Supply Chain Analytics

供应链分析

• 批准号: 538313-2019
• 负责人: AlumurAlev, Sibel
• 金额: $ 1.82万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=681183
• 关键词: Supply; Chain; Analytics

As part of its plans on creating a digital supply chain ecosystem, Unilever Canada aims to utilize big data, data analytics, and smart technologies to continuously collect data about their supply chain activities, and process/analyze the collected data to improve their supply chain practices. Unilever Canada's plans to achieve this goal include engaging in research collaborations with academia to develop data analytics methods and tools to process/analyze data to predict future adverse events (e.g., supply chain disruptions such as stock outs or late delivery) and design a robust supply chain and logistics network. In this context, the proposed research collaboration first aims to analyze Unilever Canada's demand data to specify the regional and temporal patterns in the demand using data clustering techniques. This will help to divide Unilever Canada's customer domain into demand zones, which will be analyzed separately for determining factors or trends that may be associated with adverse events. These findings will then be used to develop risk prediction models to i) estimate the risk of adverse events disrupting the supply chain performance, and ii) to determine the statistically significant factors affecting the estimated risk. Finally, all of the above findings will be used in location analysis for a new distribution center to increase the efficiency and flexibility of Unilever Canada's supply chain network in addressing possible supply chain disruptions.

作为创建数字供应链生态系统计划的一部分，联合利华加拿大旨在利用大数据、数据分析和智能技术不断收集有关其供应链活动的数据，并处理/分析收集的数据，以改善其供应链实践。联合利华加拿大公司实现这一目标的计划包括与学术界开展研究合作，开发数据分析方法和工具，以处理/分析数据，预测未来的不良事件（例如，供应链中断（如缺货或延迟交货），并设计一个强大的供应链和物流网络。在这种情况下，拟议的研究合作的第一个目的是分析联合利华加拿大的需求数据，指定的区域和时间模式的需求，使用数据聚类技术。这将有助于将联合利华加拿大的客户领域划分为需求区域，并将分别进行分析，以确定可能与不良事件相关的因素或趋势。然后，这些发现将用于开发风险预测模型，以i）估计破坏供应链绩效的不良事件的风险，以及ii）确定影响估计风险的统计学显著因素。最后，所有上述研究结果将用于新配送中心的位置分析，以提高联合利华加拿大供应链网络的效率和灵活性，解决可能的供应链中断问题。