Next Article in Journal
The Influence Mechanism of Urban Spatial Structure on Urban Vitality Based on Geographic Big Data: A Case Study in Downtown Shanghai
Previous Article in Journal
Identifying Impact Factors of MEP Installation Productivity: An Empirical Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Buildings
Volume 12
Issue 5
10.3390/buildings12050566
buildings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

An Off-Site Construction Digital Twin Assessment Framework Using Wood Panelized Construction as a Case Study

by
Yuxi Wei
1,
Zhen Lei
1,* and
Sadiq Altaf
2
1
Off-Site Construction Research Centre (OCRC), Department of Civil Engineering, University of New Brunswick, Fredericton Campus, 3 Bailey Dr, Fredericton, NB E3B 5A3, Canada
2
ACQBuilt, 4303 55 Ave NW, Edmonton, AB T6B 3S8, Canada
*
Author to whom correspondence should be addressed.
Buildings 2022, 12(5), 566; https://doi.org/10.3390/buildings12050566
Submission received: 9 March 2022 / Revised: 20 April 2022 / Accepted: 25 April 2022 / Published: 28 April 2022
(This article belongs to the Section Construction Management, and Computers & Digitization)
Browse Figures
Review Reports Versions Notes

Abstract

:
Off-site construction is an innovative type of construction with the philosophy of standardizing the process and deploying the latest technological enablers. Many technologies, such as the Building Information Model (BIM), Internet of Things (IoT), etc., are concerned with virtual representation and manipulation of the physical site. However, a holistic view of the off-site construction processes is lacking in the exploration of the technological advances, resulting in inconsistency when applying these advances in practice. The concept of Digital Twin is useful for addressing this challenge. Digital Twin is a philosophy and a collection of technologies aimed toward seamless physical and virtual connections. Therefore, a holistic Off-site Construction Digital Twin model is necessary for any research concerning this topic, and an assessment framework is useful in helping off-site construction industry companies in approaching systematic Digital Twin. This research first proposes a model for Off-site Construction Digital Twin. To quantify this model, an assessment tool named Off-site Construction Digital Twin Maturity Level is proposed. The validation and evaluation of this assessment framework are conducted through a case study with ACQBuilt, an off-site construction company in Edmonton, Canada. The resulting assessment framework contributes to the body of knowledge in two ways: Firstly, it sets the foundation for an Off-site Construction Digital Twin, which is anticipated to significantly reduce waste and to improve efficiency. Secondly, it enables easier technology application in practice by offering a holistic Digital Twin framework.

1. Introduction

In the past decade, there has been an increasing interest in transferring philosophy and methodology from the manufacturing industry to the construction industry, as building construction is moving towards a combined process of off-site prefabrication and on-site assembly [1]. This process is facilitated both by the increasing awareness of the almost-flat productivity growth in construction, which is less tolerable given the fact that “productivity in manufacturing has nearly doubled” [2], and by the development of computer software and hardware that make it possible to keep the whole building project in perspective, thus allowing all project stakeholders to participate in the design process [3]. However, as new technologies such as the Building Information Model (BIM), Artificial Intelligence (AI), and Internet of Things (IoT) have enabled potential solutions to automation and inclusion of wider application scenarios, they also bring significant challenges in dealing with exponentially increasing project data [4]. According to interviews with off-site construction industry representatives, part of the data processing challenge lies in the fact that current construction processes are using different software, following a variety of workflows using different technologies, resulting in a situation termed “drowning in data”, in which lots of waste are produced in solving interoperability problems [5]. In this context, the concept of Digital Twin has shown great potential in addressing this challenge, as it proposes a holistic linked data paradigm in which projects are fully represented digitally [4,6]. First brought up in a university course on Product Life Cycle Management (PLM) in 2003, Digital Twin, as explicitly expressed in its name, is a virtual replica of its physical counterparts, which cooperates with real-time data flow to simulate and modify projects [6,7]. Explorations in Digital Twin are mostly conceptual research in the context of the manufacturing industry, showing that the idea and its application in the construction industry are still in their infancy, even though many agree on its potential [3].
This research is motivated by the goal of improving off-site construction productivity by deploying the concept of the Digital Twin. There are generally two types of off-site construction: panelized construction and modularization. The two methods are quite similar in both concepts and processes. The fundamental difference between the two is the prefabricated unit, which for the former is the structural panel, and for the latter box-like units [8]. This research utilizes wood-panelized construction as the starting point because it is one of the most utilized off-site construction methods in North America and shows better building performance, including more flexible architectural design, better performance in energy efficiency, and reusability [8,9]. Therefore, focusing on wood-panelized construction allows us to 1. work closely with industry practitioners to test applicability in common practice and to 2. keep research focused on the optimization of construction solutions.
Given the fact that construction workflow is not standard, the philosophy of Digital Twin is rarely discussed at a higher level of the whole off-site construction process. Research tends to focus on lower levels, where Digital Twins can be applied to one simple process, such as the creation of a Digital Twin for sheet metal punching machines [10]. Although these attempts are inspiring, an overarching research framework is still lacking. Given the basic Digital Twin elements being the virtual side, the physical side, and the connection between the two [3,6,7], the very first step is to establish the relationship between all these elements and to provide an assessment framework for each of the steps. This Off-site Construction Digital Twin assessment framework has three goals: 1. to identify the differences in Digital Twins between the manufacturing and off-site construction industries; 2. to create an Off-site Construction Digital Twin Model; 3. to create an assessment framework based on the model; and 4. to verify and evaluate the assessment framework and the model through interviews and empirical evidence. These goals are achieved using a mixed-method approach including a literature review, comparative analysis, and interviews. The research contributions include 1. paving the way for future research on Off-site Construction Digital Twin and 2. offering a systematic Off-site Construction Digital Twin framework to synthesize current technologies for easier and more holistic practical application.

2. Research Methodology

This study follows a four-step process, as shown in Figure 1. The first step is to conduct a literature review on both Digital Twins in the manufacturing industry and the construction industry. This step defines Digital Twin and indicates work to be conducted if applying the concept of Digital Twin to the construction industry. The next step is to develop an Off-site Construction Digital Twin model by 1. identifying Digital Twin processes in construction, especially in off-site construction, and 2. reviewing the understanding of Digital Twin retrieved from the previous step. The third step is to create a company assessment framework by integrating the Off-site Construction Digital Twin model developed in step 2 and other assessment frameworks and standards. The final step is to validate the assessment framework through questionnaire interviews with our industry partner ACQBuilt, a leading off-site construction company in Edmonton, Alberta, Canada. ACQBuilt owns the factory where all panels are prefabricated and installed on-site. There are two types of off-site construction: penalized construction and modularization. ACQbuilt specializes in off-site panelized-wood construction (Figure 2), which sets the focus and limit for this research, as the Off-site Construction Model and validation are both conducted through interviews with ACQBUILT.

3. Literature Review

This section is conducted in two parts: Section 3.1 reviews general Digital Twin in the manufacturing industry, and Section 3.2 reviews Digital Twin and related concepts in the construction industry.

