员工培训排课系统遗传算法英文文献翻译 第7页
the EES model to design their course, layer by layer, including objects from each layer. Each object consists of one or more strategies to be implemented in order to achieve the learning objectives of the courses. This approach promises to increase the chances of successful quality and implementations.
Robinson, Lester, and Hamilton (1998) proposed a web-based method to facilitate teaching material transformation between different computer platforms. They designed teaching material utilizing web pages enhanced with Java applets, MPEG video clips and dynamic HTML. Gable and Page (1980) divided teaching design with the modern educational technology into four steps. Step one is programmed instruction, in which the instructional materials are divided into tightly connected learning units. Learning the previous unit is a prerequisite for progressing to the next unit, and all course units are connected sequentially. Step two is scrambled textbook using branching strategy. After the student answers a question, the system, in accordance with his or her choice, will jump to another learning unit. Step three is adaptive courseware, which introduces the concept of the adjustable course. The adjustable course ensures that the branching of the course is not only in accordance with the specific choice the student made on a certain question, but also the combination of the total performance of the student. And step four is generative, where the instructional materials are analyzed by artificial intelligence (AI), focusing on the needs of the student to create the most suitable materials (Liu & Lee, 1995).
Computer-managed instruction (CMI) was first introduced in the early 1970s (Lomerson & Knezek, 1991). Baker defined CMI as using computers to manage the teaching activity for individualized students (Alvarez-Valdes et al., 2002). Leiblum suggested many instructional and administrative functions for a CMI system such as setting the instructional objectives, arranging the instructional materials, providing tests and reporting learning progress (Leiblum, 1982). A CMI system is employed to manage teaching activities and knowledge through the use of computers. To be more specific, CMI is able to manage the teaching activity for knowledge, rather than to manage the teaching activity for skills and emotions. As the development of the information and communication technology, the design and application of CMI are improving and approaching to the ideal situation of individualized teaching. In the future, the goal for individualized education and training is to combine with the AI technology for administrative teaching and to implement the educational concept of the adaptive learning (Juang, 1996).
2.4 Genetic algorithms
Genetic Algorithms are adaptive heuristic search algorithms premised on the evolutionary ideas of natural selection and genetic. The basic concept of GA is designed to simulate processes in natural system necessary for evolution, following the principles of Darwinian evolution. Pioneered by John Holland in the University of Michigan, GA was shown to be an effective search algorithm (Lin, 1995, Lin et al., 1995 and Lin et al., 1997). It has been widely studied, tested and applied in fields of management science and engineering. In a comparative study, for example, Lin et al. (1995) compared five optimization algorithms for engineering problems and found that GA outperformed other optimization algorithms. 六维论文网毕业论文http://www.751com.cn/ 论文网http://www.Lwfree.com/
Using GA to solve the optimization problem, the objective function must be formulated to be a fitness function, which represents the fitness of a system to its outside environment, as the performance index of system. If the value of this fitness function approaches the optimization goal, the system performance is better. GA has developed to find the optimization solution with fitness by some of artificial operational process, which simulated natural selection and genetic, such as reproduction, crossover, and mutation. The basic principle of GA has developed from the simple genetic algorithm, SGA (Goldberg, 1989). GA is applied in a wide variety of research fields and is shown capable of finding optimal solution rapidly (Arroyo and Armentano, 2005, Chang and Chung, 2005, Elegbede and Adjallah, 2003 and Kumral, 2005).
3 Course structure analysis
For a trainee with limited training time, it is very important to receive the training as efficiently as possible. The available training time should be considered. The course combination is also configured according to each trainee’s knowledge and abilities. Furthermore, some courses with multiple knowledge aspects do not belong to only one category. For instance, a course on machine fault diagnosis utilizes many skills or knowledge to accomplish a diagnosis job, so the training material should be assigned specifically to reflect the characteristics.
The courses should be procedural for effective employee training, so trainees should take the training courses one by one, from lower levels to higher levels. However, this may not actually be true in training if the courses are not well constructed or the trainees need only applied skills. In the general enterprise training practice, most of the trainees already have basic concepts of the training subject, and only need further applied knowledge. Therefore, the training courses could be implemented focusing on applied skills, and not according to a clear procedure.
Weaver (1988) designed a training course, based on indicators of learning difficulty and importance, and found that more complex knowledge is, more difficult to learn, and this takes more time. Weaver defined the structure of course unit for knowledge analysis as four indicators: separability, complexity, importance and practicability. We modified Weaver’s indicators due to industrial practice. Each training course was assigned attribute values according to course significance, frequency, level, and training time.
The training courses in machine tool companies can be separated into three content categories: machinery, electricity and operation, based on designated maintenance tasks. Furthermore, two general course categories are aided for a complete course spectrum: programming and general. The manager expects the maintenance representatives to learn as many professionals as possible and is capable of fixing all models of machines. We considered course structure and the content of course to machine type.

上一页  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] 下一页