Determining Criteria of Human Resource Information System that Affect Human Resource Performance in Companies Using DEMATEL-Based ANP Method

Decision making in HR activities is delayed because of insufficient information. Human Resources (HR) and information system are the important roles in helping company HR works. Objective of this study is toidentifycriteria of Human Resource Information System (HRIS) that affect human resources in companies by using DEMATEL-based ANP (DANP) method.HRIS criteria are collected from several studies to be validated by experts in HR of manufacturing companies. The survey was conducted with 5 experts and their judgments are calculated in limit supermatrix. The limit supermatrix of DANP showed that 18 criteria in HRIS. This study obtained five most important criteria of HRIS in HR performance are high quality data presentation, quick and precise, accessible, information need in time, and fulfill needs of HR.


Introduction
Human Resource (HR) affect the development of an organization/company.HR problems that are often encountered such as the retention of human resources, the lack of human resources development, poor employee relations, traditional administrative system, manual payroll system, and lack of integrated system. Information system (IS) is a combination of information technology that use technology to support operational and management activity [1]. There are two important sources in the organization, human resources and information systems (IS).HR and IS require significant management to affect the overall performance of the business. Insome developed countries, HRIS weremanaged independently rather than using outsourcing services. The purpose of this study is to identify the criteria of HRIS (Human Resource Information System) that affect the performance of human resources in companies using DANP method in manufacture companies. A research conducted in UK showed modules of HRIS in the UK have been used to help facilitating the performance of human resources in the company [2].

Methodology
The questioner in this studywas reviewed by 5 expert judgments. All experts are HR manager with more than 10 years of experience in their area. Geometric mean value of criteria obtained from expert judgments that has more than 3.75 are accepted. There are 4 dimension and 18 criteria selected by experts, as follow: System quality dimension (D1) consists of four criteria: accessible (C1), quick and precise (C2), easy to use (C3), and multipurpose (C4). Information quality dimension (D2) consists of five criteria: output in useful format report (C5), comprehensive information (C6), up-to-date information (C7), information need in time (C8) and accurate information (C9). User satisfaction dimension (D3) consists of three criteria: high quality data presentation (C10), fulfill needs of HR (C11), and satisfaction (C12). Information system success dimension (D4) consist of six criteria: enhance human resource planning (C13), online system (C14), career improvement (C15), salary advice (C16), improve the assessment and training needs (C17) and enhance performance management (C18). The second queationaire of this research were distributed to 10 HR experts in different manufacturing companies and used Likert scale from 1 to 5.
The result of the respondents were generated to be an averaged matrix (A) as shown in Table 1.  The averaged matrix (A) are processed to limit supermatrix to obtain global weight. This following steps show the process: After generating average matrix (A), we form Matrix (X) which is the result of an overall matrix of direct-indirect influence. Then, we would like to find the Total Influence Matrix (T)from the multiplication of Matrix X. Furthermore, we form unweighted supermatrix (WD and WC) to generate local weight and weighted supermatrix ( ). The last step is to form a limit supermatrix to obtain a stable supermatrix and generate a global weight for the criteria.

Result and Discussion
The result of DANP method is the limit supermatrix and the global weight for the criteria. Limit supermatrix process needs normalization to make sure that the unweighted supermatrix (WD and WC) have same unit. Table 2 summarizes all the weight by DANP method. Information quality (D2) has the greatest local weight dimension (0.3268) while information system success (D4) has the smallest local weight (0.1570). Moreover, high quality data presentation (C10) also has the greatest global weight while career improvement (C15) has the smallest global weight (0.0152). The greater weight of local dimension and global weight meana higher level of importance than the other dimension dalam rangka mengambil keputusan dalam bidang HR berdasarkan sistem informasi yang ada.
According to the result of DANP method, information quality dimension helps HR to improve their performance, such as an up-to-date information will make all activities in HR like selection of new employee, promotion of employee, general affair activities, transportation and lodging expenses, and payroll system becomes easier. Information quality generates accurate data presentation that helps HR to have a valid data, so the human errors in HR activities will be minimized.Quick decision making in HR activities would improve corporate's performance.

Conclusions
As a conclusion, there were four dimensions at this research, that were system quality dimension with 4 criteria, information quality dimension with 5 criteria, user satisfaction with 3 criteria, and information system success with 6 criteria. The most important dimension was quality information (0.3268). The five biggest global weight from 18 criteria were high quality data presentation (0.1089), quick and precise (0.0884), accessible (00834), information need in time (0.0708), and meet your needs (0.0691).