The Implementation of Enterprise Resource Planning and Product Data Management in Semiconductor Related Industries: An Empirical Study in Taiwan

Article excerpt

Along with the development of information technology and fierce competition among industries, how to best use, manage and apply enterprise resources has become a critical issue in industrial transformation. In practice, management concepts can be implemented through Enterprise Resource Planning (ERP) and Product Data Management (PDM). Providing required information in real-time thus becomes a key point for enterprise survival in the efficiency-based semiconductor related industries of Taiwan. This study attempts to understand the current implementation of ERP and PDM in the semiconductor related industry in Taiwan using a field survey method. This study comprises two parts. First, a questionnaire survey concerning the implementation of ERP and PDM is conducted and analyzed statistically to explore the application of ERP and PDM in the semiconductor related industries of Taiwan. Second, the current state of the implementation of ERP and PDM as revealed in the research results and in the on-site interviews with semiconductor related manufacturers provide ideas for enterprises for implementing their ERP and PDM.

1. Introduction

Promoted by the American Production and Inventory Control Society (APICS) since 1970, MRP has been developed for several decades and has become MRP II by extension to the marketing, finance and personnel dimensions (Sum et al., 1997). MRP II is a production control system and has been adopted in the American and Taiwanese manufacturing industries to integrate limited internal resources and to overcome fierce competition. However, because of globalization, information systems should provide firms with the communication and analytical power to trade and manage their businesses on a global scale. MRP II is insufficient to cope with this trend.

To become effective and profitable participants in international markets, enterprises require more powerful information systems to help them extend their geographic reach, offer new products and services, reshape workflow, and perhaps profoundly alter their approach to business. Enterprise Resources Planning (ERP) thus is the right choice for businesses seeking to recast their process of management. It provides powerful new capabilities to help managers in planning, organizing, leasing and control (Na et al., 2003; Ralph, 1998). ERP has been among the fastest growing software and most important developments in IT in the 1990s. Davenport (1998) stated "while the rise of the Internet has received most media attention in recent years, the business world's embrace of ERP systems may in fact be the most important development in the corporate use of IT in the 1990s." Currently, many companies already use ERP to make planning, forecasting, and monitoring more precise than previously"increase planning, forecasting and monitoring precision.

ERP is a business management system that integrates all facets of a business, including planning, manufacturing, sales, purchasing, marketing, and finance, so they can cross organizational boundary and be more closely coordinated by sharing information (Davenport, 1998; Tim, 1998; Van der Aslst, 2000). The Market Intelligence Center (MIC) in Taiwan defines ERP narrowly as "resources inside an enterprise" and broadly as "an enterprise operation and management system of integrating external information" (MIC, 2003).

According to the field survey conducted by MIC in 1999, pressures from large-scale firms and the Y2K effect were two main drivers of the popularity of ERP on the software market. The prosperous development of ERP is closely related to the role played by the information, electronic and semiconductor manufacturers in the global supply chain system. Facing the trend of globalization and international competition, international large-scale manufacturers have started to ask their collaborating factories to implement ERP to achieve synergies in the supply chain management (Gould, 1997; Bragg, 1997). The upstream manufacturers place pressure on the downstream manufacturers to ensure ERP is implemented, which is one of the main reasons why plenty of Taiwanese OEM manufacturers implement ERP systems.

ERP undoubtedly offers great potential rewards and benefits, but the associated risks are equally great. If a company rushes to install or implement an ERP system without a clear understanding of the business implications, then the dream of integration can quickly become a nightmare. Davenport (1998) thinks that only 20% of ERP implementation is technology related, while the remaining 80% involves management change issues, with human problems dominating. As we know, domestic Taiwan ERP software companies spend considerable sums on consulting and training to help manufacturers implement ERP system to increase ERP acceptance. Previous studies displayed that times, costs and specifications were generally selected as criterion for measuring ERP performance. Power & Dickson (1973) showed that user satisfaction is also an important indicator of ERP performance. Delone et al. (1992) summarized previous researches and proposed a success model for integrated information systems, which included system quality, information quality, usage, user satisfaction, individual impact and organizational impacts. Khalil et al. (1999) reported that user satisfaction is positively related to system usage.

