Big Data Analytics: Optimization Models and Algorithms with Applications in Smart Food Supply Chains and Networks

大数据分析：优化模型和算法在智能食品供应链和网络中的应用

• 批准号: RGPIN-2020-06792
• 负责人: Hassini, Elkafi
• 金额: $ 4.52万
• 依托单位: McMaster University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=741436
• 关键词: Big; Data; Analytics; Optimization; Models

A data set is defined as big data if it is large in volume (e.g., measured in petabyte or exabyte units), grows with a high velocity (e.g., real time data from Internet of Things sensors) or comes in a variety of forms (e.g., from video, images and text). Big Data analytics is a three pronged process that uses data to: describe the trends and patterns in the data using visualization and graphics techniques; predict future patterns using statistical and data mining tools; and prescribe alternatives for decision makers using optimization and simulations tools. With the wide availability of data, both in large volumes and high-speed streaming data, supply chain partners are looking for ways to harness this data for optimization. In this proposal we introduce the concept of big data shared analytics and focus on developing innovative methodologies and solutions for complex problems for big data shared analytics in smart supply chains. This 5-year research program includes two theoretical areas 1) Optimization in decentralized environments with big data, 2) Multi-level optimization in decentralized environments and two application areas: 3) Sustainable operations in smart food supply chains, and 4) Network Analytics.  The impacts of this research program will be felt at two levels. 1) From a scientific point of view, we will introduce new complex problems in the growing field of optimization with big data that have not yet been treated in the scientific literature. We will study their fundamental properties and develop new models and algorithms to solve them. 2) From an economic and societal point of view, this research program addresses problems of major importance for supply chain planners. Our models and algorithms will integrate and balance economic, environmental and social concerns. In our research on collaboration logistics platforms we will propose new models for new problems in an important area of the digital economy: shared collaboration platforms. Finally, our research in the area of food supply chains will provide companies with tools to increase their efficiency while at the same time lowering their impact on the environments and the society at large. Our proposed research will support in training 20 highly qualified personnel (HQP) over five years. These HQPs will be trained in big data analytics and help fill the marketplace needs in this growing area.

如果数据集的容量很大（例如，以拍字节或艾字节为单位测量），以高速度生长（例如，来自物联网传感器的真实的时间数据）或以各种形式出现（例如，从视频、图像和文本）。大数据分析是一个三管齐下的过程，它使用数据来：使用可视化和图形技术描述数据中的趋势和模式;使用统计和数据挖掘工具预测未来的模式;使用优化和模拟工具为决策者提供替代方案。随着大量数据和高速流数据的广泛可用性，供应链合作伙伴正在寻找利用这些数据进行优化的方法。 在本提案中，我们引入了大数据共享分析的概念，并专注于为智能供应链中的大数据共享分析复杂问题开发创新方法和解决方案。这个为期5年的研究计划包括两个理论领域：1）大数据在分散环境中的优化，2）分散环境中的多级优化和两个应用领域：3）智能食品供应链的可持续运营，4）网络分析。该研究计划的影响将在两个层面上感受到。1)从科学的角度来看，我们将在不断增长的大数据优化领域引入新的复杂问题，这些问题尚未在科学文献中得到处理。我们将研究它们的基本性质，并开发新的模型和算法来解决它们。2)从经济和社会的角度来看，该研究计划解决了供应链规划人员的重大问题。我们的模型和算法将整合和平衡经济，环境和社会问题。在我们对协作物流平台的研究中，我们将针对数字经济的一个重要领域：共享协作平台中的新问题提出新模型。最后，我们在食品供应链领域的研究将为企业提供提高效率的工具，同时降低对环境和整个社会的影响。我们建议的研究将支持在五年内培训20名高素质人员（HQP）。这些HQP将接受大数据分析方面的培训，并帮助满足这一不断增长的领域的市场需求。