool to keep track of progress in the direction of lean manufacturing. It can provide answers to the important questions: are the actions we take leading us in the direction of lean manufacturing? And what progress are we making in relation to the different variables? 3. Creating and measuring value Value creation is one of the touchiest topics in the economic theory, as there is no consensus among various economic paradigms. To Marxists, the creation of value is the product of labour, while to neo-classicists, it is the product of utility. RBV focuses on resources (with the properties mentioned by Barney (1991)) as value creators. Our concern here will be with the measurement of value creation. Young and O’Byrne (2003) divide the measures into five categories: measures of residual profit, measures of the components of residual profit, market-based measures, cash flow measures and traditional profit measures. They argue that residual profit measures take into consideration the cost of “own” and “third-party” capital, unlike the market-based measures where the advantage lies in the incorporation of the market expectation. Yet, these are restricted to listed companies. However, residual profit ratios, such as economic value added, are not as limited as the generally accepted accounting principles, making them ratios that generate numbers with appropriate economic significance. On the other hand, they require complex adjustments, making them limited in calculation. Measures based on the components of residual profit, although more direct in their determination, can only be calculated at the level of sector, division or business unit. Below these levels, apportionments and allocations are necessary, which is why their results are not precise. Cash flow measures are easier to calculate and offer the advantage of associating performance with the business’s capacity to generate cash flow. Yet, in an analysis carried out by Young and O’Byrne (2003), it was verified that the calculation for these ratios requires accounting adjustments, which make them complex and arduous. Traditional measures of profit have the advantage of already being available in the required financial reports, and in spite of their weak points, are accompanied by the market and are well known and widely used, as it is the case of

earnings per share. One of the main negative aspects is that they do not consider the costs of own capital. Soriano-Meier and Forrester’s (2002) study used two economic-financial ratios: sales by employee and turnover of the asset. These ratios were used in the Brazilian study contained in this paper, but it is recognised that these are limited as measures of value creation. The first ratio is a market share ratio and market share does not necessarily lead to greater profitability. For example, a company can increase its market share by acquiring similar companies with old plants. Old facilities imply a slump in productivity and an increase of costs, which can reduce the profitability. On the other hand, it is possible that greater market share is being obtained via competition over prices, which implies a profitability downslide. To overcome these limitations in the Brazilian study, we therefore opted to work with measures of value creation rates (sundry profitability concepts, expressed in terms of a flow by an inventory) and of sums (expressed in value streams), which measure value creation. The measures are included in the category of traditional measures (accounting) of profit and profitability. These specific measures used in this study are presented in Tables I and II.


859

Model and hypotheses The Soriano-Meier and Forrester (2002) model, shown in Figure 2, tests a set of hypotheses with respect to lean production. The model refers to the performance of senior management in the transformation of the firm through the adoption of lean manufacturing. On one side, there is the commitment of senior management to investing in manufacturing infrastructure through quality leadership (QLEAD), the formation of problem-solving teams, training and empowerment, which means the release of decision-making power to levels below. On the other side, there is continuous change through the implementation of lean manufacturing concepts including elimination of waste (EW), CI, just-in-time ( JIT) deliveries, pulling of materials, multifunctional teams (MFT), decentralization of responsibilities, integration of functions (IF) and vertical information system (VIS). With the implementation of these two groups of initiatives, the company becomes leaner and improves its performance. The study aimed to corroborate three hypotheses using as a base a group of 37 Brazilian firms of the agricultural machinery and implements sector. For this, two questionnaires were designed and given to the firms. One of the questionnaires was addressed to the general manager (chief executive officer (CEO), president,

Description

Variable

Growth Growth Growth Growth Growth Growth Growth Growth

ratioturnov rationetass ratiovendas ratiorecoper ratioresultoper ratioresultair ratioresultexerc ratioasstur

of turnover between 2000 and 2003 of total assets between 2000 and 2003 of sales between 2000 and 2003 of operating revenue between 2000 and 2003 of operating result between 2000 and 2003 of result before income tax between 2000 and 2003 of result for the year between 2000 and 2003 of asset turnover between 2000 and 2003

