*本文选自知优科技于Civil + Structural Engineer 2021年8月刊文. Civil + Structural Engineer是土木和结构工程领域的国际权威性杂志，主要刊登前沿工程和建筑项目，与顶级土木、结构和多学科工程公司保持长期合作，是专业领域内的意见领袖。
By George Broadbent, Dr. Jeff Chen, Dr. Eve Lin with Kai Yin, and Jiayi Yan
With the advancement of building and information technologies, Building Information Modeling (BIM) applications have expanded from design and construction phases to support the operational stage of the building life cycle. The buzz term “digital twin,” has quickly spread throughout the Architecture, Engineering, Construction, and Operation (AECO) industry as a way to describe the extension of the digital model as a replica of the physical building for use during the operational phase. While the potential advantages of a digital twin are acknowledged, there is still a lot to investigate and explore in terms of the practical applications and transition from design and construction BIM to the operational digital twin phase.
Therefore, this five-part series dives into articles on in-depth topics regarding the practice implementation for facility management. In the first article of the series, we give a high-level background of the digital twin, followed by an overview of the fundamental needs of information representation during the facility management phase, which is different from the design and construction stages. Correspondingly, it introduces a “3X3 Principle,” which is derived from beyond Level of Detail (LOD) to tailor different LOD data analytics and reporting for different user groups, in order to more effectively support facility management during the operational phase.
Brief History of Digital Twin
The digital twin is an old and, yet, new topic, which is, as its name implies, a virtual representation that serves as the real-world, real-time digital replica of a physical object or process. As early as 2010, NASA initiated its attempt to utilize the digital twin concept to improve physical model simulation for its spacecraft. The manufacturing field then adopted the methodology in the era of Industry 4.0. However, it was not until recently that technology advanced to a stage where a digital twin is within reach for every industry.
The digital twin has played an essential role in multiple fields and achieved significant success as a game-changer. The AECO industry continues to embrace BIM and has realized significant improvements in efficiency as a result. BIM adoption has expanded into a global trend and led the industry to extend the BIM lifecycle from the project delivery phase to the operational phase, establishing it as a critical technological component of the digital twin.
In today’s AECO industry, the concept of a digital twin has gone through a progressive transformation, from snapshots in time to real-time interactive environments. Simultaneously, academic institutions are expanding studies to leverage a digital twin to benefit building operation and maintenance. The term digital twin has been coined to convey its capabilities as a holistic solution to plan, predict, report, and visualize building or infrastructure through the integration of data science, high-fidelity modeling technologies, and state-of-the-art technologies like Internet of Things (IoT), Artificial Intelligence (AI), and Geographic Information System (GIS) to name a few.
BIM-Enabled AM/FM Practices
BIM-enabled Asset and Facility Management (AM/FM) workflows have been studied and implemented across industries during recent years. A host of research has addressed the importance of BIM during the design and construction stages, particularly at the building level. BIM has been widely recognized as an invaluable digital approach that facilitates the data exchange in the project delivery lifecycle for all stakeholders. Most applications of these data include but are not limited to design coordination, clash detection, construction simulation, and cost estimation. When a BIM application is expanded to the operational phase, the story becomes a bit more complicated. Despite the availability of several established BIM-enabled AM/FM use cases, most of them focus on connecting the Building Automation System/Building Management System (BAS/BMS) and data analytics/reporting.
Most current BIM practices emerged from the standpoint of what technologies can offer rather than what the users want. This technological-centric approach presents a subtle solution design philosophy dilemma that focuses heavily on technology capability and forces the users to adopt the new workflow that might not be the most effective means. On the other hand, a user-centric approach would be a preferred method that involves process optimization and tailors the workflow based on users’ needs. Most of the existing BIM-enabled AM/FM solutions lack adaptiveness and resistance for more complicated services and functions that would meet stakeholders’ expectations. For example, some BIM to FM workflows heavily rely on the sufficiency and accuracy of the as-built model, without acknowledging that a BIM model is often not the main repository of all asset-related models. It is unfair and impracticable to push the client to change their workflows and business processes. Instead, we should take a more practical and user-centric approach to make the workflow adaptive and flexible enough to match the client’s needs.
This leads to discussions of whether this BIM/FM collaboration could be the right way to serve the industry.
Level of Detail (LOD)
One of the examples is the design and usage of (LOD) Level of Detail. The use of LOD is an interesting topic on the road of BIM to FM transition. LOD can stand for both Level of Detail and Level of Development, wherein the former relates to the way a model looks, while the latter refers to the depth of thinking embedded in the model.
Level of Development is a more recognized version of LOD that enables unbalanced development of model details and data (Level of Development = Level of Detail + Level of Data). The thinking behind it came from the understanding developed during the BIM lifecycle from implementation to operation which pushes people to rethink the true purpose of a BIM model (as-built or as-constructed). For a project that does not follow an Integrated Project Delivery (IPD) plan covering AM/FM, the as-built model could be LOD 500 on the modeling details but miss most of the AM/FM required data like serial numbers, manufacturers, etc.
Taking another point of view, LOD was developed from the perspective of project delivery – which means it speaks design/construction language. Even with its definition expanding, AM/FM professionals or owners could still be confused. Most of the time, in real-world operation and maintenance practices, the demands for data sufficiency are much higher than modeling granularity. In which case, LOD needs to be explicated even further to ensure the fulfillment of AM/FM’s data requirements. Moreover, in all cases, BIM dimensions (4D, 5D, 6D, etc.) are inevitable terminology that goes with LOD. The stacking of project delivery-oriented terminologies does not help the AM/FM stakeholders or owners better understand what data they can expect by the end of construction.