Because of the prevalence of ERP, product data management (PDM) systems have received more attention in the industry recently. With the development of IT (information technology), the functions and features of PDM technology have become more complete and clearly defined. PDM systems manage all common and secret product-related data, and provide essential product data (Leong, et al., 2003). Moreover, PDM is also an enabling technology for helping workgroups, departments, divisions, or enterprises to manage products and the development process. PDM is a tool for helping engineers to manage both engineering data and the product development process throughout the product lifecycle (Oh et al., 2001). Furthermore, PDM can be used to guarantee smooth production, and directly or indirectly as an input for decision-making processes. PDM systems presently are being used to integrate heterogeneous information systems to meet the global and distributed manufacturing environment needs of the present environment. Product data management is crucial in product development and production (Leong, et al., 2002). Manufacturers currently face more challenges than ever before. In order to remain competitive and survive, manufacturers are constantly trying to reduce their production costs.

Consequently, many companies have distributed their production plants around various different locations, including locations overseas. However, if existing PDM processes cannot meet operation and management requirements for traditional large volume production then difficulties occur in running an enterprise. As a result, the PDM system provides the best solution for ordering large quantities of products. From the perspective of management skills, PDM is not a brand-new methodology since it has long existed in various forms for a long time.

Integration of information systems usually is confined to a certain factory or company for most enterprises. However, the combination of ERP and PDM can exchange internal information inside an enterprise, make the work process more smooth and precise, improve communication among departments, prevent repetition of procedures, reduce costs and integrate global resources to allow competitive enterprises to develop and survive in the market.

An increase in ERP implementation activity swept through semiconductor related industries in Taiwan in the mid-nineties. PDM systems have also been around for the past few years. At this time, PDM systems had also been implemented for a few years to reduce product development cycle time, access company information, and improve project management, life cycle design, supply chain collaboration and interdisciplinary collaboration. Although ERP systems integrate modules for such varied functions as production, factory automation, finance, sales, purchasing, and personnel, many companies that strongly advocate ERP systems find they are not robust or flexible enough to deliver the special budgeting, international consolidation, or other financial features expected by management. Nevertheless, few studies exist on the differences between the actual outcomes achieved by ERP and PDM and the expected effects on implementation.

As a result, this study uses a questionnaire survey plus site interviews to evaluate the implementation of PDM and ERP in semiconductor related industries in Taiwan. This study comprised two steps: (1)A questionnaire was used to investigate, analyze, and generalize the problems and difficulties faced by semiconductor related industries in Taiwan while implementing ERP and PDM, as well as the solutions devised; (2) Operating Performance and organizational climate were assessed after implementing ERP and PDM in semiconductor related industries in Taiwan, using various statistical approaches; (3) Site interviews with semiconductor related manufacturers were used to enhance the research methodology and provide a reference for industries for implementing ERP and PDM in the future.

The analytical results of this study are expected to help in understanding the implementation of ERP and PDM in semiconductor related industries in Taiwan.

2. Methods

2.1 Instruments and Procedures

The measurement of this study is based on previous research, documents, filed interviews collected from several Taiwan's semiconductor companies and consulting firms.

The questionnaire includes four parts, which are the background information about firms, implementing motivations and methods, implementing process and performance evaluation before and after carrying out ERP or PDM. The implementing motivations and methods, implementing process and performance evaluation are designed with a Likert scale from 1 to 5.5-point means extremely agree and 1-point represents absolutely disagree.

There were three data collection steps in this study. In the first, samples were randomly drawn from the Taiwan Semiconductor Industry Association (TSIA, 2003). Then the researchers individually and privately called the manager of the information system department to solicit his voluntary help to arrange for this study. Finally 180 semiconductor firms were willing to participate this study. In the second step, questionnaires were mailed to the 180 managers of information system departments. Finally on-site interviews examined the results more thoroughly.

2.2 Analysis method

To achieve the objectives of the research and cope with the research structure, software SPSS 10.0 was applied as an analysis tool. The analysis procedures can be divided into five parts.