Source: Authors

Table I. Measures of flow

JMTM 21,7

860

Table II. Measures of value

Description

Variable

Growth of turnover value (turnover 2003/total asset 2003) 2 turnover 2000/total asset 2000)/(turnover 2003/total asset 2003) Growth of sales value (sales 2003/total assets 2003) 2 (sales 2000/total asset 2000)/(sales 2003/total asset 2003) Growth of operating revenue value (recoper 2003/total asset 2003) 2 (recoper 2000/total asset 2000)/(recoper 2003/total asset 2003) Growth of operating result value (resultoper 2003/total asset 2003) 2 (resultoper 2000/total asset 2000)/(resultoper 2003/total asset 2003) Growth of value of result before income tax (resultair 2003/total asset 2003) 2 (resultair 2000 2 total asset 2000)/(resultair 2003/total asset 2003) Growth of value of result for the year (resultexerc 2003/total asset 2003) 2 (resultexerc 2000/total asset 2000)/(resultexerc 2003 – total asset 2003) Growth of value of operating result (resultoper 2003/sales 2003) 2 (resultoper 2000/sales 2000)/(resultoper 2003/sales 2000) Growth of value of result before income tax (resultair 2003/sales 2003) 2 (resultair 2000/sales 2000)/(resultair 2003/sales 2003) Growth of value of results for the year (resultexerc 2003/sales 2003) 2 (resultexerc 2000/sales 2000)/(resultexerc 2003/sales 2003)

turnover

recoper resultoper resultair resultexerc resultopervendas resultairvendas resultexercvendas

Source: Authors

Continuous change towards lean manufacturing 1. Elimination of waste 2. Continuous improvement 3. Zero defect 4. Just-in-time deliveries 5. Pulling of materials 6. Multifunctional teams 7. Decentralization 8. Integration of functions 9. Vertical information system

Figure 2. Model 4-C

sales

Management

Continuous management commitment towards lean manufacturing a. Quality leadership b. Problem-solving teams c. Training d. Empowerment

The company becomes “leaner” and improves its performance Source: Soriano-Meier and Forrester (2002)

managing director, etc.) and the second to operations managers (plant manager, operations director, etc.). The hypotheses tested were as follows: H1. Companies that say they have a high management degree of commitment to lean manufacturing (measured by commitment to JIT and total quality management (TQM) programme) simultaneously demonstrate this commitment to investments in manufacturing infrastructure (structural manufacturing infrastructure (SMI) – human infrastructure that provides

support to manufacturing), measured by QLEAD; problem-solving groups (GROUP); training (TRAIN) and worker empowerment (WEMP). H2. Companies that allegedly adopt lean manufacturing principles (measured by degree of adoption (DOA)) have made changes in the direction of these principles (measured by EW, CI, zero defects (ZDs), JIT deliveries, pulling of material, MFT, decentralization, IF and VIS). H3. Companies that continuously made investments in the plant infrastructure (SMI) and changes in the direction of lean manufacturing principles are lean companies, and therefore have better performances than those in transition and those still working in the traditional model. The three hypotheses are needed to examine the conceptual structure developed in this study. H1 evaluates the first component of the 4-C model, the degree of management commitment to lean manufacturing. H2 analyses the degree of changes made in the direction of lean manufacturing. H3 associates the first two components, commitment to lean manufacturing and changes in the direction of lean manufacturing, with performance. Dependent variables The dependent variables tested in the model are commitment to JIT, commitment to total quality management, DOA of the principles of lean manufacturing and performance (this variable covers both the variables of market share in Basso et al. (2006) and the variables of value creation). Independent variables The independent variables tested in the model are QLEAD, problem-solving group, training, empowerment, EW, CI, ZDs, JIT, pulled instead of pushed, MFT, decentralization of responsibilities, integrated functions and vertical information system. 5. Analyses and results – testing the hypotheses H1. Management commitment and investment in manufacturing infrastructure. Correlation and multiple regressions were applied to test H1. The hypothesis verifies the relationship between the level of management commitment to the lean manufacturing programme (measured by commitment to JIT “Com-to-JIT” and commitment to the TQM programme “Com-to-TQM”) and the level of investments made in the manufacturing infrastructure (measured by “QLEAD”, problem-solving group “GROUP”, training “TRAIN” and empowerment “WEMP”). The CEOs of the firms surveyed confirmed that 65 per cent (24 out of 37) have an average or low level of commitment to the JIT programme. Of these 24, ten (27 per cent) reported an average level of commitment. One of the possible causes for this low commitment is that the sector is still in the initial stage of implementation of the lean manufacturing programme, since as verified during the fact finding, many of the companies that started the journey in the direction of lean manufacturing began in around the year 2000, when strong incentive was provided by the sectoral chamber of agricultural machinery and implements for companies from the sector to hear about the programme. This characteristic could be distorting the vision of the CEO when it is asked whether there is a commitment to JIT.


