SBIR Phase I: Cluster-Based Forecasting of Intermittent Demand for Service and Spare Parts

这个小型企业创新研究(SBIR)第一阶段项目将在预测备件和服务部件需求方面取得进展，这种需求通常是“间歇性的”，很难预测。间歇性需求的特点是零值的百分比很高，并夹杂着经常具有高方差的随机非零整数需求。该项目超越了传统的一次一项的方法，转向了一种新的方法，该方法基于识别和利用响应共同驱动因素的部件集群。该项目更广泛的影响/商业潜力将是改善对拥有大量备件或服务部件库存的公司的需求规划和库存管理。这是许多公司资产负债表上的第二大项目，其支出约占美国国内生产总值(GDP)的8%。由于投入了如此多的资本，最大限度地减少库存成为当务之急。另一方面，随着所有部门的周期缩短和竞争加剧，保持足够的库存以满足服务水平承诺也是至关重要的。这两种压力要求准确的需求预测，以确定库存的“甜蜜点”。

This Small Business Innovation Research (SBIR) Phase I project will advance the state of the art in forecasting the demand for spare and service parts, which is typically "intermittent" and difficult to forecast. Intermittent demand is characterized by a high percentage of zero values, interspersed with random nonzero integer demands that often have high variance. This project moves beyond the conventional one-item-at-a-time approach toward a new methodology based on identifying and exploiting clusters of parts which respond to common drivers. The broader impact/commercial potential of this project will be improved demand planning and inventory management for companies with large inventories of spare or service parts. These are the second largest item on the balance sheets of many companies, and spending on them amounts to roughly 8% of the U.S. Gross Domestic Product. With so much capital invested, minimizing inventory becomes a priority. On the other hand, with shorter cycle times and greater competition across all sectors, maintaining sufficient inventory to meet service level commitments is also essential. These twin pressures require accurate demand forecasts in order to determine the inventory "sweet spot".