Helping Traditional Industries Transform — Digital Twin

Alibaba Design
6 min readSep 10, 2020

Some of you may have heard of the term “digital twin” before you clicked on this article. What exactly is the “digital twin”, how does it apply to design, and how can it better serve the business? That’s what I’m going to talk about today.

I. The origins of the digital twin

The concept and model of the digital twin were publicly introduced in 2002 by Dr. Michael Grieves from the University of Michigan. Grieves proposed the digital twin as the conceptual model underlying product lifecycle management. Back in March 2011, it was the U.S. Air Force Research Laboratory (AFRL) that explicitly proposed the concept of the digital twin. Later on, industrial and Internet businesses began to adopt the concept.

It is explained in Wikipedia that a digital twin refers to a digital replica of potential and actual physical assets (physical twin), processes, people, places, systems, and devices that can be used for various purposes. The digital representation provides both the elements and the dynamics of how an Internet of things (IoT) device operates and lives throughout its life cycle.

Digital Twin is a concept that transcends reality and can be viewed as a digital mapping system of one or more significant, interdependent equipment systems.

While digital transformation trends in industrial works have been increasingly evident, Alibaba Cloud Industrial Brain has achieved the remote and online monitoring of machinery and equipment, the global data visualization, and has been using data and intelligence to help many enterprises on operations and strategies to create real value.

So, as designers, how do we use design to maximize the value of the digital twin?

II. Overview of the industrial digital twin panorama architecture

Using IoT technology, our product has the ability to predict equipment failures and do safety monitoring. Leveraging big data and machine learning technology, we could provide sales forecasts and industrial intelligent production, thus helping businesses to improve production capacity.

The digital twin technology, which uses data to map physical entities, has a disruptive impact on many industries. Alibaba Cloud Industrial Brain aims to “digitize, intellectualize & standardize”, and realizes interactive mapping and symbiosis between physical and digital space through 3D visualization modeling, virtual simulation, and other advanced technologies.

By building intelligent factory digital twin platforms covering the whole life cycle of the factory, as well as utilizing innovative management ideas and control systems, Alibaba Cloud Industrial Brain is to accelerate the digital transformation and upgrading of industrial enterprises.

Ⅲ.Overview of the Industrial digital twin capabilities

As stated above, the business and design value of our production capability is demonstrated in four aspects: sales and operation planning, multi-end alerting, intelligent optimization, and intelligent production line construction. To realize the above values, we have accumulated a large number of industrialized components and models to support our design.

IV. Introduction video of the industrial digital twin

SaaS End-Sales and Operation Planning and Multi-end alerting
PaaS Side-1. Intelligent Optimization
Intelligent Production Line Construction

In the process of designing the industrial model database, we went deep into the workshops and the factories of our customers, where we retrieved types of machinery that are universal and simulated a large number of real industrial models for Intelligent Production Line Construction. For example:

Preliminary effect of the model (Phase I)

V. Technical capacity

While we have upgraded our product, we are also continuously increasing the visual fidelity of the models. We explored the combination of Blender(C4D) + Substance Painter and found a way of multichannel PBR (physically-based rendering). (Channel: Base color\Metallic\Roughness\Normal\Emissive\Ambient Occlusion…) We have been utilizing this method for all of our wall screen visualization projects. By using DataV’s 3D Engine for online model rendering, we perfectly solved transparency and reflection related rendering problems. Theoretically, WebGL could work perfectly with Blender’s Eevee engine for real-time rendering. This complete GLTF/PBR node workflow could provide a well-aligned 3D product developing environment for designers and developers.

The entire workflow is modeling — UV unwrapping — texture/mapping — baking — online use.

Additionally, the newly released C4D Subscription22 version supports the direct export of online available GLTF format files and animations, greatly reducing the cost of learning new software. We are also continuously exploring the workflow.

C4D is one of the must-have tools for most visual designers today, covering a wide range of functions, including modeling, animation, rendering (offline rendering), characters, particles, newly added illustrations, and other modules. C4D provides a complete 3D creation platform and is widely used in marketing design, graphic design, motion design, character design, and etc. So integrating C4D as the entry point to the WebGL Graphic workflow is an essential part of the digital twin’s online rendering.

Therefore, in order to implement the digital twin in the industry, we explored the above two workflow paths to provide our customers with a multi-touchpoint, intelligent, and comfortable online environment.

Lastly, let’s take a look at the evolution of our product from physical simulation to the digital twin. In addition to the automatic generation of wall screen visualizations on the SaaS side, we have gradually added the twin model capability to the product side of the Industrial Brain to provide customers with an integrated product out of the box. Through the digital twin and design methods, we display complex capabilities in the form of data visualization + PBR (physically-based rendering), which greatly accelerates the pace of enterprise digital transformation.

VI. Final Thoughts

When customers interact with interfaces, they need to know immediately the meaning of the contents. Designers have to work hard to decode the content in every project. The longer it takes customers to decode the UI, the less usable it will be for them. Thus, the iterations from v1.0 data mapping to v4.0 digital twin is the process of designers refining complex data, expressing data more intuitively, and bringing customers closer to communicating with the product.

While helping to build smart factories, Industrial Brain is also infinitely lowering the threshold of factory access. The capabilities provided in our platform allow industrial enterprises without technical capabilities to also become smart factories and create quantifiable business value. For example, DunAn Energy’s O&M costs have been reduced by 30%, and the A-performance rate of Trina Solar has increased by 7%, saving the coal consumption of boilers by 2.6%…….

In summary, the digital twin in the design phase mainly includes digital model design and physical simulation design. Design tools include but not limited to C4D, Blender, Substance Painter, CAD, Rhino and etc. In this phase, the virtual design approach provides great visibility into the maintenance process, enabling design optimization predictions, and making the designer’s role irreplaceable in the industry optimization process.