SBIR Phase I: Dynamic Risk-Based Planning and Scheduling

这个小企业创新研究（SBIR）第一阶段的研究项目将确定新的软件工具的可行性，这些工具利用创新的原则为供应链管理人员提供比目前可用的更有效的计划和执行工具。这些从本研究中产生的新工具将提供一种综合的方法，可以与现有的ERP/SCM系统接口（仅用于数据），然后，利用新开发的供应链随机模型，为供应链计划者提供少量关键操作措施和控制。此外，这些新模型解决了在现有SCM产品中一直存在的关键错误。这个新工具，动态基于风险的计划和调度（DRPS），期望简化在计划层面可以做出的选择，如在不同的日子增加或减少产能，或推迟到期日期)。因此，这些可以被列举出来，从而产生一个鲁棒和最优的解决方案。如果成功，该方法将解决困扰研究人员40多年的最优调度问题。DRSP提供了一个产品流仪表板，使公司计划人员能够实时查看关键变量，并提供有用和强大的控制。这一创新通过在规划和执行之间提供有意义的直接联系，有可能彻底改变整个行业。此外，它还提供了一种方法，可以使用几个关键度量（总库存、错过发货的概率）来预测未来的性能，并使用几个关键控制（增加产能、提前工作、移除产能）。这样一个系统将有几个明显的优点，因为它：1)它是动态的，因为它可以自我纠正需求和供应的随机波动，并在需要注意的重大变化发生时发出信号；2)比目前使用的模型更准确，同样的服务需要更少的库存；3)系统明确考虑了风险和供应链的随机性；4)提供直观的规划，而不会给规划者带来不必要的细节负担；5)扩大可受益于拉动式生产的制造环境范围。

This Small Business Innovation Research (SBIR) Phase I research project will determine the feasibility of new software tools that utilize innovative principles to provide supply chain managers with more effective planning and execution tools than are currently available. These new tools emanating form this research will offer a comprehensive methodology that can interface to an existing ERP/SCM system (for data only) and then, utilizing newly developed stochastic models of the supply chain, provide a supply chain planner with a small number of key operational measures and controls. Additionally, these new models address critical mistakes that have been perpetuated in existing SCM offerings. This new tool, Dynamic Risk-based Planning and Scheduling (DRPS) anticipates to simplify the choices that can made at the planning level such as adding or reducing capacity on various days, or pushing out due dates). Consequently, these can be enumerated generating a solution that is both robust and optimal. If successful, the proposed approach will have solved the optimal scheduling problem that has eluded researchers for more than 40 years. The DRSP provides a Product Flow Dashboard that enables company planners to see critical variables in real time as well as provide useful and robust controls. This innovation has the potential to revolutionize the industry by providing a meaningful direct link between planning and execution. Moreover, it provides a means to use a few key measures (total inventory, probability of missed shipments) to predict future performance and as well as using a few key controls (add capacity, work ahead, remove capacity). Such a system would have several distinct advantages in that it: 1) it is dynamic in that it is self correcting to random fluctuations in both demand and supply and signals when a significant change occurs that requires attention; 2) Is more accurate than models used today requiring less inventory for the same service; 3) The system explicitly considers risk and the stochastic nature of the supply chain; 4) Provides planning that is intuitive without burdening the planner with unnecessary details; and 5) Extends the range of manufacturing environments that can benefit from pull production.