3.1. Digital Twin (DT) in the Manufacturing Industry

Digital Twin was first mentioned in the context of the smart manufacturing industry. The authors of [3,6,7] have identified three components of a DT: physical products in Real Space, virtual products in Virtual Space, and the connection of data linking the two spaces. The purpose of this model is to create a digital counterpart of the physical product, for better understanding and control of the physical processes. When the concept first came up, limited information technologies were available for supporting either physical manufacture or virtual modeling. As enabler technologies advanced, in 2012, the National Aeronautics and Space Administration (NASA) brought up this concept and specified it as:
A Digital Twin is an integrated Multiphysics, multiscale, probabilistic simulation of an as-built vehicle or system and uses the best available physical models, sensor updates, fleet history, etc., to mirror the life of its corresponding flying twin [11].
Even though this statement was made within the context of air-force-vehicle production, it can be generalized for the construction industry as “an integrated Multiphysics, multiscale, probabilistic simulation of a building project and uses the best available physical models, sensor updates, historical construction data, etc., to mirror the life of its corresponding twin.” Although much emphasis may be put on the process simulation and entities in the physical space, it should be noted that the data flow between the physical object and the digital object should be fully integrated and automated in both directions; otherwise, it is only a “Digital Model” or a “Digital Shadow”, with the former indicating only data inflow from the virtual side to the physical side and the latter indicating data outflow from the physical world to the virtual model [3]. Although the virtual space is supposed to make decisions and modifications in the physical space, the physical twin should also be able to feed information back into the virtual twin and make modifications. In this research paper, Digital Twin refers to a complete picture of both sides, including the virtual model, its physical counterpart, and the connection between the two. It is noted that some others use the Cyber-Physical System (CPS) for the same definition, in which Digital Twin only refers to the virtual side of the model [12,13,14,15,16]. Both definitions work for this research purpose. To reduce confusion, this paper adopts the inclusive definition of Digital Twin, which refers to the entire system. However, the literature review still includes the ones whose keywords include “Cyber-Physical System”.
The potential ability and benefits of Digital Twin are immeasurable. Its potential abilities include real-time remote monitoring and control, greater efficiency and safety, predictive maintenance and scheduling, scenario and risk assessment, better team collaboration, more efficient and informed decision support team, personalization of products and services, and better documentation and communication [4,6,11,16,17]. Ref. [17] also mentions a similar concept of Digital Siblings, in which the siblings should be able to simulate “What-if” scenarios without the necessity of actual execution in the real world, and the two concepts share similar research initiatives. All the benefits are rooted in real-time data exchanges, which form the core of Digital Twin exploration.
Digital Twin is identified as a model comprising three parts: the physical, the virtual, and the connection (Figure 3), in which an ideal Digital Twin is constantly updated almost in real-time in all three parts [6,7,17,18]. Currently, the research work is either overall conceptual work or an implementation of one of the three parts. Refs. [12,13,14,15,16,18,19] focus on the systematic Digital Twin framework in the manufacturing industry. Ref. [12] develops 5C level architecture for Digital Twin, which comprises a smart connection level, data-to-information conversion level, cyber level, cognition level, and configuration level. The five levels indicate the level of maturity or advancement in approaching Digital Twin. Ref. [13] proposes architecture for the virtual side using web services. Ref. [14] proposes a model for a cyber-physical production system in an eight-tuple form: Input, Relation, Cyber-Physical System, Internet of Services, Internet of Things, Internet of Content and Knowledge, Internet of People, and Factory and Output. Ref. [15] extends the concept of a Cyber-Physical System to manufacturing practical applications using a set of virtual engineering tools. Ref. [16] proposes a multi-model data acquisition approach to minimize the delay in Digital Twin data acquisition. Ref. [18] proposes an application framework of Digital Twin for product lifecycle management; Ref. [19] proposes a process to create a “digital factory” under the structure of factory digitalization, capability extraction, and supply chain configuration. Many concepts are inspiring for a Digital Twin in the construction industry.
Apart from overarching framework research and review, other research focuses on one of the three parts of Digital Twin that can lead to a variety of more specific research questions and objectives. For example, within the virtual side, researchers deal with modeling, simulation, and visualization [10,13,16]. For the connection part of Digital Twin, researchers focus on data exchange, interoperability, and standards [20,21]. The major research problem in the physical side is usually related to either 1. the availability and cost of technologies to enable communication with the virtual side, or 2. the standardization of the processes; therefore, data collection is possible and meaningful [22,23,24]. All these are efforts made in the context of the manufacturing industry. Although some studies can be transplanted into the construction industry, others require major modification according to the construction processes.

3.2. Digital Twin in the Construction Industry

Even though this section is dedicated to a literature review on Digital Twin in the construction industry, the research keywords are not limited to “Digital Twin” or “Construction” for two reasons. Firstly, this concept is still in its early development, so there is not much research performed on this topic. Secondly, there are other concepts, such as Building Information Models (BIM), Cyber-Physical Systems, or Virtual Design Construction (VDC), that can be included in this section, as they align with the part of the bigger goal of Digital Twin.

3.2.1. Construction Digital Twin or Off-Site Construction Digital Twin

In exploring the concept of the Digital Twin, it is important to identify the relationship between the construction industry and the off-site construction industry. No two construction projects are the same. To create a virtual replica of the physical entities, it becomes necessary to have a higher-level process that applies to most projects. In this sense, off-site construction has, by nature, adopted the philosophy of Digital Twin, as the off-site construction industry started with the idea of improving efficiency by shifting from the traditional Design–Bid–Built process model to the Design–Fabricate–Assemble model [25]. They regard off-site construction as a “strategic commitment to support a new partnering arrangement”. Construction Digital Twin requires standard processes and entities. In off-site construction, all projects can be broken down into standard pieces that are fabricated in factories and assembled on-site, which makes “Construction Digital Twin” possible. Even though some research does not mention “off-site construction”, the construction process the authors are referring to are all, to some extent, “off-site construction” [26]. Therefore, off-site construction is a step forward to Construction Digital Twin, and all contents of “construction digital twin” are included in “off-site construction digital twin”. To be more accurate, this paper will use the term “Off-site Construction Digital Twin”.

