《航运行业数字孪生体》译与评

来源：数字孪生体实验室原创

撰写：陈志强

Decision making is the central activity of allorganizations, and decision makers use explanatory or causal models eitherimplicitly or explicitly. They decide based on the anticipated effects of theirintervention. Decision making is typically improved by open sharing of decisionmodels with colleagues and calibrating them with data. The value of a decisionmodel is often determined by the quality and breadth of data used for creationand calibration. The vast and growing Internet of Things (IoT) will be a keysource of real-time data for model building and reality assess ment.

决策制定是所有组织的核心活动，决策者隐式或显式地使用解释或因果模型。他们根据干预的预期效果做出决定。通常通过与同事公开共享决策模型并用数据校准它们来改进决策。决策模型的价值通常取决于用于创建和校准的数据的质量和广度。庞大且不断增长的物联网 (IoT) 将成为用于模型构建和现实评估的实时数据的关键来源。

Thesimplest and most common decision model is based on a measured associationamong variables in a data set. Methods such as regression and machine learningfit this mould. A more advanced approach is to test interventions prior toimplementation, such as with pilot studies and experiments, aimed to validate acausal model before scaling to a larger population. Randomized fieldexperiments have been applied to test key economic principles, and the keyproponents were awarded the 2019 Nobel prize in economics. The problem with interventions is thatsome don’t work and might harm the subjects, such as when testing new drugs, orjeopardize financial sustainability, such as infrastructural investments for aport that are below the intended return. The most rigorous approach to decisionmaking is to build a hi-fidelity mathematical or biochemical model, or digitaltwin,of the environment of concern and simulate a possible range of interventions.This enables the exploration of counterfactuals, such as what if we did xinstead of y. Such models do not physically harm humans or nature and provide aconceptual foundation for decision-making for future sustainable businessoperations. You can distinguish these three approaches, respectively, as building a theory from data, testing a theory in one or more real settingsthrough interventions, or testing a theory many times using a digital twin tosimulate many possible settings. The latter is the least risky and likely to bethe most successful.

最简单和最常见的决策模型是基于数据集中变量之间的测量关联。回归和机器学习等方法适合这种模式。更先进的方法是在实施之前测试干预措施，例如试点研究和实验，旨在扩展到更大的人群之前验证因果模型。随机现场实验已被应用于检验关键经济学原理，主要支持者获得了2019年诺贝尔经济学奖。干预措施的问题在于，有些干预措施不起作用，可能让受试者受到伤害，例如在测试新药时，或危及财务可持续性，例如低于预期回报的港口基础设施投资。最严格的决策方法是针对相关环境构建高保真数学或生化模型或数字孪生模型，并模拟一系列可能的干预措施。这使得探索假设事实成为可能，例如如果我们做 x 而不是 y 会怎样。此类模型不会对人类或自然造成物理伤害，并为未来可持续业务运营的决策提供概念基础。我们可以分别区分这三种方法，如根据数据构建理论、通过干预在一个或多个真实环境中测试理论，或使用数字孪生多次测试理论以模拟许多可能的环境。后者风险最小，而且可能最成功。

Digitaltwins require the construction of a precise set of equations for each componentin the model and the interaction among the components. They also need data fortheir calibration and operation. As the digital transformation of the maritimesector proceeds, it can also create the data required to calibrate digitaltwins of the various components of a ship, a port, and other elements of thetransport infrastructure such as the goods being transported (as e.g. dry andreefer containers). In many industries, included shipping, there are “emergingopportunities to digitally represent and simulate objects and events prior todecision making”. As more devices become connected, such as a s mart containerwith data generated by diverse use cases (e.g., executed transit time,deviations alerts, and infrastructure utilization associated to containermovements and operations), digital data streams built upon standardized datasharing provide opportunities for real-time representation and simulation ofauthentic situations. Digital twins will displace simulation models because ofthe order of magnitude increase in the fidelity of representation of thephysical world and their continual recalibration via digital data streams tolocal conditions and changed circumstances.

数字孪生体需要为模型中的每个组件以及组件之间的交互构建一组精确的方程，除此以外还需要用于校准和操作的数据。随着海事领域各单位数字化转型的进行，它还可以创建校准船舶、港口和运输基础设施的其他要素（例如正在运输的货物（例如干货和冷藏集装箱）；在包括航运在内的许多行业中，“出现了在决策之前以数字方式表示和模拟对象和事件的新兴机会”；随着越来越多的设备连接起来，例如一个智能容器，其中包含由不同用例生成的数据（例如，执行的运输时间、偏差警报以及与容器移动和操作相关的基础设施利用率），基于标准化数据共享构建的数字数据流为真实情况的实时表示和模拟。数字孪生体将取代模拟模型，因为物理世界的表示保真度的数量级增加以及它们通过数字数据流根据当地条件和变化的环境不断重新校准。

Inthis article, we elaborate on the key fundamentals of digital twinning followedby how it may improve the decision making of shipping companies, port operatorsand others in the transport and shipping ecosystem, as well as in developingstandards, that support both the integration of transport supply chainoperations and the development of digital twins for operational enhancement andstrategic planning.