861

JMTM 21,7

Similar to the commitment to JIT, the commitment to TQM is not high: 62 per cent (23 out of 37) declared that the degree of commitment is low, and of these 23, seven (19 per cent) attribute an average level of commitment in relation to TQM. Among senior managers, there is an understanding that the programme is still taking its first steps and that there is still much to be accomplished until to reach a threshold where there is a commitment to JIT and to TQM.

862 Correlation analysis Correlation among variables is a ratio that ranges from 2 1 to þ 1, indicating the power of association among them. When positive, between 0 and þ 1, it indicates that when a variable increases, the other accompanies the increase in the proportion calculated. If the correlation is negative, between 2 1 and 0, this indicates that when a variable increases, the order decreases in the proportion indicated by the value calculated. Therefore, values close to 2 1 and þ 1 indicate that there is a strong relation among the variables analysed, while values close to 0 show a lesser degree of correlation. Soriano-Meier and Forrester (2002) emphasized that lean manufacturing can only be obtained with time and is not a tool for short-term problem solving. It is a strategic tool for improving resource utilisation and growing competitiveness. According to Liker (2004), knowledge about lean manufacturing at Toyota was gradually accumulated by workers and managers over 20 years in their daily activities through the constant learning of new methods and variations of methods consolidated on the plant floor. There was no blueprint or documentation of “lean” as a theory upfront. The tacit knowledge acquired was then passed on to the other production units of Toyota and subsequently also to suppliers. Sirkin (1999) perceives that correlations in social surveys found above 0.70 are rare and many tend to be in the range 0.0-0.50. As it can be seen in Table III, according to this statement, the coefficients found for both the dependent variables, commitment to JIT (measured by Com-to-JIT) and commitment to TQM (measured by Com-to-TQM) are located between 0.20 and 0.50 (although some are in the confidence interval of 5 per cent and other coefficients in the interval of 1 per cent). This contrasts with results of the original UK-based survey by Soriano-Meier and Forrester (2002) where the correlations were above 0.5. A possible explanation for the low among the variables analysed for commitment to TQM in the Brazil agricultural machinery industry is the

1 Variable

Table III. Correlation between and among Com-to-JIT, Com-to-TQM, QLEAD, GROUP, TRAIN and WEMP

1. Commitment to JIT (Com-to-JIT) 2. Commitment to TQM (Com-to-TQM) 3. QLEAD 4. Problem-solving group (GROUP) 5. Training (TRAIN) 6. Empowerment (WEMP)

Com-to JIT

2 Com-toTQM

3

QLEAD GROUP TRAIN WEMP

1

0.411 **

0.307 *

1

0.538 ** 0.351 * 0.296 * 0.368 * 0.443 ** 0.548 ** 0.597 ** 1 0.471 ** 0.33 * 1 0.728 ** 1

Note: Correlation is significance at *0.05 and **0.01 levels (one tailed)

4

5

6

0.447 ** 0.381 ** 0.319 *

very new and recent history of implementation which, according to some interviewees, only truly commenced around the year 2000. Analysing the data for this study (Table III), there is a strong correlation between empowerment and the training given to employees (0.728), which is significant at 1 per cent. This indicates that the sector seeks to release decision-making power to employees as they are trained in the tasks they perform. The multiple regression analysis was performed with the following purposes: . to verify the degree of relationship between the two dependent variables and the four independent variables, considered one by one; . to determine the relative importance of each independent variable in the forecast of the dependent variable; and . to determine the existence of co-linearity effects.


