Effective production control systems are important for any manufacturing plant. In order to obtain effective production control, several different production control systems have been proposed. MRP, Kanban and CONWIP are the most popular systems among these. In this paper, a simulation analysis compares the three control systems implemented in a semi-continuous manufacturing environment: a cold rolling plant. Simulation results show us that the CONWIP production control system is the most effecient of these. It can greatly decrease the WIP (Work-In-Process), average inventory and average inventory costs, while providing a higher throughput rate and facility utilization.

The CONWIP production control system has received a great deal of attention from researchers recently. In this paper, we introduce a method to determine the card number of the CONWIP system for a production line with a bottleneck. The simulation study compares the CONWIP system and the original control system for the four situations in a cold rolling plant. Simulation results show that the CONWIP production control system is very elcient for the production and inventory control of semi-continuous manufacturing, such as that found in the steel rolling plant. It can greatly reduce the work-in-process (WIP), decrease the average inventory and average inventory costs, and guarantee a higher throughput rate and facility utilization.

With the rapid development of Internet and the emerging of global economic, VE (Virtual Enterprise), which is the alliance of stand-alone enterprise and usually organized in project mode, becomes a new paradigm of enterprise operation. On the one hand, VEs can help enterprises to respond rapidly to market demand by sharing capabilities, resources, and so forth. One the other hand, enterprises in a VE face more risks than conventional enterprises. Moreover, the risks in VE are always dynamic. Hence, risk management for VE has received many attentions. Risk evaluation is one of the key phases for the risk management. In this paper, focus on the characteristics of project organization mode and dynamic risk of VE, the evaluation model of dynamic risk for single procedure in VE is presented based on the Markov process. In the VE, the operation of each procedure is composed of different process, and there are many states within each process. The procedure passes these states randomly and the time needed, the quality and the cost needed within each state are different, so the risk is dynamic. The Markov process is used to describe this process, and case study suggests the effective of the method.

Production inventory-control is one of the most important aspects of a flexible manufa Cturing system (FMS) design. CONstant Work In Process (CONWIP), which is a hybrid of push-and-pull type systems, offers an alternative to effective utilization of the expensive FMS equipment while still meeting customer requirements, In the selection of an FMS control method, material handling often becomes one of the capacity-constraints which forms the basis of various resarch interests, In this paper, a structure-based model for a CONWIP-controlled FMS is proposed, and within it, the node type characteristics concept is used to describe the constraints in FMS. Furthemore, simulation is used to determne the card number based on the structure-based model. The simulation results demonstrate that the model is suitable for the design and operation of FMS. The model can be used as a manufacturing execution system of enterprise resources planning. An architecture for this integrated design based on Intemet/Intranet systems is also proposed.

根据电子商务环境下企业风险的特点，在建立企业风险模糊描述机制的基础上，基于模糊ISODATA聚类法，对企业风险管理的关键问题——企业风险分类进行研究。实例表明该方法是有效的。

针对OKP企业的特点，对不同类型的CONWIP系统进行了仿真研究，结果表明多阶段混联CONWIP控制策略控制效果最佳。

虚拟企业是未来企业的潜在发展模式，虚拟企业的风险管理是当前研究的热点问题，风险规划是风险管理的重要阶段之一。本文针对虚拟企业缺少历史资料、不确定因素多等特点，提出了非线性整数规划和模糊综合评价相结合的风险规划问题描述模型以及遗传算法与模糊综合评价相结合的问题求解算法。实例仿真验证了该方法的有效性。该方法能够实现在一定风险费用投人的情况下，通过优化组合风险处理措施，达到虚拟企业整体风险水平最低的目标，是虚拟企业风险规划的科学管理方法。

Dynamic alliance and virtual enterprise (VE) are essential components of global manufacturing. Minimizing risk in partner selection and ensuring the due date of a project are the key problems to overcome in VE, in order to ensure success. In this paper, a risk-based partner selection problem is described and modeled. Based on the concept of ine6cient candidate, the solution space of the problem is reduced e7ectively. By using the characteristics of the problem considered and the knowledge of project scheduling, a rule-based genetic algorithm (R-GA) with embedded project scheduling is developed to solve the problem. The performance of this algorithm is demonstrated by a problem encountered in the construction of a stadium station and the experimental problems of di7erent sizes. The results of this trial demonstrate the real life capability of the algorithm.

Production inventory control is essential problem of manufacturing systems. System Dynamic (SD) is more suitable for simulation analysis of the manufacturing. In this paper, the system dynamic is used for modeling and analysis the CONWIP controlled lamp production line. The description of the problem is proposed, the SD model is elaborated and the simulation analysis is evaluated. The results demonstrate the effectiveness of the SD model.

Virtual enterprise is the potential mode of future enterprise under global economic environment and the risk management for it is a popular research area recently. Risk identification is the key phrase of risk management. Considering the risk identification problem in the virtual enterprise, a new identification method based on the BP network is provided in this paper. Using this method and the data related to the problem, the possible to take place of the risks could be obtained by training the neural network, so that risks can be investigated in time and managed effectively. The simulation results prove its validity.