3.2.2. Off-Site Construction Digital Twin Application

Keywords used in this literature review include “Digital Twin”, “Construction” “Building Information Modelling”, “Off-site Construction”, “Off-site Manufacturing”, “Construction Automation”, and “Virtual Design Construction”. This section first reviews general ideas and assessment methods for Virtual Design Construction and then reviews applications in the industry. Virtual Design Construction is reviewed in separate paragraphs because the concept is very similar to the virtual side of Digital Twin. In many research papers which focus on virtual simulation and decision making, “VDC” and “Digital Twin” are the same concept. For this research, the difference lies in the fact that VDC is completely virtual, missing the “physical entities” part proposed in the Digital Twin model [25].
Virtual Design Construction (VDC) is a higher-level concept highly related to the Construction Digital Twin, as it shares similar research goals to pursue full automation in the virtual world. Ref. [25] identifies three manageable aspects of the VDC model: (1) the products that can be managed, (2) the organizations that design, define, and operate, and (3) the process the organization follows. This structure is object-oriented, which is desirable for coding purposes, as some of the mainstream coding languages (JavaScript, c#, etc.) are also object-oriented. This POP model has four different levels of detail: A, B, C, and D, organized in the order of relative importance. This methodology is inspiring, as one of the big problems in performing research in the construction industry is the necessity to deal with excessive data instances. By using this method, it is possible to work out the systematic structure before working with heavy data.
VDC methods can be applied to almost all parts of construction processes, such as making inspection mission plans for progress tracking and facility monitoring; Look-ahead Schedules (LAS), in which a contractor allocates resources for maximum efficiency; building a code-checking process that takes lots of cost and time; and construction safety management, in which major challenges are concerned with the complex on-site nature of construction projects [27,28,29,30]. Apart from data models and VDC implementation, other VDC research problems concern validation and evaluation of VDC methods as well as ways to maintain and predict desired outcomes [31,32,33]. Apart from VDC application, there are also tools for assessing the maturity level of adopting VDC technologies and philosophies. Ref. [25] proposes a three-level maturity model: Visualization, Integration, and Automation. However, the three levels are not necessarily proceeded in order; instead, companies must always go back and forth among the three levels. Ref. [34] proposes a BIM Maturity Matrix to help improve performance. It introduces the assessment tool with five levels of maturity (1. Initial/Ad hoc, 2. Defined, 3. Managed, 4. Integrated, and 5. Optimized) for each aspect based on a list of maturity assessment tools from different industries. Ref. [32] proposes a VDC scorecard based on other assessment frameworks, industry standards, and Quality-Management tools. The Scorecard has four areas: Planning, Adoption, Technology, and Performance, with 10 Scorecard divisions and 56 Scorecard measurements. The scope of VDC, however, is limited to the virtual side of the Digital Twin and only covers part of the connection between the virtual side and the physical side. Therefore, a supplemental tool designed for the assessment of an organization’s level of maturity towards Digital Twin is required. This tool can refer to VDC-assessment tools as a point of departure.
Apart from VDC, research on Construction Digital Twin has showcased different application scenarios. Some research papers have applied Digital Twin to specific processes in off-site construction, forming a “micro digital twin environment” for certain processes [35,36,37,38]. Ref. [35] uses the concept of Digital Twin for prefabricated façade geometry quality assessment. Ref. [36] proposes new insight into prefabricated construction scheduling, dealing with sequence-dependent due dates and aiming for minimized earliness and tardiness. Ref. [37] proposes a bi-directional information flow model between the virtual model and physical construction. Ref. [38] proposes a BIM method to support detailed geometry design and digital fabrication of modular housings.
Other research focuses on processes on a higher level, such as manufacturing workflow or on-site installation processes [26,36,39,40]. Ref. [41] proposes a Digital Twin framework to predict the module arrival time to help project scheduling. Ref. [39] proposes a model for three-echelon supply chain management in off-site construction with stochastic constraints. Ref. [26] showcases a Digital Twin framework for on-site construction and a case study to validate and evaluate the framework. Ref. [40] proposes an ontology knowledge structure, representing the production workflow for Off-site Manufacturing.
However, other research covers building lifecycle- and construction-related topics such as construction safety. Ref. [42] conducts a literature review on BIM deployment in End-of-Lifecycle to reduce waste produced in deconstruction and demolition. Ref. [43] reviews Digital Twin applications in construction safety and comes up with research challenges to be conducted in the future.
These research papers are inspiring. However, these applications are performed in a fragmented fashion, and a holistic framework is still required for easier application in practice and future research on this topic. Apart from papers working on literature reviews, all the research mentioned above are studies on either data framework or data exchange on off-site construction processes. This finding helps establish an Off-site Construction Digital Twin.

4. Off-Site Construction Processes

This part is conducted using inductive reasoning to draw conclusions by going from the specific to the general. Firstly, by interviewing ACQbuilt representatives, we mapped out the typical project processes at the highest level of their company (Figure 4) because this research goal is to provide a holistic Digital Twin framework for the off-site construction industry. The process comprises eight parts: 1. Bid/Estimate, 2. Sales Contract, 3. Procurement, 4. Drafting, 5. Scheduling, 6. Plant Production, 7. Transportation, and 8. Field Installation. These processes can be categorized into two parts: Office Work and Field Work. Office Work includes Bid/Estimate, Sales Contract, Procurement, and Drafting; Field Work includes Scheduling, Plant Production, Transportation, and Field Installation. Secondly, the authors review how scholars identify the processes in their research. Ref. [44] identifies three dimensions in off-site construction: design, construction, and manufacturing. Ref. [45] states that the main concept of off-site construction lies in off-site manufacturing and on-site construction. Their research is also organized around the two parts. Current research on VDC and Off-site Construction Digital Twin mainly focuses on Field Work. To assure the research outcome can be generalized, this research also limits the scope of Off-site Construction Digital Twin to Field Work, as office work is not a unique part of off-site construction. The three processes in ACQBuilt fieldwork, Scheduling, Plant Production, and Field Installation, are reduced to two processes: Plant Production and Field Installation, because, in follow-up interviews, ACQBuilt representatives clarified that there are scheduling works in both processes. The factory schedule and site schedule are related. The following section builds the Off-site Construction Digital Twin model based on the two processes.