863

The results show to what extent the dependent variable is explained by each one of the independent variables. The standardized regression coefficient or the beta coefficient (b) is used to determine the relative importance of each independent variable in the dependent variable (Table IV). It explains the individual contribution that each type of investment in plant infrastructure makes in the commitment to JIT and commitment to TQM. As a conclusion, there is support to accept H1 with a basis on the previous analyses, where it was verified that all the 15 variables proved significant at statistically acceptable levels. H2. Companies that allegedly adopt the lean manufacturing principles have made changes in the direction of these principles. Correlation and multiple regression analyses were conducted to test H2. This hypothesis seeks to verify whether the companies that allegedly adopt the lean manufacturing principles (measured by DOA) have made changes in the direction of these principles (measured by “EW”, “CI”, “ZD”, JIT deliveries “JIT”, pulling of material “PULL”, “MFT”, decentralization “DEC”, “IF” and “VIS”). Paradoxically to the variables of commitment to JIT and commitment to TQM, where there is a low degree of commitment, it was verified that for the DOA, 73 per cent of the firms (27 out of 37) have a mean value of four or higher, which indicates a high degree of emphasis on the adoption of the lean manufacturing programme. This perception of DOA of the lean manufacturing programme is provided by the chief operations officer (COO), and we can report a convergence and similarity of views among the CEOs and COOs as to the stage of implementation of the programme at the companies. F Model 1 (dependent variable Com-to-TQM) Commitment to TQM (Com-to-TQM) 193.92 QLEAD Model 2 (dependent variable Com-to-JIT) Commitment to JIT (Com-to-JIT) 97.489 Problem-solving group (GROUP) Commitment to TQM (Com-to-TQM)

p

R 2 adjusted

0.000

0.843

0.000

b

t

p

0.697

13.93

0.000

0.843 0.562 0.342

4.294 2.149

0.000 0.039

Table IV. Analysis of multiple regression H1

JMTM 21,7

864

Correlation analysis Table V presents the analysis of correlation between the DOA (measured by DOA) and the changes in direction of these principles (measured by “EW”, “CI”, “ZD”, JIT deliveries “JIT”, pulling of material “PULL”, “MFT”, decentralization “DEC”, “IF” and “VIS”). In the Soriano-Meier and Forrester’s (2002) study, the variables pulling of material (PULL) and IF did not appear to be statistically significant; likewise in this study, we also verified the same behaviour in relation to the variable JIT deliveries. Even with the broad scope of a lean adoption programme, which contemplates all the areas of the manufacturing process, we can verify that there are significant correlations of the DOA of the programme with practically all the variables analysed in this hypothesis, with the exception of JIT deliveries. The statistical analysis performed corroborates the acceptance of H2. H3. Degree of leanness, degree of commitment and the relation with performance. Using the same techniques adopted for other hypotheses, correlation analyses, analysis of multiple regression, cluster analysis and one-way analysis of variance (ANOVA) were performed, followed by a Tukey honestly significant difference (HSD) test. This hypothesis verifies the relationship between the degree of leanness (measured by DOL), the degree of commitment to the lean manufacturing programme (measured by DOC) and the performance (measured by PERF). The degree of leanness (measured by DOL) was obtained through the mean of the variables that measure the DOA of the lean manufacturing programme, which are EW, CI, ZD, JIT deliveries (JIT), pulling of material (PULL), MFT, decentralization (DEC), IF and VIS. The variable degree of commitment (measured by DOC) is the mean of the variables: QLEAD, problem-solving group (GROUP), training (TRAIN) and empowerment (WEMP). Variable performance (measured by PERF) was measured by a proportion between billing by employee and turnover of assets, in comparison to the years 2000 and 2003. Correlation analysis As it can be seen in Table VI, all the three variables have high levels of correlation and in accordance with the parameters to avoid co-linearity do not exceed 0.80. The results of this correlation analysis can be considered acceptable. Therefore, the regression analysis shown below was performed in order to identify the nature of this relationship. Regression analysis Table VII shows the result of the regression analysis conducted for the coefficients of DOL and DOC. We verified that only the variable DOL explains the variable PERF at 95 per cent and with high significance ( p , 0.01). Cluster analysis The cluster analysis was performed in order to group individuals or objects in clusters; hence, individuals in the same cluster are more alike than individuals in the other cluster. The objective is to maximize the homogeneity of individuals inside the cluster, while there is also a maximisation of heterogeneity between and among the clusters (Hair et al., 1998). Table VIII shows the classification of firms into “lean”, “in transition” and “traditional”. The classification reveals that 27 per cent (ten out of 37) of the firms

