Digital Twin for Construction

Research Overview

Observed Problem:

In simple words, a digital twin is a digital replica of living or non-living physical entity. By bridging the physical and the virtual world, data is transmitted seamlessly allowing the virtual entity to exist simultaneously with the physical entity. However, digitalization and automation to fully realize the benefits of a ‘digital twin’ requires sustained top-down and bottom-up leadership and the subsequent language to discuss and prioritize opportunities. A consistent taxonomy to discuss automation and digitalization is missing in the construction industry, often leading to haphazard technological discussions and decision-making. This results in a drifting of company objectives and leadership lockstep.

Primary Research Objective:

Develop a taxonomy of digital twin for the construction industry, outlining the current and envisioned levels of the “digital twin’ concept. Through small case studies and interviews, possible digital twin levels will be determined, which will define a possibilities frontier curve to enable companies to strategically reason about and critically question their technological and digital decision making and strategy setting.

Potential Value to CIFE Members and Practice:

• Enable firms to strategically reason about and critically question their technological and digital decision making and strategy setting.
• Support construction companies to answer the following questions:
  • Why do we want to use this specific technology?
  • What are the impacts of it and Where do we see ourselves after implementing this?
  • How are we going to develop these capabilities?

Research provides relevant insights for: Construction, Design, Operations

Research and Theoretical Contributions

• Articulate and visualize the strategic and process issues facing company deployment of digitalization and automation strategies.
• Framework to articulate and benchmark the state of the digital twin concept and create a common taxonomy for researchers and practitioners

Industry and Academic Partners

TBDResearch Updates (Dec 2019)

Research Updates (Dec 2019)

A thorough literature review existing in the manufacturing and industrial sector has been done about the concept of digital twin. A brief summary of the same has been presented below:

The word ‘digital twin’ was initially coined by NASA [2] when they built an exact replica of the rockets for astronomers to rehearse. They called this as Information mirroring model (Figure 1) where the virtual space was an exact replica of the real space which enabled experimentation and simulation via physics. However, currently, the concept of “digital twin” means various “things”. For example, in automobile engineering it essentially means creating a digital replica of the product and simulate its various scenarios to optimize and gain efficiency [3]. Hence digital twin is just a real time digitized copy of a physical object with minimal involvement of humans. On the contrary, in healthcare digital twin creation of the human is being attempted for various drug trials which is totally opposite to that of manufacturing because here essentially it becomes all human focused involving emotional, intellectual and physical capabilities.

Digital twin, extremely relevant to the Architecture, Engineering & Construction (AEC) industry, is still relatively a very new and ‘ill’ defined term due to the lack of appropriate research literature. The manufacturing industry can provide us some parallels for defining the digital twin for construction and hence a comprehensive review of the literature was done. We reviewed the top definitions available in the literature [1,2,3,4,5,6] and it is interesting to note that all the definition are inspired from a physical tangible object, thus reflecting that most of the existing literature focuses on the design of a ‘product’.

We are quite aware of the fact that construction isn’t just focused on a product or people. It is in fact a combination of Product, Process and an Organization. Therefore, it becomes quite interesting and challenging to see how these entities can be represented digitally, both individually and together along with their interactions, which serves as the steppingstone for our further research.

[1]  David Kadleček, Digital Twin, IBM Cognitive IoT.
[2]  M. Grieves, "Digital Twin: Manufacturing Excellence through Virtual Factory Replication," 2014.
[3]  "Digital Twin," [Online]. Available: https://en.wikipedia.org/wiki/Digital_twin.
[4]  L. W. Aaron Parrott, "Industry 4.0 and the digital twin," 12 May 2017. [Online]. Available: https://www2.deloitte.com/insights/us/en/focus/industry-4-0/digital-twi….
[5]  Roland Rosen, Georg von Wichert, George Lo, Kurt D. Bettenhausen, "About The Importance of Autonomy and Digital Twins for the Future of Manufacturing," in International Federation of Automatic Control, 2015.
[6]  Michael Chui, James Manyika, and Mehdi Miremadi, What AI can and can’t do (yet) for your business, McKinsey & Company, 2018.

Research Updates (Aug 2020)

We summarized the existing literature around Digital twin in AEC and sister industries like manufacturing, aerospace etc. Our main takeaway from the literature review are (1) Digital twin is a very fuzzy term owing to the wide spectrum of technologies and market needs, and (2) The current Digital twin literature is very product focused, ignoring the other elements of POP. The findings have been summarized in section 1. Therefore, we questioned ourselves, what exactly is ‘digital’? and what are we ‘twinning’? The current ongoing progress on these research questions have been summarized in section 2.

1. Related work and Points of Departure from Digital Twin literature

A digital twin is a virtual representation of the world which enables us to use technology and at the same time establishes a bidirectional link with the physical world. Digital twins gained initial momentum from high-value product manufacturing industries such as automotive and aerospace, but the concept is increasingly being adapted by the building and construction industry. It is widely accepted in the literature that the term ‘Digital Twin’ was coined by Michael Grieves in 2002 in the context of an industry presentation concerning product lifecycle management (PLM) (Grieves 2014). The concept was later adopted by NASA in Modelling, Simulation, Information Technology & Processing roadmap (2010) and the digital twin was defined as- “an integrated multi-physics, multi-scale, probabilistic simulation of a vehicle or system that uses the best available physical models, sensor updates, fleet history, etc., to mirror the life of its flying twin”.