5. Off-Site Construction Digital Twin Model: 4 Parts

As mentioned in previous paragraphs, the Digital Twin consists of three parts: the virtual part, the physical part, and the connection between the two [3,6,7]. This model is developed in the context of the manufacturing industry, in which mass production is enabled by automation and process standardization [46]. There are two types of off-site construction: modulization and panelized construction. All three types require both off-site fabrication and on-site assembly. The literature review shows that, even though the manufacturing industry and off-site construction industry are often compared together by their similarities, the construction industry is much more complicated. Manufacturing and construction are similar because they both engage multiple stakeholders in Concept, Design, Planning, Control, Manufacture, and Assembly [46], and they both “produce engineered products that provide a service to the user” [47]. The differences are apparent: construction is site-specific, one-of-its-kind, and of bigger volume; therefore, it is highly customized, more expensive, and requires non-standard workflow [47,48,49].
Off-site construction, however, has set out to eliminate much of the customized processes, poor efficiency in information flow, and high cost of correcting errors by making projects with standard elements (e.g., walls, floor, etc.), prefabrication in the factory, and system building that showcases dimensional co-ordination [47,50]. Although plant production is conducted mostly in a factory environment, the biggest difference between the manufacturing industry and off-site construction industry remains valid; the off-site construction industry’s practices still assemble the final product in situ [46]. The loading and assembly process still requires special handlings and arrangements that may vary from project to project. Moreover, the uncertainty on site may result in constant changes in factory and site schedules [48]. The literature review in Section 3 also demonstrates research concern on data flow and status with stochastic constraints. Given the complex nature of construction projects [36,37,38,39], Off-site Construction Digital Twin relies heavily on constant information exchange between the virtual side and the physical side. Therefore, research on Off-site Construction Digital Twin cannot be limited to either one of the three entities (the virtual side, the physical side, and the connection between the two); instead, current research always focuses on the relationship among the three [26,35,36,37,38,39,40,41,42]. To create a framework with no ambiguity, it is necessary to re-define the components for Off-site Construction Digital Twin. The new definition is based on the relationship between the components. To enable the “flow” of information, an abstraction of the processes is required. In the virtual world, this is usually represented as a holistic data structure based on process mapping [25]. The information flow between the virtual side and the physical side is a two-way process, with each facing very different research problems and using different technologies and processes [37]. The proposed framework assesses the two processes separately: 1. the information inflow from the physical side to the virtual side, and 2. the information outflow from the virtual side to the physical side. Between inflow information and outflow information, there is always a decision process that happens completely within the virtual world. This part shares common research interests and goals with Virtual Design Construction. Therefore, the authors identify four aspects or components of Off-site Construction Digital Twin: 1. the data structure, 2. data inflow from the physical side to the virtual side, 3. virtual twin modeling and decision making, and 4. data outflow from the virtual side to the physical side. The four processes happen simultaneously, as none of the processes is preliminary to the other (Figure 5). The four parts together form an organic whole dynamics of a Digital Twin.
1. Data structure: Data structure is a systematic abstraction of the current processes, including Scheduling, Plant Production, and Field Installation. There are existing standards for similar purposes, such as IFC. However, IFC does not take into consideration the manufacturing and assembly process for off-site construction. The data structure requires a standard process; otherwise, it falls into studies on qualitative tools, which are helpful in organizing works completed and decided by people. Off-site construction has offered this standard process with plant production and field installation, in which “people” can be counted as resources to be used in the process rather than in which decisions are made. The data structure is preliminary to the establishment of the Digital Twin, as all other parts need to develop based on it.
2. Data Inflow: Data inflow refers to the process by which data reflecting the physical side are collected and transmitted to the virtual side. The data are the materials for decision making. Challenges of this part include the organization of required data from the factory and site, the type of sensors required to collect data, data storage, etc.
3. Virtual Twin Modeling and Decision Making: Virtual twin modeling and decision making use data collected from the physical side. This process collects data from its physical entities and processes the data to make the best decisions. There are challenges concerning Digital Twin modeling, visualization, improvement management, and optimization. This part decides what kind of data are required in part 1, and it bridges part 2 and part 4.
4. Data Outflow: Data outflow refers to the process by which the processed data become a series of orders being called into factories or on-site. This is a reverse process to part 2, usually involving actuators such as robotic arms or automated schedule changes.

6. Off-Site Construction Digital Twin Maturity Level (OCDTML)

6.1. OCDTML Design

The OCDTML is designed based on the Off-site Construction Digital Twin model to guide the off-site construction industry on how to control the process of approaching Off-site Construction Digital Twin. OCDTML helps to make the improvement plan by identifying the current maturity level. Section 4 mentions off-site construction processes and has limited the construction assessment process to Plant Production and Field Installation, and Section 5 identifies four parts of Off-site Construction Digital Twin. These together form the assessment framework. Currently, most organizations are somewhere between conventional construction and Digital Twin construction, usually adopting the concept and technology of Digital Twin in a fragmented fashion on some of the processes. The OCDTML, therefore, is necessary to have maturity indexes to assess the current maturity level and to indicate improvement plans.
Different maturity models present similar maturity indexes. The software Maturity Index has five levels: 1. Initial, 2. Repeatable, 3. Defined, 4. Managed, and 5. Optimized [51]. Ref. [34] also reviews 11 maturity models related to BIM application and summarizes 5 similar ones indexed for the same purpose: 1. Initial/Ad hoc, 2. Defined, 3. Managed, 4. Integrated and 5. Optimized. These are similar descriptions for the level of maturity. For this research purpose, this assessment framework uses the definition from [51]. The following paragraphs specify the indication of each level for Off-site Construction Digital Twin.
Level 1: Initial. This stage indicates the lowest level of Off-site Construction Digital Twin Maturity Level. In this stage, the changes are made ad hoc, usually in work when there are failures or problems. Decisions and operations are mostly made manually.
Level 2: Repeatable. In this stage, there are some standardized processes and automation technologies within project groups. The processes and technologies of new projects are always based on previous projects. It is repeatable for those who perform the work. However, neither of them are standardized between project groups, nor are they formally documented. Some changes are made to streamline the processes, but still include lots of manual inputs and executions that the workflow is not standard to circulate on a larger scale. Some technologies are applied, but the application between projects may vary.
Level 3: Defined. At this level, there are systems across the company that all projects can follow. The processes are mostly standardized and defined for all projects that Digital Twin modeling is enabled for in order to realize automated data collection, transferring, translation, and decision making.
Level 4: Managed. There are quantitative measurements to control the quality of each process. The criteria are clear and quantified. There are troubleshooting processes for monitoring the performance of the Off-site Construction Digital Twin.
Level 5: Optimized. At this level, the organization keeps bridging in new technologies and processes. There are plans to continuously educate employees to keep them updated about the new technologies and processes.
Table 1 shows detailed examples and scenarios in which the levels are assessed for each part. In each of the buckets, there are explanations of the application scenarios and examples. There are two tables: one for Plant Production, and another for Field Installation. However, the criteria are quite similar, because only the data that are collected are different. Therefore, in the table, the authors list the criteria, but the data collected are specified in the next session.