1

1. Degree of adoption (DOA) 2. Elimination of waste (EW) 3. Continuous improvement (CI) 4. Zero defects (ZD) 5. Just-in-time deliveries (JIT) 6. Pulling of material (PULL) 7. Multifunctional teams (MFT) 8. Decentralization 9. Integration of functions (IF) 10. Vertical information system (VIS)

3 0.849 ** 0.832 **

2 0.831 ** 1

4 0.762 ** 0.640 ** 0.643 ** 1

Note: Correlation is significance at *0.05 and **0.01 levels (one tailed)

1

Variable 0.026 0.019 0.062 0.087 1

5 0.745 ** 0.535 ** 0.527 ** 0.476 ** 20.076 1

6 0.803 ** 0.582 ** 0.635 ** 0.537 ** 2 0.115 0.751 ** 1

7 0.756 ** 0.503 ** 0.505 ** 0.439 ** 20.118 0.590 ** 0.562 ** 1

8

10 0.735 ** 0.608 ** 0.677 ** 0.504 ** 20.056 0.352 ** 0.501 ** 0.671 ** 0.390 ** 1

9 0.686 ** 0.514 ** 0.513 ** 0.422 ** 0.124 0.401 ** 0.517 ** 0.538 ** 1


865

Table V. Correlation between and among DOA, EW, CI, ZD, JIT, PULL, MTF, DEC, IF and VIS

JMTM 21,7

866

studied are classified as lean firms, 46 per cent (17 out of 37) as firms in transition and 27 per cent (ten out of 37) are traditional according to the criterion adopted. One-way ANOVA test After the classification of the firms in these criteria, a one-way ANOVA test was conducted in order to verify how these differ in terms of the variables that we are analyzing for H3: degree of leanness (measured by DOL), degree of commitment (measured by DOC) and performance (measured by PERF). Table IX shows the main results for each type of firm, with statistical significance. In essence, the one-way ANOVA test examines the total difference in the mean, and it can only indicate that the mean values of the groups are not equal and therefore reject the null hypothesis that the mean values of the groups are equal. Hence, the Tukey HSD test is necessary. Tukey HSD test The one-way ANOVA test helped to conclude that there are differences among the groups but it is unable to indicate where they are. Carrying out a procedure similar to that of the survey by Soriano-Meier and Forrester (2002), the Tukey HSD test was applied to the variables degree of leanness (measured by DOL), degree of commitment Variable

Table VI. Correlation between and among DOL, DOC and PERF

Table VIII. Firm classification criterion

Table IX. One-way ANOVA test

DOC

PERF

1

0.294 * 1

0.625 * * 0.599 * * 1

Degree of “leanness” (DOL) Degree of commitment (DOC) Performance (PERF)

Note: Correlation is significance at *0.05 and * *0.01 levels (one tailed)

Variable Table VII. Analysis of regression between PERF and DOL

DOL

1. Performance (PERF) 2. Degree of leanness (DOL)

F

p

R 2 adjusted

657.39

0.000

0.954

1. Lean firms 2. Firms in transition

b

t

p

0.977

0.000

25.640

For DOL . 4.57 and DOC . 4.70 For DOL . 4.57 and DOC , 4.70 or for DOL , 4.57 and DOC . 4.70 For DOL , 4.57 and DOC , 4.70

3. Traditional firms

Variable

df

F

p

DOL DOC PERF

2 2 2

19.009 10.224 128.655

0.000 0.000 0.000

(measured by DOC) and performance (measured by PERF). Table X presents the comparison of the Tukey HSD test for the three variables: degree of leanness (measured by DOC), degree of commitment (measured by DOC) and performance (measured by PERF). The test was aimed to verify whether lean firms have higher significant mean values than traditional firms, in the three variables studied in the model, and as it can be verified in Table X, this is confirmed. Another important analysis was to verify whether lean firms also have higher mean values than firms in transition, which was confirmed in the table, as lean firms have higher statistically significant mean values than firms in transition for the three variables studied. It was concluded that, with a basis on the tests carried out and on the results of the correlations, multiple regression analysis, cluster analysis and one-way ANOVA test followed by the Tukey HSD test, the results corroborate H3.


