Digitalisation for operational efficiency and GHG emission reduction at container ports

数字化可提高集装箱港口的运营效率并减少温室气体排放

• 批准号: EP/W028492/1
• 负责人: Dongping Song
• 金额: $ 5.43万
• 依托单位: University of Liverpool
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FW028492%2F1
• 关键词: Digitalisation; operational; efficiency; GHG; emission

Ports are regarded as concentrated areas producing air pollutants and greenhouse gas (GHG) emissions. Container ports play an important role in the global economy as they handle over 50% of seaborne world trade by value. Due to surging trade volume, disruptive events, and lack of coordination across relevant stakeholders, container ports often experience inefficiency and severe congestion. Port congestion creates the requirements for extra and unproductive moves when containers are stacking or retrieving, resulting in longer turnaround times for vessels and trucks.According to the Environmental Report 2019-20 produced by the Port of Felixstowe, about 60% GHG emissions (equivalent to 34.3K tons of CO2) from port operations originate from fossil fuelled yard cranes and internal trucks. The deployed fleet of trucks travels more than 14 million km a year, consuming about 4.2 million litres of diesel fuel per year and producing 26.5K tons of CO2 per year. The fleet of cranes consumes around 6.0 million litres of diesel fuel per year and generates nearly 7.8K tons of CO2 yearly. The port acknowledges that nearly 30% crane movement is unproductive, and improvements in yard management, reducing the empty travel time, can dramatically reduce both fuel consumption and GHG emissions (potentially by 15%, i.e. 1.5 million litres of fuel and 6.1K tons of CO2). This project applies digital technologies such as machine learning and optimisation techniques to develop a new decision support system to reduce unproductive crane movement and truck travel distance. As a result, the product productivity and efficiency will be improved, more containers can be handled within time windows, and vessel and truck turnaround times will be reduced. GHG emissions from trucks, ocean-going vessels and cargo handling equipment will be reduced. The project will directly benefit container ports, by improving ocean freight efficiency. The decision support system will work as a part of a physical and digital ecosystem which will facilitate the development of maritime autonomy and support the UK's transition towards 'zero-emission' shipping. The project will also indirectly benefit other stakeholders including shipping lines, rail operators and shippers, by automating process, reducing their costs, boosting trading volume and economic growth. Our innovation focuses on: (i) the pioneering attempt to apply digital technologies to predict import containers' out-terminals at the point when they are discharged from vessels to improve stacking operations; (ii) using the ground-breaking approach of combining predictive models with prescriptive models to support yard container allocation decisions; (iii) advance the knowledge on the relative importance of determinant factors (container attributes) to predict containers' out-terminals and quantify the contributions made by each factor to the prediction. The quantifiable information will inform maritime policy making, for example, introducing appropriate regulations or incentive programs, to encourage information sharing between ports and the stakeholders, so as to improve operational efficiency and reduce GHG emissions at ports.

港口被认为是产生空气污染物和温室气体排放的集中地区。集装箱港口在全球经济中发挥着重要作用，因为它们处理了50%以上的海运世界贸易价值。由于贸易量激增，破坏性事件以及相关利益相关者之间缺乏协调，集装箱港口往往效率低下，严重拥堵。港口拥堵导致集装箱堆放或回收时需要额外的非生产性移动，导致船舶和卡车的周转时间更长。根据费利克斯托港编制的2019-20年环境报告，港口运营中约60%的温室气体排放（相当于34.3万吨二氧化碳）来自化石燃料的堆场起重机和内部卡车。部署的卡车车队每年行驶超过1400万公里，每年消耗约420万升柴油，每年产生2.65万吨二氧化碳。起重机每年消耗约600万升柴油，每年产生近780万吨二氧化碳。该港口承认，近30%的起重机移动是非生产性的，改善堆场管理，减少空行程时间，可以大大减少燃料消耗和温室气体排放（可能减少15%，即150万升燃料和6.1万吨二氧化碳）。该项目应用机器学习和优化技术等数字技术开发新的决策支持系统，以减少非生产性起重机移动和卡车行驶距离。因此，产品生产力和效率将得到提高，更多的集装箱可以在时间窗口内处理，船舶和卡车的周转时间将减少。卡车、远洋轮船和货物装卸设备的温室气体排放将减少。该项目将通过提高海运效率直接使集装箱港口受益。该决策支持系统将作为物理和数字生态系统的一部分，促进海上自治的发展，并支持英国向“零排放”航运过渡。该项目还将通过自动化流程，降低成本，促进贸易量和经济增长，间接惠及其他利益相关者，包括航运公司，铁路运营商和托运人。我们的创新重点在于：（i）率先尝试应用数码科技，预测入口货柜卸船时的出运时间，以改善堆置运作;（ii）采用开创性的方法，将预测模式与规范模式结合，以支援堆场货柜分配决定;（iii）增进对决定性因素的相对重要性的认识（集装箱属性）来预测集装箱的出码头，并量化每个因素对预测的贡献。这些可量化的信息将为海事政策制定提供信息，例如，引入适当的法规或激励计划，以鼓励港口和利益相关者之间的信息共享，从而提高运营效率并减少港口的温室气体排放。

项目成果

Text classification in shipping industry using unsupervised models and Transformer based supervised models

• DOI: 10.48550/arxiv.2212.12407
• 发表时间: 2022-12
• 期刊: ArXiv
• 作者: Yingyi Xie;Dongping Song

Predicting out-terminals of import containers at seaports through data analytics: incorporating unstructured data and measuring operational costs induced by misclassifications

通过数据分析预测海港进口集装箱的出港：纳入非结构化数据并衡量错误分类引起的运营成本

• 发表时间: 2023
• 作者: Xie Y