6.2. OCDTML Assessment Process

The typical assessment processes comprises five steps, as shown in Figure 6. 1. Select which process to assess: Plant Production or Field Installation. 2. Identify which of the four parts (Data Structure, Data Inflow, Decision Making, Data Outflow) to assess. 3. The five levels serve as a tool to assess the level of maturity of each of the four parts. 4. Iterate until all the processes and parts are assessed; 5. Finally, calculate the final score.
Recursive interviews were held with ACQbuilt representatives to modify and test the effectiveness of the Off-site Construction Maturity Level.
The four parts indicate different processes when assessing “Plant Production” and “Field Installation”, but the questionnaire format is identical. The difference between Plant Production and Field Installation lies in the data collected when doing the assessment. Take “Data Inflow” as an example; there is one question in the questionnaire: how are the factory data be collected? This question is referencing the machine status in the factory, product status, progress status, etc. The same question is also asked referencing Field Installation: how are the field data collected? This question indicates the material locations, installation progress, crane locations, etc. Both questions can be assessed by the five levels specified in Table 1. The questionnaire comprises 8 sections with 36 questions (Figure 7). The questions are based on the OCDTML reference, as shown in Table 1. The 8 sections are Plant Production–Data Structure, Plant Production–Data Inflow, Plant Production–Decision Making, Plant Production–Data Outflow, Field Installation–Data Structure, Field Installation–Data Inflow, Field Installation–Decision Making, and Field Installation–Data Outflow. The set of questions between “Plant Production” and “Field Installation” are identical, but the process each of them are referring to are different. Therefore, the following section is presented in four parts: 1. Data Structure, 2. Data Inflow, 3. Virtual Twin Modeling, and 4. Data Outflow. Each part comprises two sections: one from Plant Production and the other from Field Installation.
Data Structure Assessment: In this section, the questions focus on the level of standardization and on the progress of creating an abstract data model for all the processes. The design consideration is based on the fact that full standardization is a prerequisite for data abstraction. The questions include: Approximately what percentage of work in the factory/at the field is performed manually/ad hoc? Are there any typical processes of plant production/Field Installation for all the projects? Are there any quality management/check methods? What is the progress of process simulation? Is there any department working on the improvement of the processes? These questions are designed based on the five levels to examine the progress of standardization and improvement intentions.
Data Inflow Assessment: This section examines how the factory/field data are collected for virtual twin modeling and decision making. The questions include: Are there any typical data collection types/processes for all projects? How are factory data (e.g., for the factory: product data, progress data; for the field: progress data, location data, modifications, etc.) collected? (Manually? Machine data? Sensors?) What percentage of it is automated? Are there any projects working on the improvement of data collection?
Virtual Twin Modeling Assessment: This section assesses the decision-making process and the level of automation. Questions in this section include: Is there any flowchart that can be referenced for decision making? What percentage of decisions (e.g., for the factory: resource allocation; for the field: order of installation, crane placement, change in schedule, etc.) are made automatically? If any, are employees familiar with automated decision-making mechanisms? Are there any projects working on the improvement of decision making?
Data Outflow Assessment: This section assesses how the decisions are carried out. This becomes significant when there are changes to be made in the system. The factory is a collaborative environment in which all changes require actions from multiple parties. Questions in this section include: Is there a standard process of decision execution? How are the decisions carried out in the factory? (At the factory, for example, if it has been rescheduled, how is this information passed on to all parties (loading crew, site crew, factory machines, etc.)? In the field, for example, if it is necessary to change part of the design in a house, how is this information passed on to all parties (site crew, truck driver, etc.)?) What percentage of it is automated? If any, are employees familiar with automated execution mechanisms? Answer N/A if not applicable. Are there any projects working on the improvement of decision execution?

7. Results and Discussion

This assessment uses a scoring system that is designed for use at an organizational scale. The five levels are assigned a fixed number of points, with a maximum of 20 points each: level one is assigned 4 points, level two 8 points, level three 12 points, level four 16 points, and level five 20 points. The full score is 160. The maturity score shown in Table 1 is the percentage taken on a scale of 160 to suit the habit of calculating a score on a scale of 100. The result is generated based on answers given by the ACQBuilt representatives and the OCDTML. The questions can be categorized into two parts: 1. questions on the standardization process for conducting things, and 2. questions on the automation of the processes. The following paragraphs elaborate on factory/site facts based on which of the authors assign maturity scores.
In Plant Production–Data Structure, there are standard processes for each project. There is a list of “Master Operations”, in which each project comprises a different number of operations selected from this list. The difference between projects has been reduced to a minimum. All projects are constructed with the same sets of panels. The panels are categorized into three types: wall, floor, and roof. According to the questionnaire, this process is well-managed, but around half of the work is performed manually. They are working on improvement projects on process simulation, which is still in the development phase. Taking all these into consideration, the authors assign level 3 to this section because the process is not fully automated, so it is not fully managed in the sense of a Digital Twin.
In Field Installation–Data Structure, the progress is quite like Plant Production, but with a higher percentage of manual work. Almost all works on site are performed manually. Field Installation faces more challenges in automation, as site work has, by nature, heavier requirements for manual work. Process simulation faces challenges of digitalization on a lower level because each project is different in design. This particularity is a research problem that requires technological advancement. Based on these facts, the authors consider this session to be at maturity level 2. There are still processes to be defined on a lower level to enable full automation.
In the section Data Inflow for both Plant Production and Field Installation, data are collected manually/ad hoc. Therefore, both sessions are clearly at maturity level 1. For this part to improve, improvements in Decision Making are considered a prerequisite, as automated decision making can ask for consistent data inflow, thus making automated data inflow possible.
In the section Decision Making part for both Plant Production and Field Installation, there are some standard processes to make decisions (for example, the order to process jobs, schedule, crane allocation, etc.), but all daily works are performed manually. Take scheduling, for example. Both scheduling in the factory and in the field are performed ad hoc, and there is some logic behind the operation (e.g., the project due date, availability of crane and machine, etc.). However, it is far from optimized. Most of the waste (time, machine idle time, labor waiting time, etc.) is generated at this stage. Currently, the lower-level decision-making process is based on the resource condition (crane location, trailer status, machine availability, etc.), so there are not enough foreseeing decisions that can be made. This section for both Plant Production and Field Installation is assigned to Level 2.
For Data Outflow, Plant Production is clearly at a higher maturity level than Field Installation. In Plant Production, there is a clear process for execution once decisions have been made. Around half of the works are automated. The decisions are passed around via dashboards and emails. The maturity level assigned to this section is level 3. Data Outflow in Field Installation, however, is completely manual work, with processes on a higher level but not with daily works at the lower level. Therefore, this section is still considered to be at maturity level 1.
According to their answers, ACQBuilt has established standard processes on a higher level for both Plant Production and Field Installation. However, there are still massive manual/ad hoc operations in daily work, which is true for all the four parts. For the four parts, the answers to the questionnaire have shown a stronger profile in Data Structure than in the other three parts. Referring to Section 5, where the four parts are defined, Data Structure does go first in the making of the Off-site Construction Digital Twin. Only after the data inputs and outputs are made clear can researchers work on automated decision making, technology enablers to build Data Inflow, and Data Outflow for Off-site Construction Digital Twin simultaneously.
Reviewing factory/site processes and assessment results (Table 2), it is possible to identify the next step of development. The biggest problem at the current stage seems to be Decision Making, because this part is the source of most waste (time, labor, resource, etc.), as well as Data Inflow/Outflow guidelines. Take scheduling, for example. Manual scheduling is always ad hoc. If there are changes that must be made, the person making the changes cannot always guarantee the best way of allocating resources and coordinating multiple projects. Decision Making is also a pre-requisite to automated data outflow and inflow because the consistency of input and output data is necessary for automated data flows.
This assessment framework helps practitioners identify the path towards a complete Off-site Construction Digital Twin by 1. generating a report of progress that has been made and by 2. pointing out the gaps and suggesting what the next steps are to take action. The entire assessment process is based on the wood-panelized construction process, in which the focus in the factory is framing, window/door placement, trimming and loading, etc., and the focus at the site is unloading, panel transportation, installation, etc. Therefore, the assessment criteria can be generalized to other off-site construction methods given other construction processes, such as modular construction.

