China Paper Network /3/view- 1550800.htm
[Keywords:] human resources; Data warehouse; Multidimensional data model; decision support
doi : 10。 3969 / j . issn。 1673 - 0 194 .20 12 .07.026
【 China Library ClassificationNo. 】 f272.92tp31.13 【 ID number 】 a 【 ItemNo. 】1673-0194 (2012) 00.
1 quotation
Human resources are the available abilities and talents that an enterprise can regard as people who can make potential contributions to the creation and realization of its mission, vision, strategy and goals [1]. In the era of global competition and knowledge economy, human resources are the first resource and fundamental competitive advantage of enterprises, and human resource management has become an important force supporting the core competence and competitive advantage of enterprises and has a very important strategic position. However, at the same time, it is difficult to give full play to the strategic role of human resources because of the various levels of enterprise departments involved, complex business processes and huge data and information, and the extremely complicated and difficult management of human resources. Even though the business process has been optimized through the human resource management information system in recent years, the daily work efficiency has been improved, and the efficient and stable operation of human resource management has been realized at the business process level, there are still some problems to be solved urgently in human resource management at the strategic decision support level. These problems are manifested in two aspects: it is difficult to obtain decision information and the correctness is difficult to guarantee.
The main reasons why it is difficult to obtain and ensure the correctness of decision-making information are as follows: 1. Human resource decision-making information includes human resource inventory, human resource demand forecast, position information, personnel competency information, recruitment and configuration information, personnel performance information, salary information and training information. There are many kinds and complicated data; Secondly, the data organization mode of human resources internal management information system is oriented to specific business processes, not to the theme of decision analysis. Obtaining a certain decision information will inevitably lead to the re-extraction and induction of all relevant data, which is complicated and time-consuming; Thirdly, human resource management involves all departments and levels of the enterprise, and a large amount of decision-making information comes from the databases of other departments outside the human resource management department. The data in various external databases are not uniform in definition, granularity and updating strategy, so it is impossible to make statistics, induction and calculation directly, and it needs a long data preprocessing, which further increases the difficulty of obtaining decision information.
Aiming at the above problems, this paper puts forward a multidimensional data model of human resource data warehouse. With the help of data warehouse technology, the data is reorganized with the theme of decision analysis to realize the correct unification of data, thus solving the problems that decision information is difficult to obtain and correctness is difficult to guarantee.
2 data warehouse
In 1992, William H. Inmon put forward the definition and design method of data warehouse, which is very different from the traditional database in data organization and application objectives. Data warehouse has changed the form of data organization. By organizing data by subject and extracting subject information by dimension, the purpose of integrating and analyzing massive data is achieved. Data warehouse is a subject-oriented, integrated, time-specific and stable data collection, which is used to support the decision-making process in enterprise management [2]. According to the concept of data warehouse, data warehouse has the following four characteristics:
(1) subject orientation. A topic is an abstract collection of information that users pay attention to when making decisions. Data warehouses are organized by subject areas. Data warehouse organizes data with the theme as the center, which is beneficial to realize the goal of decision information support.
(2) synthesis. The data in the data warehouse is mainly used for analysis and decision-making, and it is necessary to summarize, sort out and synthesize detailed data. At the same time, most of the data in the data warehouse come from multiple heterogeneous data sources, so it is impossible to simply summarize and copy the data. It is necessary to clean and transform the data to keep the consistency of the data.
(3) It has the characteristics of the times. In order to ensure that the decision information provided by the data warehouse can be properly adjusted over time, the data warehouse needs to add new content over time and delete old data over time. The key codes of data tables in data warehouse all contain time items to represent the historical period of data.
(4) stability. Data warehouse changes with time, but it is relatively stable. The data in the data warehouse is mainly used for enterprise decision analysis, and the service provided is mainly data query. Once a certain data enters the data warehouse, it will be kept for a long time and generally will not be modified. Data in a data warehouse usually only needs to be loaded and refreshed periodically [3].
An analytical data warehouse needs to summarize and classify a large amount of data, and needs to complete complex queries quickly and flexibly according to customers' requirements, and the query results should be easy for users to understand. This requires that the data warehouse must eliminate data conflicts and inconsistent expressions when storing data; When faced with the user's query demand, in order to shorten the response time, the required information must be prepared in advance. In order to achieve this goal, the basic architecture of data warehouse includes at least three basic components: data acquisition, data storage and management and information access.
(1) data acquisition data acquisition refers to the process of obtaining original data from multiple heterogeneous data sources, obtaining subject-centered integrated data through extraction, cleaning, transformation and aggregation, and loading the data into the data warehouse.