Introducing the digital twin into the AECO industry raises the bar of LOD for facility operations to a new height. In this scenario, the phrase digital twin explains itself very effectively, and institutions and facility owners will get the point without a second thought – it’s a digital replica of the physical assets. As a result, we need to have a new set of vocabulary terms that define the adaptability of the digital twin’s LOD and how it is used in the operations phase. These terms would focus on asset information requirements and the accuracy of data rather than geometric types.
Digital Twin in AECO
The emerging trend of digital twins allows for the introduction of new possibilities and solutions that meet the currently unsatisfied needs from both methodological and technical angles. Due to the complexity of AM/FM scenarios, various stakeholders’ needs, limitations of existing data collection, and handoffs, there is no common agreement on digital twin definitions yet. Therefore, the digital twin implementation needs to be tailored to suit specific end users. It is essential to ensure that the functional requirements are collected to map the multi-stakeholder expectations. This is accomplished through system architecture and uses software development techniques to organize complex relationships of tasks. An example is a goal-oriented engineering approach for data warehouse development. Moreover, introducing state-of-the-art technologies like AI, IoT, and Big Data can maximize BIM’s potential in the AM/FM phase to enhance the efficiency of operations.
A digital twin is a systematic approach that establishes the program from a bottom-level common data environment and policy. It reaches each part of the value chain with a data flow that mitigates data risks during the handoff along the lifecycle of the business process.
The Need for Multi-Stakeholder Solution
In addition to LOD, there are several emerging topics that state the current needs of multi-stakeholders to build an efficient and effective building/infrastructure-level digital twin system during the operation and maintenance stages.
• The need to bypass interoperability issues:
– The data gap between construction and actual operation causes data loss in the BIM lifecycle implementation and creates substantial difficulties in efficiency, effectiveness, and sufficiency of project information. A common data environment needs to be established and agreed upon to regulate the flow of information and enforce Quality Assurance/Quality Control (QA/QC) to ensure data interoperability.
• The need for system integration:
– Many different systems, such as Enterprise Resource Planning (ERP), Customer Relationship Management (CRM), BMS/BAS, security management systems, and maintenance systems, are deployed during the operational phase. An integration solution plays a critical role in connecting all isolated systems to perform queries and reports more synergistically.
• The need to support sustainable operations and monitoring:
– Working closely with BAS/BMS, sustainability consideration is essential to creating an environmentally friendly, physically comfortable, and safe building/infrastructure for its occupants. Different thresholds need to be examined and set up meticulously to trigger notifications or alerts for decision making.
• The need for quantitative feedback and assessment:
– Data analytics are required based on accumulated data from different systems to generate assessment grades for professionals and non-professionals to learn the current operational status of a building/infrastructure or a specific location/asset.
• The need for visualization to assist navigation:
– 3D visualizations of all disciplines are necessary to locate equipment for maintenance staff and help to complete emergency simulation, especially when alarms occur. 3D information is the most important way to improve problem-solving efficiency.
The digital twin’s development aims to satisfy stakeholders’ different levels of requirements during the operational phase and provide intuitive, efficient, and intelligent functions for defined users. To this end, users are categorized into three groups. There are property owner-level stakeholders who make major investment decisions, management teams that take responsibility for daily asset and facility management tasks, and the technician team that deals with daily work orders. A 3×3 principle has been created to implement this concept into the digital twin framework.
The Proposed AM/FM Digital Twin Framework – The 3X3 Principle
Based on the discussion above, this article introduces the high-level framework that is designed to serve the challenges facing the AECO industry while integrating all technologies and processes using a modular programmatic structure. One of the critical topics this series will address is the adaptiveness of the framework and how it can be used for different levels/types of stakeholders.
For example, the information requirements will differ from a facility operator to a facility director and from a leasing officer to a budget estimator. Although it contains all the data and information, the digital twin also needs to be capable of adapting to perform data analytics and aggregations and report in a way that fulfills the end user’s needs without overwhelming them or creating unnecessary distractions.
As a result, a 3×3 principle has been developed to be applied throughout the entire framework development. A short description has been listed below. Detailed use cases will be included in future installments.
• 30 Seconds
– Intent: Gather as much facility performance information as possible with minimum effort and distractions.
– Example: Facility Director will take 30 seconds for a quick glance at the facility dashboard/big screen to absorb enough knowledge and information to draw a big picture of the structure’s overall condition, including energy, security, leasing, operations, and maintenance, etc., to perform decision making.
• 3 Minutes
– Intent: Locate problems and provide corresponding responses with clear notification and guidance.
– Example: Facility Manager will take 3 minutes to review a notification or alert on the dashboard across the entire facility covering multi-disciplines and decide what the next step should be.
• 30 Minutes
– Intent: Provide explicit rich asset information for the executor to perform required tasks.
– Example: Facility Operator will take 30 minutes to dive into specific issues assigned to create service requests, send the right personal, and collect the required information.
In the following installments, the previously discussed framework will be divided into four issues that cover all five areas with real-life case studies and best practices. The articles are intended to trigger innovative discussions regarding the cradle-to-cradle digital twin lifecycle solution that makes AM/FM more effective, efficient, and predictive.