Firstly, Cronbach's α coefficient was used to test consistency and stability of the questionnaire. Validity of the questionnaire was determined by content validity and structure validity. Secondly, a correlation analysis was performed. Variables to be studied included capital amount invested by manufacturers while implementing ERP and PDM, implementation willingness, product delivery date, repetitive investment and waste, product design capability and customer satisfaction. A correlation analysis was carried out based on the results of these variables. Thirdly an ANOVA was performed. The significant impact of each basic variable upon ERP and PDM systems introduced by enterprises can be revealed through ANOVA. Next, where operation performance is affected by the interaction among each characteristic variable, the variables are analyzed. The fourth part is factor analysis. The results of practice and major influential factors while implementing ERP can be extracted through factor analysis. The last part is regression analysis. The influence of each dimension can be explored via regression analysis and a regression model established accordingly.

3. Results, Analysis and Conclusion

To appreciate the current state of ERP and PDM implementation in Taiwan, a questionnaire survey was conducted. The sampling was based on 180 semiconductor related manufacturers provided by the Taiwan Semiconductor Industry Association (TSIA, 2003).

3.1 Reliability and validity analysis

The process of introducing and implementing ERP and PDM are analyzed in this research. The internal reliability coefficient of the scale is 0.8265. According to Gay (1992), a reliability coefficient above 0.8 for any test or scale is the minimum acceptable; however, a coefficient over 0.7 was acceptable for DeVellis (1991). Accordingly, the reliability coefficient of this questionnaire survey is quite high, which meant the results are stable and consistent. Since the design of the questionnaire in this study was based on pertinent documents in Taiwan and overseas, and the opinions of project team members of several semiconductor related companies in Taiwan who were interviewed, content validity was appropriate. The dimensions obtained by factor analysis were consistent with the proposed structure; thus, the structural validity of the questionnaire is 'quite high'. As a result, the validity of the questionnaire seems convincing.

3.2 Correlation analysis

3.2.1 Capital of corporation

Pearson's test was used to measure the correlation between corporate capital and software purchase price. The correlation coefficient of 0.437 suggests that the more capital a corporation has, the greater the flexibility of the software purchase price.

3.2.2 Implementation willingness of senior managers

Pearson's test was used to measure the correlation among implementation willingness of higher level managers, the cognition of the long-term partnership on both sides and the cognition of the departments involved on both sides concerning the introduction of ERP and PDM. The correlation coefficients of 0.517 and 0.595 were significantly positive; i.e., the higher the implementation willingness of superiors, the greater the common view of long-term cooperation on both sides.

3.2.3 Product delivery date

Pearson's test assessed the correlation between shortening product delivery date and reducing operation cost. The positive correlation coefficient of 0.569 was significant suggesting that the shorter the product delivery date, the lower the operating cost.

3.2.4 Repetitive investment and waste

Pearson's test measured the correlations among decreasing repetitive investment and waste, shortening product delivery date and reducing operation cost. The correlation coefficients of 0.517 and 0.595 were significantly positive; the shorter the product delivery date and the lower the operation cost, the less repetitive investment and waste.

3.2.5 Product design capability

Pearson's correlation test examined the relations among increasing product design capability, shortening product delivery date and reducing operation cost. The significantly positive correlation coefficients of 0.503 and 0.385 suggest that the higher the product design capability, the shorter the product delivery date and lower are operation costs.

3.2.6 Customer satisfaction

Pearson's correlation test examined the relations among increasing customer satisfaction, shortening product delivery date, controlling supply records of suppliers and enhancing material turnaround rate. The significantly positive correlation coefficients of 0.531, 0.482 and 0.351 suggest that the shorter the product delivery date, the better the control of information about suppliers and the higher the material turnaround rate, the higher is customer satisfaction.

3.2.7 Enhanced competitiveness after adopting ERP and PDM

Pearson's correlation test examined the relations among competitiveness after adopting ERP and PDM, shortening product delivery date, reducing operation cost, decreasing repetitive investment and waste, promoting product design capability, controlling supply records of suppliers, increasing material turnaround rate and enhancing customer satisfaction. The correlation coefficients of 0.672, 0.349, 0.615, 0.484, 0.443, 0.406, 0.684 and 0.443 were all significantly positive suggesting that when competitiveness after adopting ERP and PDM gets higher, product delivery date becomes shorter, operation cost becomes lower, repetitive investment and waste become less, product design capability becomes better, supply records of the suppliers under control improve, material turnaround rate becomes shorter and customer satisfaction gets higher.