(2) Data storage and management. Data storage and management is responsible for the internal maintenance and management of the data warehouse, and the services provided include data storage organization, data maintenance, data development and reasonable maintenance of the data warehouse.
(3) Information acquisition. Information access belongs to the front end of data warehouse. The front end of data warehouse is oriented to end users. End users can get information to support decision-making through the information access part. The information acquisition part generally consists of report generator, OLAP tool, data mining tool and visualization tool.
The basic architecture of data warehouse [4] is shown in figure 1.
3 Basic structure of human resource data warehouse
The end users of human resources data warehouse are the senior decision makers and daily staff of human resources department. Human resource data warehouse can provide decision-making information support for human resource strategic planning, competency analysis, training development management, salary management, recruitment and configuration management [5-7]. The basic structure of human resource data warehouse consists of data warehouse server, query server and front-end display tools [8]. The basic structure of human resource data warehouse is shown in Figure 2.
The information needed for strategic planning and daily management of human resources determines the analysis theme, dimension, granularity, measurement requirements and data extraction, transformation and loading rules of data warehouse. According to the rules of data extraction, transformation and loading, the human resource data warehouse extracts data from multiple external heterogeneous data sources, and then organizes the extracted data according to the theme, so as to query the server for data extraction. Finally, the front-end display tool extracts the corresponding data from the query server, organizes it into decision support information, and displays it according to the needs of end users.
4 human resource management process analysis
According to the modern enterprise human resource management theory, the human resource management system can be divided into seven parts: strategic planning, position management, ability management, recruitment and distribution, performance management, salary management and training management. The functional modules of human resource management system are interrelated and interact with each other, and operate systematically under the action of human resource management mechanism. The process of human resource management is shown in Figure 3.
Human resource strategic planning runs through the whole human resource management work and determines all links of human resource management work. First of all, the strategic planning of human resources determines the strategy, objectives, principles and policies of human resources by understanding the development strategy and strategic objectives of enterprises, analyzing the operating conditions and organizational structure of enterprises. Then, the strategic planning of human resources takes stock of the current situation of human resources, analyzes the differences between the current situation of human resources and the future strategic demand, finds the gap of human resources through the prediction of the supply and demand of human resources, and makes clear the specific problems faced by human resources planning. After the specific problems are identified, the human resource strategic planning will design systematic solutions to the human resource strategic problems, adjust the business functions of the human resource management system, make policy and institutional arrangements for the implementation of the human resource strategic planning, and formulate specific measures and action plans. Finally, human resource strategic planning will evaluate and control its specific implementation in real time to ensure the smooth implementation and timely adjustment of human resource strategic planning to meet the needs of enterprises and strategic development. In the process of determining strategic direction, identifying current problems, making specific plans and controlling the implementation of the plans, human resource strategic planning must be supported by strong information from specific functional departments such as position management, capacity management, recruitment and configuration management, performance management, salary management and training development management.
Based on the profound knowledge and understanding of the business structure, organizational structure and process of the enterprise, post management designs and constructs the system of functions, post categories and post categories to form a complete post system of the enterprise.
Ability refers to the potential and deep-rooted characteristics of an individual, which can distinguish those who perform well at work from those who perform poorly. It includes knowledge, cognitive or behavioral skills, attitudes or values, self-image, traits and motivations, that is, any individual characteristics that can be reliably measured or described. Competence should include the actual performance to be achieved and the personality characteristics needed to achieve these performances. Competency management defines employees' potential ability by analyzing their academic qualifications, professional titles and work performance, and provides a basis for further tapping their potential and cultivating their abilities. Competency management provides quality requirements for personnel recruitment and selection; It provides standards and basis for the allocation of human resources; Provide the most basic standard, basis and framework for the design of salary system.
The basis of recruitment and distribution is position management and ability management. According to the positions of enterprises and the internal characteristics of employees, effective recruitment methods, channels and selection methods are selected to realize the rational allocation of employees. Recruitment and allocation includes five links: enterprises find the demand for human resources acquisition, formulate the strategy for human resources acquisition, analyze the relative position of the labor market, recruit personnel, and select and allocate personnel, with the ultimate goal of personnel selection.