867

6. Conclusion The objective of this paper was to study the performance of manufacturing companies from the agricultural machinery sector in Brazil, which strategically opted for the implementation of lean manufacturing. The variables defined and studied in this paper were based on the survey by Soriano-Meier and Forrester (2002), where nine independent variables were defined to measure the DOA of the programme, which was summarised in a dependent variable-designated degree of leanness. Another metric was also developed to identify the degree of commitment to the programme. Once these two variables were obtained, the authors verified their relationship with performance, which was measured through the ratios of billing by employee and turnover of assets, calculated using secondary data. A group of 37 firms from the sector of agricultural

Variable DOL

(I) Type 1 2 3

Degree of commitment (DOC)

1 2 3

Performance (PERF)

1 2 3

Tukey HSD ( J) Type Difference of mean values (I 2 J) 2 3 1 3 1 2 2 3 1 3 1 2 2 3 1 3 1 2

0.651 0.927 2 0.651 0.276 2 0.927 2 0.276 1.095 1.510 2 1.095 0.415 2 1.510 2 0.415 2.212 4.700 2 2.212 2.488 2 4.700 2 2.488

p 0.000 0.000 0.000 0.130 0.000 0.130 0.003 0.000 0.003 0.000 0.003 0.387 0.000 0.000 0.000 0.000 0.000 0.000

Table X. Tukey HSD test

JMTM 21,7

868

machinery and implements took part in the survey. Two questionnaires per firm were applied: the first for the CEO and the second for the COO. The agricultural machinery and implements sector are characterized by complex, hard-to-make products often presenting long lead times, facts which might delay or hamper the real benefits of the implementation of lean principles as viable strategy in this particular sector. Correlation and multiple regression analyses were conducted in order to test H1. With a basis on the statistical analyses performed, it was verified that there is enough support to accept H1: firms with a high degree of management commitment to the simultaneously support this commitment to investments in support of the plant infrastructure, measured by QLEAD, problem-solving groups (GROUP), training (TRAIN) and empowerment (WEMP). However, for H2, it was verified that the relationship of DOA of the model (measured by DOA) proved significant when compared with the variables: measured by “EW”, “CI”, “ZD”, pulling of materials “PULL”, “MFT”, decentralization “DEC”, “IF” and “VIS”. The relationship with the variable JIT deliveries “JIT” did not prove statistically significant. Although this single variable did not prove statistically significant, it can be accepted that for the sector of agricultural machinery and implements in Brazil, this hypothesis is true. A possible explanation of this result is the lack of continuous support by top management – a tendency to view the adoption of lean principles as a short-term panacea to achieve immediate improvements, thus failing to embrace the real benefits of the strategic element of the lean approach. This near term perspective inhibits the longer term continuous support of lean principles. As a consequence, and in line with Schonberger’s (2008) analogy, waste reduction should not be the dominant lean target. Pursuing lean provides a system that encourages value creation, improves capabilities and provides a longer financial health, strategic concerns that tend not to be identified with a short-term view, particularly in this sector. Finally, correlation analyses, multiple regression, cluster and one-way ANOVA analyses followed by a Tukey HSD test were applied to test H3: firms that made continuous investments in plant infrastructure (SMI) in the direction of the lean principles (measured by degree of leanness “DOL” and DOA “DOC”) have better performances (measured by PERF). With a basis on the data and the analyses proposed, it was verified that the data corroborates the validation of H3. This paper, therefore, provides a detailed and original analysis of the implementation of lean operations in the agricultural sector in Brazil. It provides a “state of play” in terms of the degree of leanness and the degree of commitment – and the link to business performance. The paper demonstrates that the adoption of lean manufacturing is taking grip within the agricultural machinery and implements sector of Brazilian industry and also demonstrates that early adopters and those “in transition” to lean have improved their performance and competitiveness over competitors. It is clear that the adoption of JIT is less central a concern to businesses in Brazil (perhaps due to logistical and geographical reasons), compared with those in other economies of the world, and so not as central to their interpretation of lean. The paper also shows that the Soriano-Meier and Forrester’s (2002) model is applicable to other industries and economies, beyond the original UK context, for measuring and comparing degrees of leanness. But clearly further investigation is required to understand the drivers for change and adoption in Brazil. So further research will test the model in other sectors of the Brazilian economy and also in the same sector but in different national contexts.