3.3 ANOVA

3.3.1 ANOVA of each basic variable

The results of the tests of the impact of each basic variable after the introduction of ERP and PDM by enterprises are listed in Table 1.

Table 1 shows that none of the F values in the homogeneity analysis, including the number of employees, capital, time of implementing ERP and PDM, business reengineering prior to implementing ERP and PDM, software purchase price and time spent on implementing modules are significant. Therefore, all variables can be regarded as homogeneous. The insignificant P values in the table indicate no impact for the number of employees, capital, time of implementing ERP and PDM, business reengineering prior to implementing ERP and PDM, software purchase price and time spent on introducing modules.

3.3.2 ANOVA of interaction among feature variables

Results of the effect of the interaction of each variable after ERP and PDM were introduced by enterprises are analyzed by two-way ANOVAs and listed in Table 2. The results show that the F values of all manufacturer variables are insignificant, confirming their homogeneity. However, there is no correlation among all variables due to insignificant P values. Hence, a comparison of major effects cannot be made.

3.4 Factor analysis

3.4.1 Introduction process

Ten factors obtained during the introduction process were extracted and analyzed by factor analysis through using principle components, with the results listed in Table 3.

Fig. 1 is the factor scree plot of the introduction process. From the plot the slope tends to be even after five dimensions. Therefore, it is appropriate to reserve three to five dimensions.

The result indicated a KMO (Kaiser Meyer Olkin) of 0.567. According to Kaiser (1974), when KMO is less than 0.5, it is not proper to conduct a factor analysis. As the KMO here is 0.567, a factor analysis can be conducted.

Based on the factor matrix and the scree plot, three dimensions are most suitable for factor analysis during the introduction process. Names of the factors in accordance with the results in Table 1 are as follows:

* Factor 1 includes full communication between both sides before cooperation, cognition of long-term partnership on both sides, strong implementation inclination of higher level, incoordination among departments, mutual capability assessment before cooperation, mutual credit assessment before cooperation and mutual assessment on size before cooperation. As it is related to communication inside an organization, it is called 'Communication.'

* Factor 2 includes employees are willing to cooperate and exchange of technical resources and information on both sides. As it is related to technical interchange, it takes the name of Technical Exchange.'

* Factor 3 includes a common knowledge of ERP and PDM for departments involved on both sides. As it is related to the system and organization, it is called 'System Organization.'

3.4.2 Performance factor analysis after implementing ERP and PDM

14 factors concerning the implementing of ERP and PDM were extracted by factor analysis. The extraction method was based on principal components. The results of the analysis are listed in Table 4 as follows:

Fig. 2 is the factor scree plot of performance after implementing ERP and PDM. It is known from the plot that the slope tends to be even after six dimensions. Therefore, it is more appropriate to limit the analysis to four to six dimensions.

The result indicated a KMO (Kaiser Meyer Olkin) of 0.700. According to Kaiser (1974), when KMO is less than 0.5, it is not proper to conduct a factor analysis. As the KMO here is 0.700, a factor analysis can be conducted.

Based on the factor matrix and the scree plot, four dimensions are suitable for analysis of performance factors after implementing ERP and PDM. Names of the factors in accordance with the results in Table 4 are as follows:

* Factor 1 includes increased competitiveness after adopting ERP and PDM, quality problems resulted from design improvement, increased customer satisfaction, access to all product information from design to production, reduced repetitive investment and waste, shortened delivery duration, enhanced material turnaround rate, increased product design capability and reduced operation cost. As it is related to the performance of operation, it is named as Operation Performance.'

* Factor 2 includes convenience of ERP and PDM operation, corporate satisfaction with ERP and PDM operation and correct information control exactly. As it is related to management of the system, it takes the name of 'System Management.'

* Factor 3 consists of better interaction among departments. As it is related to coordination in the organization, it is named as Organization Coordination.'

* Factor 4 comprises control of supply records of the suppliers. As it is related to management of supplies, it is named as 'Supply Management.'