在本文中，我们详细阐述了数字孪生体的关键基础，以及它如何改进运输和航运生态系统中航运公司、港口运营商和其他人的决策以及制定标准，这既支持了运输供应链运营的整合，也支持开发用于运营增强和战略规划的数字孪生体。

Adigital twin is a dynamic digital representation of an object or a systemdescribing its characteristics and properties as a set of equations. Complexprocesses involving a multitude of actors are often difficult decision-makingenvironments that are best modelled digitally prior to action. A digital twinincludes both the hardware to acquire and process data and the software torepresent and manipulate these data. Digital twins are more powerful thanmodels and simulations because they leverage digital data streams to bridge thebarrier between the physical entity and its representation. This means thatdigital twin a nalytics relies on historical data (e.g., a data lake), andreal-time digital data streams (e.g., IoT generated data), to a nalyze possibleoutcomes (Figure 1). A digital twin is a generic model of a situation that canbe tailored to a specific situation by specifying relevant parameters to provideanswers to “what happens if …” or “what happens if this does not …” to supportdecision-making.

数字孪生体是对象或系统的动态数字表示，将其特征和属性描述为一组方程。涉及众多参与者的复杂流程通常是艰难的决策环境，最好在行动之前进行数字建模。数字孪生体包括获取和处理数据的硬件以及表示和处理这些数据的软件。数字孪生体比模型和模拟更强大，因为它们利用数字数据流来弥合物理实体与其表示之间的障碍。这意味着数字孪生体分析依赖于历史数据（例如，数据湖）和实时数字数据流（例如，IoT 生成的数据）来分析可能的结果（图 1）。数字孪生体是一种情况的通用模型，可以通过指定相关参数来针对特定情况进行定制，以提供“如果……会发生什么”或“如果不……会发生什么”的答案，以支持决策。

图 1：数字孪生体的组成部分

Adigital twin can be continually calibrated through its entire lifecycle byintegrating real time digital data streams. This also means that a model can becontinuously refined to so that it converges to a very high-fidelity model ofreality.

通过集成实时数字数据流，可以在整个生命周期内持续校准数字孪生体。这也意味着可以不断改进模型，使其收敛到非常高保真度的现实模型。

Traditionally,we have used data modelling to surface that core component within a standardand to ensure compatibility across standards. This has been followed by effortsof defining standardised interfaces for communication, so-called APIs(Application Protocol Interfaces). Now, we need to recognize that data have a dualrole: transaction processing and data ana lytics, such as that facilitated by adigital twin. Thus, a digital twin is another use case that needs to besupported by standardized digital data streams using standardised APIs. We needto redesign business processes to support the generation of IoT derived datanecessary for digital twin creation and operation, so that they become powerfultools for risk management an alysis and mitigation, as well as effectivedecision making aids To prepare for the era of digital twins, standardisationbodies, such as UN/CEFACT, GS1, WCO, and DCSA have developed various buildingblocks in support of the digital twin concept, namely the UN/CEFACT SmartContainer data model and the DCSA IoT connectivity infrastructure. Extra standardsare still needed to build and deploy fully the digital twins. Standards need toserve both the transactions of today and the digital twins of the future. Threeareas of operations are now discussed for which maritime sector digital twinswould serve as an important foundation for strategic and operationaldecision-making to enable ecological and financial sustainable maritimetransport.

传统上，我们使用数据建模来显示标准中的核心组件，并确保跨标准的兼容性。紧随其后的是定义用于通信的标准化接口，即所谓的API（应用程序协议接口）。现在，我们需要认识到数据具有双重作用：事务处理和数据分析，例如由数字孪生体提供的便利。因此，数字孪生体是另一个需要使用标准化 API 的标准化数字数据流支持的用例。我们需要重新设计业务流程，以支持数字孪生体创建和运营所需的物联网衍生数据的生成，使其成为风险管理分析和缓解的强大工具，以及有效的决策辅助工具，为数字孪生体时代准备好标准化机构，如 UN/CEFACT、GS1、WCO 和 DCSA，开发了各种构建模块来支持数字孪生体概念，即 UN/CEFACT 智能容器数据模型和 DCSA 物联网连接基础设施。仍然需要额外的标准来完全构建和部署数字孪生体。标准需要为当今的交易和未来的数字孪生体提供服务。现在讨论了三个运营领域，航运行业数字孪生体将作为其战略和运营决策的重要基础，以实现生态和金融可持续的海上运输。

Digitaltwinning is an acknowledged opportunity for maritime sector improvement. “Thereis no doubt that the digital twin is the future. Being able to predictpotential dangers and create the optimum design, will enhance safety andoperation greatly. With the element of the unknown significantly limited, thedigital twin concept can help the shipping industry make better use ofdigitalization and move to a new era”. Three areas that will likely benefitfrom digital twins are fleet optimisation, port optimisation, end-to-end supplychain optimisation and increasing key stakeholders’ situational awareness,which we now elaborate upon.

