GOALI: Implementing Global Factory Schedules in Automated Facilities

GOALI项目旨在开发一套完整的进度和结构控制器，以实施全球进度计划。调查人员不会探索制定全球时间表的新方法，而是寻求制定一种系统的办法来执行这些时间表，其隐含的假设是，全球时间表准确地反映了管理优先事项。调度控制器将全局调度产生的作业优先级作为输入，在面临中断的情况下对作业进行重新调度，以最大限度地减少与全局调度的偏差，同时提高系统性能。结构控制器管理系统中的工作分配，以避免系统易受中断影响的状态。因此，结构控制器的功能是通过保持作业相对于生产和材料处理能力的“良好”分配，使调度控制器的任务更容易。研究人员将制定一种新的方法来重新安排问题，在利用全局计划中的信息的同时，尽可能保持其性能。将调查避免系统状态的结构控制政策，即在工厂内分布极易受到干扰的批次。这项研究将在半导体晶片制造的背景下进行，在那里，调查人员拥有丰富的经验，而工业合作伙伴是半导体行业此类软件的主要供应商。因此，该团队是唯一有资格解决这一内在跨学科问题的团队，并为其提供新颖、实用的解决方法。

This Grant Opportunities for Academic Liaison with Industry (GOALI) project is to develop an integrated set of scheduling and structural controllers to implement a global schedule. The investigators will not explore new methods of developing global schedules, but rather will seek to develop a systematic approach to implementing them, with implicit assumption that the global schedule accurately reflects management priorities. The scheduling controller takes the job priorities developed by the global schedule as input, and reschedules jobs in the face of disruptions to minimize deviations from the global schedule as well as improve system performance. The structural controller manages the distribution of work in the system to avoid states in which the system is vulnerable to disruptions. Thus the function of the structural controller is to make the tasks of the scheduling controller easier by maintaining a "good" distribution of jobs relative to production and material handling capacity. The researchers will formulate a novel approach for rescheduling problems, which exploit the information in a global schedule while maintaining its performance as closely as possible. Structural control policies that avoid system states, i.e., distributions of lots throughout the factory that are unduly vulnerable to disruptions will be investigated. The research will be pursued in the context of semiconductor wafer fabrication, where the investigators have extensive experience and the industrial partner is the primary supplier of such software to the semiconductor industry. Hence the team is uniquely qualified to address this inherently interdisciplinary problem and present novel, practical approaches for its solution.