3.5 Regression analysis

3.5.1 Correlations between implementation performance and cognition of system implementation, technical exchange and communication

The three performance factors extracted earlier serve as independent variables and internal performance as the dependent variable for multiple regression. These three factors that influence internal performance are Communication, (X^sub 1^), Exchange & Cooperation (X^sub 2^) and perception of system implementation (X^sub 3^). Results of multiple regression analysis are as follows:

Y = -0.335X^sub 1^ -0.025X^sub 2^ -0.168X^sub 3^

Adjusted R^sup 2^ (corrected correlation coefficient) = 0.052, F value = 1.583 and P value = 0.215 > α= 0.05 as Tables 5 and 6 show, t values of all factors are -1.943, -0.148 and 0.974 respectively as Table 7 indicated. As P value was not significant, it means that an insignificant regression relationship existed between integral performance factors and overall performance satisfaction. When Adjusted R^sup 2^=0.052, it means that these three performance factors could explain the difference of overall internal performance by 5.2%.

3.5.2 Correlation between operation performance and cognition of system implementation, technical exchange and communication

The three performance factors extracted earlier served as independent variables and operation performance as the dependent variable for multiple regression. These three factors that influence internal performance are Communication, (X^sub 1^), Exchange & Cooperation (X^sub 2^) and cognition of system implementation (X^sub 3^). Results of standardized regression analysis are as follows:

Y = 0.067X^sub 1^ + 0.307X^sub 2^ + 0.273X^sub 3^

Adjusted R^sup 2^=0.083, F value = 1.965 and P value = 0.141 > α=0.05 as Tables 8 and 9 indicated. t values of all factors are 0381, 2.243 and 0.848 respectively shown as Table 10. As P value was not significant, it means that an insignificant regression relationship existed between operation performance factors and overall performance satisfaction. When Adjusted R^sup 2^=0.088, it means that these three performance factors could explain the difference of overall internal performance by 8.8%.

4. Conclusion and Suggestions

Interviews with manufacturers, a questionnaire survey and statistical analysis were used to assess the success of the implementation of ERP and PDM in semiconductor related industries in Taiwan, as well as the views of those in the industry regarding its implementation.

4.1 Conclusion

4.1.1 Research results

According to the results of the correlation analysis of corporate capital, software purchase capability increases with company capitalization. Correlation analysis of the implementation inclination of higher level managers indicates that the long-term partnership of both sides concerning ERP and PDM implementation gets better with increasing level of the inclination to implementation.

Moreover, the correlation analysis of repetitive investment and waste shows that product delivery shortens and operation costs reduce when repetitive investment and waste is decreased. Additionally, correlation analysis of customer satisfaction reveals that enhancing customer satisfaction can control supplier supply records, increase material turnaround rate and shorten product delivery. Furthermore, regarding the issue of whether or not business reengineering should be conducted before implementing ERP and PDM, the F value of 1.565 and the P value of 0.226, means that business reengineering prior to implementing ERP and PDM has little impact. Factor analysis identifies three dimensions during the introduction process, namely communication, technical exchange and system organization. Meanwhile, factor analysis of ERP and PDM implementation reveals four dimensions of operation performance, system management, organization coordination and supply management. According to the results of the multiple regression analysis among implementation performance, system implementation cognition, technical exchange and communication, none of the F values are significant, indicating that there is no clear relationship between integral performance factor and integral performance satisfaction. The F values of the multiple regression analysis among operation performance, communication, technical exchange and cooperation and system implementation cognition were all insignificant, indicating that no significant relationship exists between operation performance factor and integral performance satisfaction.

4.1.2 Site interviews

According to site interviews with semiconductor related manufacturers in Taiwan, these manufacturers did not use any special introduction procedures to cope with customer requirements. System introduction does not necessarily mean introducing all ERP modules. The main concern of manufacturers is how to integrate ERP with the software used internally for information sharing. Besides, some manufacturers that introduce ERP and PDM to cope with customer requirements consider information exchange between the customer end and the supply chain when selecting software and systems.