数字孪生体是航运行业各单位改进的公认机会。“毫无疑问，数字孪生体是未来能够预测潜在危险并创建最佳设计，将大大提高决策安全性和操作性。在未知元素明显受限的情况下，数字孪生体概念可以帮助航运业更好地利用数字化，迈向新时代。”可能从数字孪生体中受益的三个领域是船队优化、港口优化、端到端供应链优化和提高关键利益相关者的态势感知能力，我们现在对此进行详细说明。

Fleet optimisation  船队优化

Typically,a shipping company serves multiple clients at the same time, and clients mayuse different shipping companies simultaneously. Thus, a shipping company needsto maintain and gain in competitiveness by optimizing its fleet in terms ofships and their cargo carrying capacity. This need for sensitivity a nalysiscould be served by a digital twin based on historical, ongoing, and predictionsof business transactions. This digital twin could form the basis of strategicdecision-making by testing a variety of scenarios for trade patterns andshipping fleets.

通常，一家航运公司同时为多个客户提供服务，而客户可能同时使用不同的航运公司。因此，航运公司需要通过优化其船队的船舶及其载货能力来保持和提高竞争力。基于业务交易的历史、正在进行和预测的数字孪生可以满足这种敏感性分析的需求。这个数字孪生体可以通过测试贸易模式和船队的各种场景来形成航运公司战略决策的基础。

Furthermore,a digital twin for fleet optimisation could also enhance operational decisionmaking under diverse contextual factors, such as weather conditions that createatypical situations, and various options need to be rapidly reviewed.

此外，用于船队优化的数字孪生体还可以增强航运公司在各种环境因素下的运营决策能力，例如天气条件产生非典型情况需要快速审查各种选项。

Port and terminal optimisation  港口码头优化

Portefficiency relies on balancing demand and supply in a flexible way andintegration within the entire transport system. A port is dependent of acontinuous inbound and outbound flow of cargo and passengers arriving anddeparting from the port by different means of transport. For strategicplanning, a port and its partners need to capture historical, ongoing, andpredicted future trade in a digital twin. Such a model should incorporate thedifferent parameters and relationships that port decision-makers should includein their strategic decisions, such as investment in infrastructure, portdesign, and terminal capacity. Typically, questions that such a model shouldaddress are how many berths are needed for the port need to meet punctualitygoals, or how much yard space is needed to allow for different customers tostore their cargo as it moves between transport services, either shipping orother modes.

港口运营效率依赖于以灵活的方式平衡需求和供应以及在整个运输系统中的整合。一个港口依赖于通过不同运输方式连续不断的实现货物和乘客进出港口。对于战略规划，港口及其合作伙伴需要在数字孪生体中捕捉历史、正在进行和预测的未来贸易。这种模型应包含港口码头决策者在其战略决策中应包括的不同参数和关系，例如基础设施投资、港口设计和码头容量。通常，这种模型应该解决的问题是港口需要多少个泊位才能满足准点率目标，或者需要多少堆场空间来允许不同的客户在运输服务（运输或其他方式）之间移动时存储他们的货物。

Adigital twin, fed by multiple data streams of real-time data and historicaldatabases, is also an operational planning tool for the coordination andsynchronization of port operations. It could be an essential foundation forvirtual arrival processes and green steaming and for the hinterland window tosupport efficient use of trucks, trains, and infrastructure for diverse needs.

由实时数据和历史数据库的多个数据流提供的数字孪生体也是港口运营协调和同步的运营规划工具。它可以成为虚拟到货流程和绿色航运以及腹地的重要基础窗口，以支持高效使用卡车、火车和基础设施以满足各种需求。

Situational awareness: short and long term  态势感知：短期和长期

Cargoowners, transport buyers, and end-customers seek enhanced visibility andpredictability on the state of the transport of goods in their movement fromorigin to destination. To enhance situational awareness for these groups, it isfeasible to consider a parallel linking of relevant digital twins so that therepercussions of a delay in one stage can be thoroughly ana lysed, adjustments m ade, and situational awareness updated. In addition, connected digital twinsare a tool for investigating the coordinated development of infrastructureinvestments across a web of ports that frequently interact so that keystakeholders also gain long-term situational awareness. This allows them tocollaboratively make decision to serve the common goals of the eco-system likeminimizing emission in ports. Understanding a complex interacting world isincreasingly beyond the cognitive capabilities of humans, and we must build anduse high-fidelity models of that world that enable them to perceive the stateof the present and the future.