8. Conclusions

This research was designed with two goals: 1. to pave the way for future research in Off-site Construction Digital Twin by creating a model, and 2. to enable smoother new technology applications in practice by introducing a holistic assessment framework which synthesizes the progress and points out the next step of development. The assessment framework is a knowledge tool that incorporates both knowledge in practice and academia (Figure 8). This assessment framework was developed on top of the Off-site Construction Digital Twin Model, which serves as a starting knowledge framework for future research. This model was developed based on a literature review on Digital Twin and recursive interviews with industry practitioners on the construction process. This framework covers all aspects of wood-panelized construction, and progress is reflected in the assessment framework scoring system. Apart from that, assessment results help organizations identify gaps and point out research directions. This framework is validated by a case study with ACQBuilt, who specializes in wood-panelized construction. The framework synthesizes the technologies that ACQBuilt has put into use to form a foundation for a unified workflow and technology application platform, aiming for a holistic and standardized process, which is necessary for a complete Off-site Construction Digital Twin. The assessment result patterns for the case study show that ACQBuilt has a well-defined data structure for both Plant Production and Field Installation, but data collection and execution are still ad hoc. However, Decision Making, or Virtual Twin modeling, should be the next step of development because decision making defines the data required and data output in Data Inflow and Data Outflow.
Although this research presents a plausible Off-site Construction Digital Twin, there are limitations. This research is limited by the number of off-site construction companies we can reach out to. Organization assessment can be more informative if more organizations use this assessment framework. More organizations can make the scoring system in percentages, so each organization can be informed by how they are performing in relation with the proven practices of the rest of the industry. Another limitation is related to the particular type of off-site construction we work with. As we work with ACQbuilt, whose main specialty is in off-site panelized-wood construction, process mapping is based on this type of construction. There are other types of off-site construction, including module construction, precast/prefabricated construction, etc., that are not considered in this research. However, among all the off-site construction types, wood-panelized construction is the most common type in North America. Additionally, off-site construction companies share similar procedures, and the data model can be extended to wider application by further studies.

Author Contributions

Conceptualization, Y.W. and Z.L.; methodology, Y.W. and Z.L.; validation, Y.W., Z.L. and S.A.; data curation, Y.W. and S.A.; writing—original draft preparation, Y.W.; writing—review and editing, Z.L.; visualization, Y.W.; supervision, Z.L.; funding acquisition, Z.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Discovery Grant (Grant #: RGPIN-2020-04126) of Natural Sciences and Engineering Research Council (NSERC) of Canada.