Performance management refers to the regular assessment of employees in order to ensure the realization of the business objectives of enterprises, analyzing the performance assessment standards and actual completion values of their posts, and making a correct evaluation of employees. Performance management includes the determination of objectives and plans, performance evaluation and performance appraisal. The determination of goals and plans refers to the formulation of goals and plans, and the definition of work content and evaluation criteria according to the requirements of enterprise strategic goals. Performance evaluation refers to the objective evaluation of employees' work according to pre-set evaluation indicators. Performance appraisal refers to analyzing the causes of the gap and formulating work improvement measures.
Salary management system is the core of the whole enterprise incentive mechanism and the key to attract and retain talents. Salary management can effectively adjust the salary system through the analysis of data such as personnel, positions and performance appraisal results, and realize effective control of labor costs and effective salary incentives. Salary management includes keeping the consistency of salary strategy with enterprise human resources strategy and enterprise business objectives, and improving the strategic management ability of salary management; Determine the appropriate salary level, ensure the external competitiveness and internal fairness of salary, and correctly handle the relationship between labor cost and human capital investment; Design multiple salary incentive elements and salary structure to reflect the characteristics of different types of employees and meet the uncertain, multi-level and complex employee needs; Establish a hierarchical and classified salary management system and dynamically adjust the salary so that the salary design can reflect the needs and labor characteristics of employees of different levels and categories; The design of salary mechanism and system is fair in procedure, and it is matched with other mechanisms and systems of human resources, especially with performance appraisal system and post qualification system; The design of salary system and management system should be legal and compliant, so as to avoid enterprises paying illegal costs.
By analyzing the training needs and the training effect reflected by historical data, the training development management makes a training plan for enterprise employees, ensures that the quality and skills of employees are improved to the maximum with limited training expenses, and realizes the maximum growth of enterprise benefits. Training development management is embodied in training base management, training budget management and expense accounting management. The specific work of training development management includes understanding and mastering the company's strategic development and the ability and quality of employees, and providing basis for the formulation of the company's training development plan; According to the potential characteristics of employees and the needs of enterprises, help employees make career development and personal ability development plans; Control the training implementation process; Evaluate the effect of training development.
The human resource management system is based on posts and personnel, with strategic planning as the main line, with competence management, recruitment and allocation, performance management, salary management and training development management as the main functions to ensure the normal operation of human resource management in enterprises and maximize the value of human resources. Through the analysis of the framework of human resource management, it can be concluded that the decision-making information needed in the actual operation of human resource management is organized around seven centers: position information, basic personnel information, competence, recruitment and distribution, performance, salary, training and development. Therefore, the human resource data warehouse will establish a multidimensional data model with the theme of position, basic information of personnel, competence, recruitment and distribution, performance, salary, training and development.
5 multidimensional data model of human resource data warehouse
Multidimensional data model is the best way for decision support data modeling, and it is also a commonly used data modeling method in data warehouse. Multidimensional data model organizes data in an intuitive way and supports high-performance data access. Each multidimensional data model is represented by multiple multidimensional data patterns, and each multidimensional data pattern consists of a fact table and a set of dimension tables. Multidimensional data modeling of data warehouse includes determining the theme, selecting facts and dimensions, determining granularity, and determining the attributes of dimensions and fact tables.
5. 1 Determine the subject area
Topic is a standard for classifying data at a higher level, and each topic basically corresponds to a macro-analysis field. Theme is an abstract concept, which integrates, transforms, classifies and analyzes the data in enterprise information system at a higher level. Through the analysis of the above-mentioned human resource management system architecture, the themes of human resource data warehouse are determined as position theme, personnel basic information theme, competence theme, personnel adjustment and allocation theme, training and development theme, performance appraisal theme and salary theme.
(1) Position theme. Position theme determines whether to add a new position by analyzing the job category, function and job category of the position in the current enterprise organization system, as well as the position and staffing information of employees, and designs and explains it.
(2) Subject of personnel basic information. The topic of personnel basic information forms an overall understanding of the current enterprise human resources by analyzing the information of all employees' academic qualifications, skills, professional titles and work experience, and provides data support for human resources inventory.
(3) Competency characteristics theme. The ability theme provides information support for personnel recruitment, selection, adjustment of configuration and salary through the analysis of basic information such as employee education, skills, professional titles and job adaptability.
(4) Personnel adjustment and configuration theme. The theme of personnel adjustment and allocation is to determine the reasonable transfer and adjustment of employees through the study of their competence and post functions.
(5) Training and development theme. Training development theme Through the analysis of employee basic information, training base information, training budget and expense management information, the training plan is made to realize the maximum value growth of human resources at the lowest cost.
(6) Performance appraisal theme. The theme of performance appraisal refers to the objective evaluation of employees according to the performance appraisal standards of their posts in order to ensure the realization of business objectives.
