داده کاوی برای بهبود انتخاب پرسنل و بهبود سرمایه انسانی : مطالعه موردی در صنعت فن آوری پیشرفته
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|4690||2008||11 صفحه PDF||سفارش دهید||6530 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 34, Issue 1, January 2008, Pages 280–290
The quality of human capital is crucial for high-tech companies to maintain competitive advantages in knowledge economy era. However, high-technology companies suffering from high turnover rates often find it hard to recruit the right talents. In addition to conventional human resource management approaches, there is an urgent need to develop effective personnel selection mechanism to find the talents who are the most suitable to their own organizations. This study aims to fill the gap by developing a data mining framework based on decision tree and association rules to generate useful rules for personnel selection. The results can provide decision rules relating personnel information with work performance and retention. An empirical study was conducted in a semiconductor company to support their hiring decision for indirect labors including engineers and managers with different job functions. The results demonstrated the practical viability of this approach. Moreover, based on discussions among domain experts and data miner, specific recruitment and human resource management strategies were created from the results.
Human capital is one of the core competences for high-tech companies to maintain their competitive advantages in the knowledge economy. Personnel recruitment and selection directly affect the quality of employees. Hence, various studies have been conducted on resumes, interviews, assessment centers, job knowledge tests, work sample tests, cognitive tests, and personality tests in human resource management to help organizations make better personnel selection decisions. Indeed, the existing selection approaches focus on work and job analysis that are defined via specific tasks and duties based on their static properties. However, owing to the changing nature of knowledge workers in high-tech industry, jobs cannot be easily delineated especially for jobs in the management level. As globalization and technology advance, cross-functional tasks are also increased while new jobs are also constantly created. The requirements of personnel quality in high-technology companies are increasingly strict, while the work processes in these companies are becoming diversified and complicated. Thus, the conventional personnel selection approaches that are developed on the basis of static job characteristics will no longer suffice (Lievens, Van Dam, & Anderson, 2002). In order to find the right people to do the right things for the right jobs, developing effective selection approaches is very critical. A high-tech industry such as semiconductor industry has many unique or unusual characteristics including complex and highly uncertain manufacturing processes, short product life cycles, low yield problems, and difficulties in acquiring human capital (Chien and Wu, 2003 and Sattler and Sohoni, 1999). Thus, the quality of their human resource is very crucial in increasing their competitiveness. In addition, Appleyard and Brown (2001) analyzed the firm-level data from semiconductor manufacturers in the United States, Asia, and Europe and found that engineers play important and growing roles in creating high-performance semiconductor factories. Nevertheless, semiconductor companies, as well as other high-technology companies, often suffer from high turnover rates and difficulties in recruiting the right talents. In order to attract good applicants, companies provide attractive compensation and welfare benefits. However, despite the willingness of many companies to do all that they can to recruit the best people, they usually have difficulties at the selection stage in predicting which applicants would have better work performance and would have longer service time after they are hired. Therefore, selecting the right engineers who can demonstrate the best performance and who will stay with the company for a long time is of great urgency for every high-technology company. Recently, owing to the advancements in information technology, researchers have developed decision support systems and expert systems to improve the outcomes of human resource management. In particular, data mining is recognized as one of the most salient topics. Data mining refers to the extraction of useful patterns or rules from a large database through an automatic or semi-automatic exploration and analysis of data (Berry and Linoff, 1997 and Chen et al., 1996). With the help of data mining techniques, computers are no longer limited to passively storing or collecting data. They can also help the users to actively excerpt the key points from huge amounts of data, and make use of analysis or prediction. Data mining techniques have been widely applied in many fields and have exhibited outstanding results. However, the applications of data mining in the semiconductor industry are mostly related to engineering data analysis and yield enhancement (Braha and Shmilovici, 2002, Kusiak, 2001, Chien et al., 2004 and Chien et al., 2007). Little research has been done in human resource management. This study aims to develop a data mining framework for personnel selection to explore the association rules between personnel characteristics and work behaviors, including work performance and retention. An empirical study for indirect labor (IDL) including engineers with different job functions in one of the world largest semiconductor foundry company located in the Hsinchu Science Park in Taiwan is studied to demonstrate the validity of this approach. In particular, we employ decision tree analysis to discover latent knowledge and extract the rules to assist in personnel selection decisions. Furthermore, using the information gathered, domain experts from this company can also generate recruiting and human resource management strategies. Some of the findings have been implemented in this company and the results have shown the practical viability of this approach.
نتیجه گیری انگلیسی
High-tech companies rely on human capital to maintain competitive advantages. This study developed a data mining framework to extract useful rules from the relationships between personnel profile data and their work behaviors. Furthermore, we developed useful strategies with domain experts in the case company and most of the suggestions have been implemented. With an effective personnel selection process, organizations can find the suitable talents at the first time to improve retention rate and generate better performance. In addition, the mined results have also assisted in improving human resource management activities including job redesign, job rotation, mentoring, and career path development. This study used applicants’ demographical data such as age, gender, marital status, education background, and work experience to predict their work performance and retention. Indeed, other demographical data and test results may be considered to improve the accuracy of the prediction or generate other potentially useful rules. Future study can be done to collect possible input variables such as address, the rank or scores in school, and number of owned licenses and to uncover buried relationships. Furthermore, although we have examined turnover reasons, the HR staff thought the derived rules to be divergent for implementation. Future research should be done to redesign the survey for collecting true turnover reasons for in depth analysis to help managers understand the root causes and thus take actions to effectively improve retention rate. Decision tree is used for data mining in this study because it is easier to understand and it offers an acceptable level of accuracy. The empirical study has shown practical viability of this approach for extracting useful rules for human resource management in the semiconductor industry. Alternative data mining techniques such as neural network can be studied in future research to compare various approaches and may thus integrate them for better exploration of complex interrelationships among the input personnel variables and target work behaviors. Furthermore, this methodology can also be applied to other jobs like operators or management level jobs, and to other industries to find matched talents to enhance human capital. The validated results can be integrated into the human resource information system (HRIS) as a preliminary screening mechanism for the large amount of resumes gathered from external recruiting channels such as internet to reduce the workload of recruiters and save on both visible and invisible costs.