References Barney, J. (1991), “Firm resources and sustained competitive advantage”, Journal of Management, Vol. 17 No. 1, pp. 99-120. Basso, L., Shimizu, U. and Nakamura, W. (2006), “Produçaõ Enxuta e Desempenho de Mercado, Uma Ana´lise para o Setor de Ma´quinas e Implementos Agrıćolas no Brasil”, IX SIMPOI 2006, paper presented at: Simpo´sio de Administraçaõ da Produçaõ, Logı´stica e Operaço˜es Internacionais. Cheng, T. and Podolsky, S. (1996), Just-in-time Manufacturing: An Introduction, 2nd ed., Chapman & Hall, London. Collis, D. and Montgomery, C. (1997), Corporate Strategy: A Resource-based Approach, McGraw-Hill, New York, NY. De Oliveira, E. and Fensterseifer, J. (2003), “Use of resource-based view in industrial cluster strategic analysis”, International Journal of Operations & Production Management, Vol. 9 No. 23, pp. 995-1009. Goddard, W. (1986), Just-in-time, Oliver Wight, Brattleboro, VT. Grant, R. (1991), “The resource-based theory of competitive advantage: implications for strategy formulation”, California Management Review, Vol. 33 No. 3, pp. 114-35. Hair, J.F. Jr, Anderson, R., Totha, R. and Bloch, W. (1998), Multivariate Data Analysis, 5th ed., Prentice-Hall, Englewood Cliffs, NJ. Harrison, A. (1992), Just-in-time Manufacturing in Perspective, Prentice-Hall, Hemel Hempstead. Hines, P., Holweg, M. and Rich, N. (2004), “Learning to evolve: a review of contemporary lean thinking”, International Journal of Operations & Production Management, Vol. 24 No. 10, pp. 994-1011. Huber, G. (1991), “Organizational learning: the contributing processes and the literature”, Organization Science, Vol. 2 No. 1, pp. 88-115. Karlsson, C. and Ahlstrom, P. (1996), “Assessing changes towards lean production”, International Journal of Operations & Production Management, Vol. 16 No. 2, pp. 24-41. Katayama, H. and Bennett, D. (1996), “Lean production in a changing competitive world: a Japanese perspective”, International Journal of Operations & Production Management, Vol. 16 No. 2, pp. 8-23. Learned, E., Christensen, C., Andrew, K. and Guth, W. (1969), Business Policy, Irwin, Homewood, IL. Lewis, M. (2000), “Lean production and sustainable competitive advantage”, International Journal of Operations & Production Management, Vol. 20 No. 8, pp. 959-78. Liker, J. (2004), The Toyota Way – 14 Management Principle from the World’s Greatest Manufacturer, McGraw-Hill, New York, NY. Metalworking Insider Report (2004), “Brazil’s machinery sector launches and export drive”, available at: www.allbusiness.com/manufacturing/fabricated-metal-product-manufacturing/ 184393-1.html (accessed June 7, 2009). Monden, Y. (1983), Toyota Production Systems, Industrial Engineering & Management Press, Norcross, GA. NIST (2000), Principles of Lean Manufacturing with Live Simulation, Manufacturing Extension Partnership, National Institute of Standards and Technology, Gaithersburg, MD. Ohno, T. (1988), The Toyota Production System: Beyond Large-scale Production, Productivity Press, Portland, OR. Porter, M. (1981), “The contribution of industrial organization to strategic management”, Academy of Management Review, Vol. 6, pp. 609-20.