Digital twin, extremely relevant to the Architecture, Engineering & Construction (AEC) industry, is still relatively a very new and ‘ill’ defined term due to the lack of appropriate research literature. The manufacturing industry can provide us some parallels for defining the digital twin for construction and hence a comprehensive review of the literature was done (Elisa Negria 2017). Figure 1 shows the top 100 words of the word cloud constructed by compiling the different definitions available in the literature. It is interesting to note that among the top words are ‘physical’ and ‘product’, all representing that the definition is inspired from a physical tangible object, thus reflecting that most of the existing literature focuses on the ‘product’.  Circling back, to the POP framework: it isn’t sufficient to just focus on the product, process and organization are equally important aspects. Hence, the product focus is one of the major short comings of the current digital twin literature.

2.  What is digital and What are we twinning?

Digitalization is the use of digital technologies to change a business model and provide value-producing opportunities resulting in more revenue (Ross n.d.). The term to emphasize here is ‘value’. Every digital optimization that we do is essentially to improve value addition. Hence, we looked at digital twin from the lens of value addition.

‘Value’ is a commonly used measure in lean manufacturing. In construction, the concept of value was introduced by (Koskela 1992). Essentially, value adding activities are the “Activity that converts material and/or information towards fulfilling the requirement of a customer, and the customer is willing to pay for it”. It brings in the customer centric view, which is a very important factor in the construction industry. This definition doesn’t necessarily imply, that all the activities which aren’t value adding must be removed. For e.g. regular project meetings of the participants is one of the key elements for a successful project, but essentially is a non-value adding activity. This further motivates for better classification activities as: Value adding, Non-Value adding but essential, Nonvalue adding.
In summary, the ways of adding value are by:

1. Increasing value adding activities itself i.e. enhancing the product delivered. This essentially is controlled by the product or the service being provided itself.
2. Streamline the non-value adding but essential activities: The process and the organization are the things which support in making the product happen but doesn’t have an inherent value in itself and hence are non-value adding but essential activities. So, this can be controlled by the process and the organization arms of the POP.

We observe that value addition for AEC is essentially controlled by POP, therefore being the appropriate thing to twin. Hence, we argue that twinning the POP is important, rather than just having a product focus.

But still digitalization in this form is very fuzzy. Is documenting in excel or having a full-fledged predictive 3D BIM digitalization? This led to the development of levels of digitalization inspired by (Harris 2007), representing the different levels at which digitalization can take place, each having a different value proposition.

Figure 2 shows the different levels of digitalization in a Maslow’s hierarchy format suitable for the construction industry. Description about each level is as follows:

• Digitization: The most basic level, essentially converting analog data to digital format. This acts as the base for all the other upper levels. It comes with limited benefits like better document management, enhanced security features etc. which have become an operation necessity now a days. All the technologies like digital document managements etc. lie in this category.
• Visualization: This level essentially helps us to visualize what is happening in the world in a better way. This may include wireless monitoring, Augmented reality/Virtual Reality, BIM, Drone technology etc.
• Analysis: Here we start to get insights from the data and are answering the questions around why certain things or situations are occurring. Data analysis become a common part at this stage. The progression towards a digitally mature organization starts happening at this stage. All the basic data analytics, parametric methods etc. lie in this zone.
• Prediction: Here we start answering what will happen in the future and make decisions according to that. All the generative methods, design simulations, artificial intelligence algorithms come at this level.
• Prescriptive: Finally, here the machine starts recommending us what is the ideal thing to for a given situation therefore reaching the maximum degree of sophistication possible.

Ongoing work and Next Steps

The current framework provides a good broader picture of the digitalization strategy, but still lacks fine grained actionable insights. Hence, we are working towards detailing important elements of the POP and digitalization pyramid. Framework validation and detailed case studies would follow after this.

Bibliography

• Elisa Negria, Luca Fumagallia, Marco Macchia. 2017. "A review of the roles of Digital Twin in CPS-based production systems." 27th International Conference on Flexible Automation and Intelligent Manufacturing. Elsevier.
• Gerald C. Kane, Doug Palmer, Anh Nguyen Phillips, David Kiron and Natasha Buckley. 2015. Strategy, not Technology, Drives Digital Transformation. MIT Sloan Management Review.
• Grieves, Michael. 2014. ""Digital Twin: Manufacturing Excellence through Virtual Factory Replication."
• Harris, Thomas H. Davenport and Jeanne G. 2007. Competing on Analytics: The New Science of Winning Heavily. Harvard Business Review.
• Jay Lee, Behrad Bagheri, Hung-An Kao. 2014. "A Cyber-Physical Systems architecture for Industry 4.0-based manufacturing systems." Society of Manufacturing Engineers (SME).
• Koskela, Lauri. 1992. Application of new production philosophy to construction. Centre for Integrated facilities engineering (CIFE).
• Martin Fischer, Howard W. Ashcraft, Dean Reed, Atul Khanzode. 2017. Integrating Project Delivery. Wiley.
• Porter, Michael E. 2008. "The five competitive forces that shape strategy." Harvard Business Review.
• Ross, Jeanne. n.d. "Don’t Confuse Digital with Digitization."

Original Research Proposal

Proposal 2019-15

Funding Year: 

2020