货主、运输买家和最终客户寻求提高货物从始发地到目的地的运输状态的可见性和可预测性。为了提高这些群体的态势感知能力，可以考虑将相关数字孪生体并行连接起来，以便对某一阶段延迟的影响进行彻底分析、做出调整并更新态势感知。此外，互联数字孪生体是一种工具，用于调查频繁交互的港口网络中基础设施投资的协调发展，以便关键利益相关者也获得长期态势感知。这使他们能够协作做出决策，以实现生态系统的共同目标，例如最大限度地减少港口排放。理解一个复杂的交互世界越来越超出人类的认知能力，我们必须建立和使用那个世界的高保真模型，使他们能够感知现在和未来的状态。

Optimization of container flows in theend-to-end supply chain

端到端供应链中集装箱流的优化

Recently,s mart containers supporting IoT connectivity standards for have beenintroduced.9 There are numerous use cases for s m art containers that overcomesome of the pain points that the transport industry experiences. The datastreams generated by s mart containers are a valuable input for fleetoptimization, port and terminal optimization, and situational awareness aselaborated previously. Containers pass through many transport hubs and aremanaged by different carriers (of the same and different type) in theend-to-end supply chain. As a result, data generated by connected containers isa very valuable source for data for digital twins, whether retrieved from adata lake or handled realtime as a data stream.

最近，支持物联网连接标准的智能集装箱已经被推出。智能集装箱有许多用例可以克服运输行业遇到的一些痛点。如前所述，智能集装箱生成的数据流是船队优化、港口和码头优化以及态势感知的宝贵输入。集装箱经过许多运输枢纽，由端到端供应链中的不同承运人（相同和不同类型）管理。因此，连接容器生成的数据对于数字孪生体来说是非常有价值的数据来源，无论是从数据湖中检索还是作为数据流实时处理。

Adigital twin for supply chain optimization will provide transport buyers andcoordinators opportunities to optimize the choice of transport mode and routefor serving their clients. This should strengthen their strategic relationshipto transport producers, such as carriers and transshipment hubs. Furthermore, adigital twin will be a basis for optimizing the flow of empty containers.Connected containers are an electronic necessity for "s mart" supplychains and an essential building block for digital twins of supply chains.

用于供应链优化的数字孪生体将为运输买家和协调员提供优化运输模式和路线选择以服务客户的机会。这应该加强他们与运输生产商的战略关系，例如承运人和转运枢纽。此外，数字孪生体将成为优化空集装箱流动的基础。互联集装箱是“智能”供应链的电子必需品，也是供应链数字孪生体的重要组成部分。

Adigital twin is constructed by generic mathematical representations of manycomponents (e.g., a container crane, a container, the machine of a ship, and abollard) and their relationship with other components (e.g., a container craneunloading a container ship, the utilisation of a berth for a visiting ships).

数字孪生体是由许多组件（例如，集装箱起重机、集装箱、船舶机器和系船柱）的通用数学表示及其与其他组件（例如，卸载集装箱船的集装箱起重机、为来访船舶使用泊位）。

Thesegeneric representations are parameterized so they can be tailored to specificcircumstances, such as the unloading speed of a crane given its position, theposition of a container on a ship, and the prediction of berth slots occupiedby visiting ships.

这些通用表示是参数化的，因此它们可以根据特定情况进行定制，例如给定位置的起重机的卸载速度、集装箱在船上的位置，以及来访船只占用的泊位的预测。

Thosegroups with deep knowledge of each component, such as crane manufacturers, portinfrastructure designers, and ship designers, need to develop, or advise ondevelopment of, a standard model of their component. Standardized digitalmodels of all components in the shipping industry is the next wave ofstandardization if the industry is to achieve higher levels of capitalproductivity through a nalytics based operational and strategic decision making.The physical instances of all components need to have embedded sensors thatgenerate standardized data stream to calibrate their associated digital model.Current operation and future needs can be both guided by digital twins providedthe maritime industry cooperates to standardize digital data streams and modelsof digital components.

那些对每个组件有深入了解的团队，例如起重机制造商、港口基础设施设计师和船舶设计师，需要开发或建议开发其组件的标准模型。如果航运业要通过基于分析的运营和战略决策来实现更高水平的资本生产力，那么航运业所有组件的标准化数字模型将成为下一波标准化浪潮。所有组件的物理实例都需要嵌入传感器，生成标准化数据流以校准相关的数字模型。如果航运业合作标准化数字数据流和数字组件模型，则当前的运营和未来的需求都可以由数字孪生体指导。

陈志强 安世亚太战略合作部咨询顾问

二十年企业管理咨询和信息化规划，三年增材制造生态圈业务拓展。在企业质量管理、供应链管理、企业管理体系规划等方面积累了丰富经验。