4.2 Suggestions

Based on these results, those manufacturers who introduced ERP and PDM and achieved excellent performance are not all large-scale enterprises with extensive capital. Small and medium-sized enterprises outside of semiconductor related industries with capital of between 20 million and 25 billion also may implement ERP. Some suggestions for enterprises that are about to implement ERP and PDM are:

1. It is better to implement PDM before ERP because this approach can save on personnel training and system and hardware adjustment costs.

2. Internal information systems in enterprises should be conducted before implementing any management system, to reduce the cost of introducing that management system.

3. Communication and learning are important during the implementation of ERP and PDM, since knowledge management is closely related to ERP and PDM implementation. As a result, software suppliers and enterprises should incorporate the ideas of knowledge management into ERP and PDM to design software that can integrate closely with the enterprise culture, and should implement methods that can encourage or establish a positive learning atmosphere within an enterprise.

[Reference]

References

[1] Bragg, Simon, "ERP for Manufacturers", Cambashi Ltd., http://www.cambashi.co.uk/ERP.htm, 1997.

[2] Cyert, R.M. and March, J.G.. Behavioral theory of the firm. Prentice-Hall, Englewood Cliffs, NJ, 1963.

[3] Davenport, Thomas H., "Putting the Enterprise into the Enterprise System", Harvard Business Review, Jul-Aug., 1998, 121-131.

[4] Delone, William H. and Ephraim R. McLean. Information systems success: the quest for the dependent variable. Information Systems Research, 3(1), 60-95, 1992.

[5] Gould, Lawrence, "Planning and scheduling today's automotive enterprises," Automotive Manufacturing & Production, 109(4), pp.62-66, 1997.

[6] Guthrie, A.. Attitudes of user-managers towards MIS. Management Informatics, 3, 5, 1974.

[7] Khalil, O. E. M. and Elkordy, M. M.. The relationship between user satisfaction and systems usage: empirical evidence from Egypt. Journal of End User Computing, 11(2), 21-28, 1999.

[8] Leong, K. K., Yu, K. M. and Lee, W.B.. A. Product data allocation for distributed product data management system, Computer in Industry, 47, 289-298, 2002.

[9] Leong, K. K. ,Yu, K. M. and Lee, W. B.. A security model for distributed product management system. Computer in Industry, 50, 179-193, 2003.

[10] Market Intelligence Center (MIC), http://mic.iii.org.tw/index.asp. Access date: 2003-10.

[11] Larcker, D. F. and Lessig, V. P.. Perceived usefulness of information: A psychometric examination. Decision Science, 11(1), 121-134, 1980.

[12] Na Dong-Gil, Jang Wooseung, Kim Dong-Won, Development of a resource planning system for compound semiconductor wafer manufacturing, Journal of Materials Processing Technology, 138(1-3), July 20, pp. 372-378, 2003

[13] Oh, Y, Han, S. H., and Suh, Y. Mapping product structures between CAD and PDM systems using UML. Computer-Aided Design, 33, 521-529, 2001.

[14] Power, R.F. and G.W. Dickson. MIS project management: myths, opinions and realities. California Management Review, 15(3), 147-156, 1973.

[15] Ptak, C. A.. ERP tools, techniques, and applications for integrating the supply chain. St. Luice Press. APICS, 2000.

[16] Ralph Kappelhoff, Integration of ERP to the final control elements, ISA Transactions. 36(4), pp229-238, 1998.

[17] Saleem, N.. An empirical test of the contingency approach to user participation in information systems. Journal of Management Information System, summer, 13(1), 145-166, 1996.

[18] Sum, Chee-Chuong, James S K Ang, Lei-Noy Yeo., "Contextual0elements of critical success factors in MRP implementation," Production & Inventory Management Journal, 38(3), pp.77- 83, 1997.

[19] Taiwan Semiconductor Industry Association (TSIA), "http://www.tsia.org.tw/tsia_En/". Access date: 2003-10

[20] Tim Minahan, Enterprise resource planning: strategies not included, Purchasing, p.112-127, 1998.

[Author Affiliation]

Wen-Tssan Lin

National Chin-Yi Institute of Technology, Taiwan

Hui-Jen Yang

National Chin-Yi Institute of Technology, Taiwan

Ming-Yi Lin

National Chin-Yi Institute of Technology, Taiwan

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.