Acknowledgments

ACQBuilt is acknowledged for its academic support of this research.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Gerth, R.; Boqvist, A.; Bjelkemyr, M.; Lindberg, B. Design for construction: Utilizing production experiences in development. Constr. Manag. Econ. 2013, 31, 135–150. [Google Scholar] [CrossRef]
  2. Changali, S.; Mohammad, A.; van Nieuwland, M. The Construction Productivity Imperative; McKinsey & Company: New York, NY, USA, 2015. [Google Scholar]
  3. Kritzinger, W.; Karner, M.; Traar, G.; Henjes, J.; Sihn, W. Digital Twin in manufacturing: A categorical literature review and classification. IFAC-PapersOnLine 2018, 51, 1016–1022. [Google Scholar] [CrossRef]
  4. Boje, C.; Guerriero, A.; Kubicki, S.; Rezgui, Y. Towards a semantic Construction Digital Twin: Directions for future research. Autom. Constr. 2020, 114, 103179. [Google Scholar] [CrossRef]
  5. Strother, J.B.; Ulijn, J.M.; Fazal, Z. Drowning in Data: A Review of Information Overload within Organizations and the Viability of Strategic Communication Principles. In Information Overload; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2012; pp. 231–250. [Google Scholar] [CrossRef]
  6. Grieves, M. Digital Twin: Manufacturing Excellence through Virtual Factory Replication. 2015. Available online: https://www.researchgate.net/publication/275211047_Digital_Twin_Manufacturing_Excellence_through_Virtual_Factory_Replication (accessed on 10 September 2021).
  7. Tao, F.; Zhang, H.; Liu, A.; Nee, A.Y.C. Digital Twin in Industry: State-of-the-Art. IEEE Trans. Ind. Inform. 2019, 15, 2405–2415. [Google Scholar] [CrossRef]
  8. Lopez, D.; Froese, T.M. Analysis of Costs and Benefits of Panelized and Modular Prefabricated Homes. Procedia Eng. 2016, 145, 1291–1297. [Google Scholar] [CrossRef]
  9. Altaf, M.S.; Al-Hussein, M.; Yu, H. Wood-Frame Wall Panel Sequencing Based on Discrete-Event Simulation and Particle Swarm Optimization. In Proceedings of the Wood-Frame Wall Panel Sequencing Based on Discrete-Event Simulation and Particle Swarm Optimization, Sydney, Australia, 9–11 July 2014. [Google Scholar] [CrossRef] [Green Version]
  10. Moreno, A.; Velez, G.; Ardanza, A.; Barandiaran, I.; de Infante, Á.R.; Chopitea, R. Virtualisation process of a sheet metal punching machine within the Industry 4.0 vision. Int. J. Interact. Des. Manuf. 2017, 11, 365–373. [Google Scholar] [CrossRef]
  11. Glaessgen, E.; Stargel, D. The Digital Twin Paradigm for Future NASA and U.S. Air Force Vehicles. In Proceedings of the 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference, Honolulu, HI, USA, 23–26 April 2012. [Google Scholar] [CrossRef] [Green Version]
  12. Lee, J.; Bagheri, B.; Kao, H.-A. A Cyber-Physical Systems architecture for Industry 4.0-based manufacturing systems. Manuf. Lett. 2015, 3, 18–23. [Google Scholar] [CrossRef]
  13. Schroeder, G.; Steinmetz, C.; Pereira, C.E.; Muller, I.; Garcia, N.; Espindola, D.; Rodrigues, R. Visualising the digital twin using web services and augmented reality. In Proceedings of the 2016 IEEE 14th International Conference on Industrial Informatics (INDIN), Poitiers, France, 19–21 July 2016; pp. 522–527. [Google Scholar] [CrossRef]
  14. Yao, X.; Zhou, J.; Lin, Y.; Li, Y.; Yu, H.; Liu, Y. Smart manufacturing based on cyber-physical systems and beyond. J. Intell. Manuf. 2019, 30, 2805–2817. [Google Scholar] [CrossRef] [Green Version]
  15. Konstantinov, S.; Ahmad, M.; Ananthanarayan, K.; Harrison, R. The Cyber-physical E-machine Manufacturing System: Virtual Engineering for Complete Lifecycle Support. Procedia CIRP 2017, 63, 119–124. [Google Scholar] [CrossRef]
  16. Uhlemann, T.H.-J.; Lehmann, C.; Steinhilper, R. The Digital Twin: Realizing the Cyber-Physical Production System for Industry 4.0. Procedia CIRP 2017, 61, 335–340. [Google Scholar] [CrossRef]
  17. Rasheed, A.; San, O.; Kvamsdal, T. Digital Twin: Values, Challenges and Enablers From a Modeling Perspective. IEEE Access 2020, 8, 21980–22012. [Google Scholar] [CrossRef]
  18. Zheng, Y.; Yang, S.; Cheng, H. An application framework of digital twin and its case study. J. Ambient Intell. Human Comput. 2019, 10, 1141–1153. [Google Scholar] [CrossRef]
  19. Ameri, F.; Sabbagh, R. Digital Factories for Capability Modeling and Visualization. In Proceedings of the IFIP International Conference on Advances in Production Management Systems (APMS), Iguassu Falls, Brazil, 3–7 September 2016; pp. 69–78. [Google Scholar] [CrossRef] [Green Version]
  20. Laakso, M.; Kiviniemi, A. The ifc standard—A review of history, development, and standardization. J. Inf. Technol. Constr. (ITcon) 2012, 17, 134–161. [Google Scholar]
  21. Lee, Y.-C.; Eastman, C.M.; Lee, J.-K. Validations for ensuring the interoperability of data exchange of a building information model | Elsevier Enhanced Reader. Autom. Constr. 2015, 58, 176–195. [Google Scholar] [CrossRef]
  22. Wilson, J.S. Sensor Technology Handbook; Newnes: Amsterdam, The Netherlands, 2005; Available online: https://login.proxy.hil.unb.ca/login?url=https://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=117182&site=ehost-live&scope=site (accessed on 14 February 2022).
  23. Lu, W.; Tan, T.; Xu, J.; Wang, J.; Chen, K.; Gao, S.; Xue, F. Design for manufacture and assembly (DfMA) in construction: The old and the new. Archit. Eng. Des. Manag. 2021, 17, 77–91. [Google Scholar] [CrossRef]
  24. Pandey, P.S.; Raghuwanshi, S.K.; Kumar, S. Recent Advances in Two-Dimensional Materials-Based Kretschmann Configuration for SPR Sensors: A Review. IEEE Sens. J. 2022, 22, 1069–1080. [Google Scholar] [CrossRef]
  25. Kunz, J.; Fischer, M. Virtual Design and Construction: Themes, Case Studies and Implementation Suggestions; Center for Integreated Facility Engineering: Stanford, CA, USA, 2009. [Google Scholar]
  26. Pan, Y.; Zhang, L. A BIM-data mining integrated digital twin framework for advanced project management. Autom. Constr. 2021, 124, 103564. [Google Scholar] [CrossRef]
  27. Eastman, C.; Lee, J.; Jeong, Y.; Lee, J. Automatic rule-based checking of building designs. Autom. Constr. 2009, 18, 1011–1033.Dec. [Google Scholar] [CrossRef]
  28. Hamledari, H.; Sajedi, S.; McCabe, B.; Fischer, M. Automation of Inspection Mission Planning Using 4D BIMs and in Support of Unmanned Aerial Vehicle–Based Data Collection. J. Constr. Eng. Manag. 2021, 147, 04020179. [Google Scholar] [CrossRef]
  29. Shafiq, M.; Afzal, M.; Aljassmi, H. Improving construction safety with Virtual-Design Construction Technologies—A review. J. Inf. Technol. Constr. 2021, 26, 319–340. [Google Scholar] [CrossRef]
  30. Dong, N. Automated Look-Ahead Schedule Generation and Optimization for the Finishing Phase of Complex Construction Projects. Ph.D. Thesis, Stanford University, Stanford, CA, USA, 2012. Available online: https://www.proquest.com/docview/2458565230/abstract/2AC75ACDE9894C57PQ/1 (accessed on 18 January 2022).
  31. Yee, P.; Fischer, M.; Kam, C. Prospective validation of virtual design and construction methods. J. Inf. Technol. Constr. 2013, 18, 214–239. [Google Scholar]
  32. Kam, C.; Song, M.H.; Senaratna, D. VDC Scorecard: Formulation, Application, and Validation. J. Constr. Eng. Manag. 2017, 143, 04016100. [Google Scholar] [CrossRef]
  33. Rischmoller, L.; Fischer, M.; Kunz, J. A study of virtual design and construction implementation and benefits using a bayesian approach. Rev. De La Constr. 2012, 11, 74–87. [Google Scholar] [CrossRef] [Green Version]
  34. Succar, B. Building Information Modelling Maturity Matrix. In Handbook of Research on Building Information Modeling and Construction Informatics: Concepts and Technologies; IGI Global: Hershey, NY, USA, 2009; pp. 65–103. [Google Scholar] [CrossRef]
  35. Tran, H.; Nguyen, T.N.; Christopher, P.; Bui, D.-K.; Khoshelham, K.; Ngo, T.D. A digital twin approach for geometric quality assessment of as-built prefabricated façades. J. Build. Eng. 2021, 41, 102377. [Google Scholar] [CrossRef]
  36. Yazdani, M.; Kabirifar, K.; Fathollahi-Fard, A.; Mojtahedi, M. Production scheduling of off-site prefabricated construction components considering sequence dependent due dates. Environ. Sci. Pollut. Res. 2021. [Google Scholar] [CrossRef]
  37. Akanmu, A.; Anumba, C.J. Cyber-physical systems integration of building information models and the physical construction. Eng. Constr. Archit. Manag. 2015, 22, 516–535. [Google Scholar] [CrossRef]
  38. He, R.; Li, M.; Gan, V.J.L.; Ma, J. BIM-enabled computerized design and digital fabrication of industrialized buildings: A case study. J. Clean. Prod. 2021, 278, 123505. [Google Scholar] [CrossRef]
  39. Salari, S.A.-S.; Mahmoudi, H.; Aghsami, A.; Jolai, F.; Jolai, S.; Yazdani, M. Off-Site Construction Three-Echelon Supply Chain Management with Stochastic Constraints: A Modelling Approach. Buildings 2022, 12, 119. [Google Scholar] [CrossRef]
