A Framework for Integrated Production Planning and Scheduling in a Hybrid Assembly Job Shop Environment Under Uncertainty

9300568计算机技术的进步和计算机公司之间的竞争导致计算机辅助设计、计算机辅助工艺设计和计算机辅助制造的应用增加。虽然新技术的发展对于科学的进步是必要的，但缺乏这些技术的整合将证明对该系统不利，因此可能导致这些技术被抛弃。今天，人们比以往任何时候都更需要高效的生产管理技术。目前改进生产管理的方法集中在几种独立的改进方法上。这项研究的目标是在某种意义上发展一种“最大限度”的生产管理方法。“最大”方法是一种利用经理可利用的灵活性和响应的主要可变来源的方法：选择批量、工艺计划(要加工的零件特征)、用于创建特征的过程、每个过程的顺序、创建的库存、接受短缺或授予分包合同。这种方法适用于混合装配作业车间，该车间制造零部件，其中一些部件被送往装配线。由于建模和计算方法的进步、计算机能力的增强以及数据库组织的进步，这种方法现在是可能的。各个数据字段(时间标准、工艺计划、持有成本、可靠性和不良率数据等)仅限于研究人员已知和探索的实际制造系统中当前可用的那些。必须协调各个选择，因为优化单个变量(如批量)可能会使瓶颈资源超载，通过最短处理时间优化流程时间可能会增加拖期，并再次超载临时瓶颈。将快速模拟工艺规划、多准则优化、带补偿的随机规划技术与调度和排序技术相结合是一种创新，并显示出对现有理论的改进前景。

9300568 Foote The advancement in computer technology and the competition among computer companies have resulted in an increase in the application of computer aided design (CAD), computer aided process planning and computer aided manufacturing (CAM). Although the development of new technologies is necessary for the advancement of science, lack of integration of these technologies will prove detrimental to the system and hence may result in the desertion of such technologies. Today, there is greater need for efficient production management techniques than ever. Current methods of improving production management focus on several approaches as independent methods of improvement. The goal of this research is to develop in some sense a 'maximal' production management methodology. A 'maximal' methodology is one that takes advantages of the major variable sources of flexibility and response available to a manager: choosing lot size,process plan (which part feature to process), process used to create feature, sequence at each process, inventory created, shortage accepted or subcontract awarded. The methodology applies to a hybrid assembly job shop, which manufactures components some of which go to an assembly line. This approach is now possible due to advances in modeling and computational methodologies, increased computer power, and advances in data base organization. The individual data fields (time standards, process plans, holding costs, reliability and defect rate data, etc.) are limited to those currently available in actual manufacturing systems known and explored by the researchers. The individual choices must be coordinated since optimizing an individual variable such as lot size may overload a bottleneck resource, optimizing flow time by shortest processing time may increase tardiness and again overload a temporary bottleneck. The integration of fast simulation process planning, multiple criteria optimization, stochastic programming with recourse techniques with the scheduling and sequencing techniques is original and shows promise of improvements over the current theory.