869

JMTM 21,7

870

Schonberger, R. (2008), World Class Manufacturing, The Free Press, New York, NY. Sirkin, M. (1999), Statistics for the Social Sciences, Sage, Thousand Oaks, CA. Sitkin, S. (1991), “Learning through failure: the strategy of small losses”, in Staw, B. and Cummings, L. (Eds), Research in Organizational Behaviour, Vol. 14, JAI Press, New York, NY. Soriano-Meier, H. and Forrester, P. (2002), “A model for evaluating the degree of leanness of manufacturing firms”, Integrated Manufacturing Systems, Vol. 13 No. 2, pp. 104-9. Williams, K., Haslam, C., Johal, S. and Williams, J. (1994), Cars: Analysis, History, Cases, Berghahn, Oxford. Womack, J., Jones, D. and Roos, D. (1990), The Machine that Changed the World, Rawson Associates, New York, NY. Young, D. and O’Byrne, S. (2003), EVA e Gestaõ Baseada em Valor – Guia Pra´tico para Implementaçaõ, Bookman, Porto Alegre. About the authors Paul L. Forrester is a Senior Lecturer in Operations Management at the Birmingham Business School of the University of Birmingham, UK, where he is also the Director of full-time MBA and DipBA programmes. Paul L. Forrester has held various research and teaching appointments including eight years at Keele University where he became the Director of MBA Programmes and three years prior to joining Birmingham at Aston Business School (Director of part-time MBA programmes and Convenor of the Technology and Operations Management Research Group). His research interests lie in the strategic, design and organisational issues of managing operations and projects, the extension of operations management concepts to service organisations and the use of virtual learning for management education. He has supervised a number of successful PhD projects and has over 30 academic journal and book chapter publications in addition to over 100 conference papers. Ullisses Kazumi Shimizu is a Lecturer in Cost Accounting, Planning & Control and Scientific Methodology at Mackenzie University, Brazil. He has 26 years of professional experience in the financial area (administration, accounting, planning and control, cost accounting and plant controlling) that was acquired while working for a German multinational pharmaceutical company, an American multinational medical devices company and a general private hospital. Horacio Soriano-Meier is a Senior Lecturer in Operations Management at the Northampton Business School of the University of Northampton. Horacio Soriano-Meier received his BSc in Civil Engineering in 1978 from the “Universidad Central de Venezuela”, Caracas; his MBA from Bryant College (now Bryant University), Smithfield, USA, in 1985 and his PhD in Operations Management at Keele University, UK, in 2001. He has held previous academic posts in the area of Operations Management in the Universities of Cardiff and Birmingham in the UK, and ULA, UCV and INTEL in Venezuela. Horacio Soriano-Meier has supervised a number of successful PhD and MSc projects, and prior to his academic career, he worked as an Engineer and Management Consultant. Jose Arturo Garza-Reyes is a Lecturer in Manufacturing Engineering at the School of Technology of the University of Derby, UK. Jose Arturo Garza-Reyes was graduated in 1998 from the Autonoma de Nuevo Leon University (UANL) in Mexico as Mechanical Administrator Engineer. In 2001, he was graduated as Master in Sciences, with major in Production and Quality, from the same University. In 2003, he was awarded a scholarship from the Mexico’s National Council of Science and Technology (CONACYT) to pursue a PhD in Manufacturing Systems and Operations Management at the University of Manchester, from where he graduated in 2008. He has recently completed an MBA at the Northampton Business School of the University of Northampton and is also a certified Six Sigma-Green Belt. Jose Arturo Garza-Reyes has six years of industrial experience working as Production Manager, Production Engineer and

Operations Manager for several international and local companies in the UK and Mexico. Jose Arturo Garza-Reyes is the corresponding author and can be contacted at: [email protected] Leonardo Fernando Cruz Basso is a Professor at the Universidade Mackenzie and financed by Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico, Brazil. He has experience in the field of economics, with emphasis in economic theory. Fernando Leonardo Cruz Basso’s research interests are in exchange rate, monetary economy and money valuation. He graduated from the Instituto Tecnolo´gico da Aeronaútica, ITA in 1974 as Mechanical Engineer and specialised in Economic Theory at the Universidade de Saõ Paulo, USP, in 1978. Leonardo Fernando Cruz Basso has a Master’s degree in Economics from the New School for Social Research (1981), a doctorate in Economics from the same university (1984) and a post-doctorate from Bielefeld University (1993).

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints


871