  40. Ayinla, K.; Vakaj, E.; Cheung, F.; Tawil, A.-R.H. A Semantic Offsite Construction Digital Twin- Offsite Manufacturing Production Workflow (OPW) Ontology. In Proceedings of the Second International Workshop on Semantic Digital Twins, Hersonissos, Greece, 6 June 2021; Available online: https://openreview.net/forum?id=KGwOglqpSp7 (accessed on 30 March 2022).
  41. Lee, D.; Lee, S. Digital Twin for Supply Chain Coordination in Modular Construction. Appl. Sci. 2021, 11, 5909. [Google Scholar] [CrossRef]
  42. Akbarieh, A.; Jayasinghe, L.B.; Waldmann, D.; Teferle, F.N. BIM-Based End-of-Lifecycle Decision Making and Digital Deconstruction: Literature Review. Sustainability 2020, 12, 2670. [Google Scholar] [CrossRef] [Green Version]
  43. Hou, L.; Wu, S.; Zhang, G.; Tan, Y.; Wang, X. Literature Review of Digital Twins Applications in Construction Workforce Safety. Appl. Sci. 2021, 11, 339. [Google Scholar] [CrossRef]
  44. Goulding, J.S.; Rahimian, F.P.; Arif, M.; Sharp, D.M. New offsite production and business models in construction: Priorities for the future research agenda. Archit. Eng. Des. Manag. 2015, 11, 163–184. [Google Scholar] [CrossRef]
  45. Sutrisna, M.; Ramanayaka, C.D.D.; Goulding, J.S. Developing work breakdown structure matrix for managing offsite construction projects. Archit. Eng. Des. Manag. 2018, 14, 381–397. [Google Scholar] [CrossRef]
  46. Winch, G. Models of manufacturing and the construction process: The genesis of re-engineering construction. Build. Res. Inf. 2003, 31, 107–118. [Google Scholar] [CrossRef]
  47. Sanvido, V.E.; Medeiros, D.J. Applying Computer-Integrated Manufacturing Concepts to Construction. J. Constr. Eng. Manag. 1990, 116, 365–379. [Google Scholar] [CrossRef]
  48. Koskela, L. Application of the New Production Philosophy to Construction; Stanford University: Stanford, CA, USA, 1992. [Google Scholar]
  49. Tan, T.; Lu, W.; Tan, G.; Xue, F.; Chen, K.; Xu, J.; Wang, J.; Gao, S. Construction-Oriented Design for Manufacture and Assembly Guidelines. J. Constr. Eng. Manag. 2020, 146, 04020085. [Google Scholar] [CrossRef]
  50. Crowley, A. Construction as a manufacturing process: Lessons from the automotive industry. Comput. Struct. 1998, 67, 389–400. [Google Scholar] [CrossRef]
  51. Paulk, M.C.; Curtis, B.; Chrissis, M.B.; Weber, C.V. Capability maturity model, version 1.1. IEEE Softw. 1993, 10, 18–27. [Google Scholar] [CrossRef]
Buildings 12 00566 g001 550
Figure 1. Research design and methods.
Figure 1. Research design and methods.
Buildings 12 00566 g001
Buildings 12 00566 g002 550
Figure 2. Wood-panelized off-site construction: factory and construction site. (Photo is courtesy of ACQBuilt, Edmonton, AB, Canada).
Figure 2. Wood-panelized off-site construction: factory and construction site. (Photo is courtesy of ACQBuilt, Edmonton, AB, Canada).
Buildings 12 00566 g002
Buildings 12 00566 g003 550
Figure 3. Three components of a Digital Twin.
Figure 3. Three components of a Digital Twin.
Buildings 12 00566 g003
Buildings 12 00566 g004 550
Figure 4. Typical company processes.
Figure 4. Typical company processes.
Buildings 12 00566 g004
Buildings 12 00566 g005 550
Figure 5. Four parts of Off-site Construction Digital Twin.
Figure 5. Four parts of Off-site Construction Digital Twin.
Buildings 12 00566 g005
Buildings 12 00566 g006 550
Figure 6. Assessment process.
Figure 6. Assessment process.
Buildings 12 00566 g006
Buildings 12 00566 g007 550
Figure 7. Assessment framework based on the off-site construction process in ACQBuilt.
Figure 7. Assessment framework based on the off-site construction process in ACQBuilt.
Buildings 12 00566 g007
Buildings 12 00566 g008 550
Figure 8. OCDTML components.
Figure 8. OCDTML components.
Buildings 12 00566 g008
Table
Table 1. Off-site Construction Digital Twin Maturity Level.
Table 1. Off-site Construction Digital Twin Maturity Level.
DT: 4 PartsDTML-1: Data Structure
DTML-1 refers to the idea that the company defines standard processes in its off-site construction supply chain, factory, and site. This standard process is abstracted into a data structure that sets the foundation for Off-site Construction Digital Twin.		DTML-2: Data Inflow
DTML-2 refers to the idea that the company establishes automated data inflow from the supply chain, factory, and construction site to the virtual side.		DTML-3: Decision Making/Virtual Twin Modeling
DTML-3 refers to the idea that the company implements a virtual decision-making program to use data inflow from the physical side to make decisions/optimizations.		DTML-4: Data Outflow
DTML-4 refers to the idea that the company establishes a data outflow stream to execute virtual decisions in the physical world.
Level
Level 1: InitialThe process is unorganized, and the processes may vary each time, depending on the person in charge.Technologies are not adequate. Data collection only happens when required. Hardware updates and installation are operated when unavoidable. Manually detect where to optimize/modify based on observations in factories and on site.Manual execution upon receiving orders from the virtual side.
Level 2: RepeatableThe data structure is partially defined for certain operations that make it repeatable for those who do the work but are not systematically organized.Mostly manual inputs of the data from the physical side to the database, with some automated sensors collecting data on-site and in the factory. The automated process has multiple application scenarios.Automated decision making is unified within a project team. The planning and management of projects are always based on previous similar projects. Some project teams adopt actuators (sensors, robotic arms) in certain processes on site to execute orders from the virtual side, and the work schedules can be generated automatically.
Level 3: DefinedMost of the processes are standardized, and most of the product data, process data, and resource data are categorized (data structure well-defined for a project cycle, e.g., ER diagram fully established).The process of automated data collection is well-defined and deployed in certain processes for all projects, e.g., panel cutting and punching processes are reflected in computer systems in real time.All decision makings necessary for the physical side are completed automatically on the virtual side, and the typical process is well-defined across the organization for all projects.All the execution work is based on the Virtual Twin decision making and has typical processes defined for all projects.
Level 4: ManagedAll processes are well-synchronized across projects and tightly integrated with business processes. Interoperable data usage, storage, and exchange are treated as part of organizational strategy.The technologies and data collection process have clear standards across the organization. There are quantitative goals to measure the quality of data inflow. There are quantitative measurements for the level of quality achieved in the process. There are measurements for the quality of the decisions.The quality of this automated execution is well-controlled with quantitative measurement tools.
Level 5: OptimizedBased on Level 5, there are projects to continuously optimize data usage, storage, and exchange. All the required data are collected automatically via a system that can bridge in new processes and new technologies.Continuous software improvement including user interface, code efficiency, visualization, etc.The data outflow process is continuously improving with new technologies, better adoption, and more efficient processes.
Table
Table 2. The assessment results of ACQBuilt.
Table 2. The assessment results of ACQBuilt.
Digital Twin Maturity Level Assessment Result: ACQBuilt
Level 1: 4 ptsLevel 2: 8 ptsLevel 3: 12 ptsLevel 4: 16 ptsLevel 5: 20 pts
Plant ProductionData Structure ●
Data Inflow●
Decision Making ●
Data Outflow ●
Field InstallationData Structure ●
Data Inflow●
Decision Making ●
Data Outflow●
Subtotal12242400
Total Points60
Maturity Score37.5
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Wei, Y.; Lei, Z.; Altaf, S. An Off-Site Construction Digital Twin Assessment Framework Using Wood Panelized Construction as a Case Study. Buildings 2022, 12, 566. https://doi.org/10.3390/buildings12050566

AMA Style

Wei Y, Lei Z, Altaf S. An Off-Site Construction Digital Twin Assessment Framework Using Wood Panelized Construction as a Case Study. Buildings. 2022; 12(5):566. https://doi.org/10.3390/buildings12050566

Chicago/Turabian Style

Wei, Yuxi, Zhen Lei, and Sadiq Altaf. 2022. "An Off-Site Construction Digital Twin Assessment Framework Using Wood Panelized Construction as a Case Study" Buildings 12, no. 5: 566. https://doi.org/10.3390/buildings12050566

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop