Supply Chain Planning for Random Demand Surges: Reactive Capacity and Safety Stock

• 成果类型: Article
• 署名作者: Huang, Lu; Song, Jing-Sheng; Tong, Jordan
• 署名单位: Alphabet Inc.; Google Incorporated; Duke University; University of Wisconsin System; University of Wisconsin Madison
• 刊物名称: M&SOM-MANUFACTURING & SERVICE OPERATIONS MANAGEMENT
• ISSN/ISSBN: 1523-4614
• DOI: 10.1287/msom.2016.0583
• 发表日期: 2016
• 页码: 509-524
• 关键词: supply chain risk management; inventory planning; demand surges; Markov modulated demand; reactive capacity; safety stock; stochastic comparison

摘要

Globalization, innovation, social media, and exposure to natural and man-made disasters have increased organizations' need to cope with demand surges: random, significant increases in demand in an otherwise relatively stable demand environment. To build supply chain capabilities, organizations face a choice between two fundamentally different sourcing strategies-reactive capacity and safety stock. We develop a framework to guide the joint sourcing strategy that minimizes the long-run average cost under a target service level. A salient feature of our modeling framework is a novel demand model that captures important continuous-time, non-Markovian characteristics of demand surge trajectories. In addition to the total magnitude as typically modeled in the literature, we define several other metrics of surges, including duration, intensity, compactness, peak position, volatility, and frequency. The resulting optimization problem is challenging because it requires evaluating, for any sample path, whether surge demand can be satisfied at every point in time-a refined feature that traditional models do not have. To identify the optimal strategy, we first characterize the optimal production and deployment policy for any given strategy and then transform the original problem into two equivalent but more tractable problems. Finally, through stochastic comparison techniques, we show how the magnitude and predictability of surge demand characteristics (mentioned above) and the cost profiles of each strategy impact the optimal joint strategy.