(7) Salary theme. By analyzing the ability, position and performance of employees, the salary theme can effectively adjust the salary system, achieve effective cost control and maximize the incentive effect.
5.2 Determine the particle size
The so-called granularity refers to the level of detail and hierarchy of data units in the data warehouse. The more detailed the data, the smaller the granularity and the lower the level; The higher the comprehensive degree of data, the greater the granularity and the higher the level. The granularity selected must meet the business requirements and be selected based on the data in the source database. The general granularity selection method adopts the lowest level of detail, that is, the granularity of data warehouse can be regarded as a row of records in each source database.
5.3 Select size
Dimension refers to the angle from which people observe things. According to the organization angle of information in the process of human resource management, the dimensions of each theme are selected as follows:
(1) Position subject dimensions include time dimension, personnel dimension, company dimension and job category dimension.
(2) The dimensions of personnel basic information include time dimension, education dimension, skill dimension, position dimension, professional title dimension and company dimension.
(3) The main dimensions of competence include time dimension, position dimension, personnel dimension and company dimension.
(4) The main dimensions of personnel adjustment and allocation include time dimension, position dimension, personnel dimension, performance dimension and company dimension.
(5) The dimensions of training development theme include time dimension, company dimension, position dimension, personnel dimension, training base dimension, professional title dimension and expense dimension.
(6) The main dimensions of performance appraisal include time dimension, company dimension, position dimension, professional title dimension, personnel dimension and appraisal index dimension.
(7) Salary subject dimensions include time dimension, position dimension, personnel dimension, company dimension, salary category dimension and appraisal dimension.
5.4 Establish a multidimensional data model
The human resource data warehouse adopts the star mode to build a multidimensional data model, and according to the selection of the above-mentioned subject fields and dimensions, the multidimensional data model is built, as shown in Figure 4~ Figure 10.
(1) Position theme. Position theme takes time, company, staffing information and position category as analysis dimensions, and provides comprehensive information such as position name, position content, position characteristics, position qualification and so on. The time dimension includes three dimensions: year, month and day. Staffing dimensions include personnel type, personnel level and personnel quantity. Company dimensions include company name, company category and company level. The job category dimension includes job category, department office and job level.
(2) Subject of personnel basic information. The topic of personnel basic information takes time, education, skills, position, professional title and unit as the analysis dimensions, and provides comprehensive statistical information such as the name, gender, date of birth, native place and quantity of relevant personnel. Education dimensions include school name, major name, education, graduation time, degree and awarding time. Skill dimension includes skill name, skill category and skill level. Title dimensions include category, level and review time.
(3) Competency characteristics theme. Competency quality takes time, personnel, post and unit as analysis dimensions, and provides comprehensive statistical information of personnel competency quality. Personnel dimensions include skill dimension, education dimension, name, gender, date of birth and place of origin.
(4) Personnel adjustment and configuration theme. The topic of personnel adjustment and allocation takes time, personnel, position, company and performance as the analysis dimensions to provide comprehensive statistical information of personnel adjustment and allocation. Performance dimension includes assessment results.
(5) Training and development theme. The theme of training development takes time, personnel, post, unit, professional title, training base and cost as analysis dimensions, and provides comprehensive statistical information on training plan and training effect. The dimensions of training base include usage time, training base name and training base category. The expense dimension includes expense budget and expense settlement.
(6) Performance appraisal theme. The theme of performance appraisal takes time, personnel, post, unit, professional title and appraisal indicators as analysis dimensions, and provides comprehensive statistical information of performance appraisal results. The dimension of appraisal standard includes appraisal standard.
(7) Salary theme. Salary theme provides comprehensive statistical information of salary quantity with time, personnel, position, company, salary type and performance appraisal as analysis dimensions. Salary categories include basic salary, performance salary increase, bonus, allowance, equity, security benefits, health and relief benefits, retirement benefits and paid vacation benefits.
6 conclusion theory
Aiming at the problem that it is difficult to obtain and ensure the correctness of human resource management decision information at present, it is an ideal solution to establish human resource data warehouse. Through the analysis of human resource strategic planning and daily business process, seven themes of organizing human resource work information are defined: position, basic information of personnel, ability and quality, personnel adjustment and allocation, training and development, performance appraisal and salary, and the dimensions of each theme are determined, and a multidimensional data model of human resource data warehouse is established, which lays a foundation for solving the problem of difficulty in obtaining human resource decision information and ensuring correctness.