Industrial Internet "ruthless people" gathered together to discuss how to predict the future in the key year of implementation? | CCF-GAIR 2020

Author | Guo Renxian

In 2017, the State Council issued the "Guiding Opinions on Deepening the Development of Industrial Internet through 'Internet + Advanced Manufacturing'". This year marks the final year of the Three-Year Action Plan for the Innovation and Development of the Industrial Internet, and it is also the first year that the Industrial Internet has been classified as "new infrastructure".

At this time of transformation from old to new driving forces, originally, "ruthless people" refer to people who have extraordinary abilities in certain aspects. Just like now, under the baptism of the epidemic, the complex international environment, and the new demands of the manufacturing industry and other multiple factors, a wave of industry backbone forces have emerged in actual combat in the industrial Internet.

On August 9, Shenzhen, China, the 2020 Global Artificial Intelligence and Robotics Summit (CCF-GAIR) kicked off at the JW Marriott Hotel Shenzhen Qianhai OCT. The conference was hosted by the China Computer Federation (CCF), co-organized by the Chinese University of Hong Kong (Shenzhen) and Leifeng.com, and co-organized by Pengcheng Laboratory and Shenzhen Institute of Artificial Intelligence and Robotics. Among them, the "Industrial Internet Special Session", which was held as the finale of the conference, brought together the industry leaders in the current front line of China's industrial Internet industry. In 2020, the key year for the implementation of the industrial Internet, together they helped industry professionals to understand the future trend.

Liu Jiangchuan, academician of the Canadian Academy of Engineering, IEEE Fellow, and distinguished professor at Simon Fraser University in Canada, Liu Zongchang, chief data officer of Foxconn Industrial Internet and general manager of the Technology Service Group, Huang Luchuan, senior vice president and CMO of TreeRoot Interconnect, Qin Cheng, CTO of PTC China, Yang Baogang, senior vice president of UFIDA Network, Zhao Jianchun, vice president of Tencent Cloud, and Cai Shenglong, vice president of Saiyi Information and general manager of the industrial Internet subsidiary, all gave important reports, sorting out the "new infrastructure" of the industrial Internet from multiple perspectives, including the practical applications and latest explorations of industrial Internet platforms, industrial AI, industrial brains, edge computing, industrial software, etc.

At the beginning of the conference, Liu Jiangchuan, a member of the Canadian Academy of Engineering, IEEE Fellow, and distinguished professor at Simon Fraser University in Canada, who has been working in the academic world for many years, analyzed the three waves of the Internet mentioned by famous Silicon Valley investors for the participants. The specific situation is as follows:

• The first wave was from 1985 to 1999. Their carriers were mainly PCs, and their purpose was to connect PCs distributed around the world.

• The second wave, from about 2000 to 2015, was the time to load applications on top of connections. At the same time, carriers gradually shifted to smaller mobile phones starting in 2007.

• The third wave, which started in 2016, saw the emergence of various sensors, especially industrial sensors, as new carriers of Internet access.

Leifeng.com learned that both the first and second waves were basically centered around the consumer Internet industry. Now we are experiencing the third wave, and the industrial Internet not only connects everyone to communicate with each other, but also gives birth to distributed data and data computing engines.

"The so-called industrial Internet is not just about interconnection, and today's Internet is not just about interconnection like it was at the beginning," said Professor Liu Jiangchuan.

"From the scenarios of consumer Internet, industrial Internet, and energy Internet, we can see that there is some inertial thinking in consumer Internet. These people believe that a highly reliable and high-performance infrastructure has been established in the past decade or so. For example, long-term uninterrupted work and backup power supply in the event of power outages consume a lot of electricity, which is exchanged for the value of applications. On the other hand, the network infrastructure is also very good, with high-throughput and low-latency network infrastructure, now reaching Gbps. With 5G, the delay can be reduced to about 10 milliseconds or even 1 millisecond. On massive clients, massive amounts of data can be obtained, and data can be obtained without restrictions. Privacy is not a big problem, and digital value is reflected."

When it comes to the hottest topic this year, "new infrastructure" is definitely the top priority, and the industrial Internet has attracted much attention as a key element. Huang Luchuan, senior vice president and CMO of TreeRoot Interconnect, took the theme of "New Infrastructure Industrial Internet Empowers Industrial Chain Transformation and Upgrading" and analyzed in detail how China's manufacturing transformation and upgrading is accelerating under the impact of the COVID-19 pandemic and the Sino-US trade frictions.

Huang Luchuan said: "Industrial Internet is actually equivalent to the highway in infrastructure. After the construction of Industrial Internet, related industries will be driven. IoT gateways, sensors, corresponding 5G communications, industrial Internet platforms, industrial software, and employment are the driving force of new industries. In the process of building Industrial Internet, after the completion of Industrial Internet, we can compare it to the highway, which will form an information and data highway, and build a communication highway that can be used for upper-level applications based on on-site equipment and on-site industrial production data. This highway is long-term and can help our country achieve real digital transformation and reform of new infrastructure and improve the efficiency of the entire industrial chain."

Foxconn Industrial Internet Co., Ltd., also known as Foxconn Industrial Internet Co., Ltd., is a corporate representative that not only understands industrial manufacturing but can also serve as a "testing ground" itself. How does it plan its layout?

It has been two years since they were listed on the Shanghai Stock Exchange in June 2018. During this period, they officially launched the service main architecture of the industrial cloud platform (Fii Cloud), professional cloud (Micro Cloud) and industrial artificial intelligence (IAI). At the same time, they also established the dual-wheel drive strategy of "intelligent manufacturing" and "industrial Internet".

At the forum, Liu Zongchang, Chief Data Officer of Foxconn Industrial Internet and General Manager of Technology Service Group, shared with participants "The Value Opportunities Brought by Industrial Artificial Intelligence and Industrial Internet to the Manufacturing Industry".

"We say that the smart manufacturing system will go through five stages of development, and there is no way to skip these five stages," said Liu Zongchang.

"If we don't do lean well, it may be difficult to do informatization. If we don't do informatization well, let alone big data and intelligence. Now we have a dedicated technology service business group to serve many companies. We have contacted hundreds of companies and we have noticed many excellent companies that have done well in terms of scale and profitability. If they can improve the first two parts, they will become the top companies."

In the era of consumer Internet, high-market-cap companies such as "BAT" were born, profoundly affecting our lives and studies. In the era of industrial Internet, how will "BAT" continue to write brilliant achievements?

Tencent Cloud Vice President Zhao Jianchun gave an in-depth interpretation of their implementation practices in the field of industrial AI with the theme of "Tencent Cloud Artificial Intelligence Practice in the Field of Industrial Internet". Tencent Cloud relies on its C-end accumulation and three major AI laboratories, Tencent AI-Lab, Tencent Youtu, and WeChat Artificial Intelligence, to build Tencent's full-stack AI capabilities.

Then, Zhao Jianchun focused on interpreting the overall architecture of Tencent Cloud's industrial products. He said:

"The underlying capabilities are the well-known IaaS platform, IoT platform and 5G communication. Above them are four major bases, including industrial big data base, industrial business base, industrial marketing base and industrial intelligence base. Above them are products and ecosystems that support the service ecosystem, including industrial ecosystem capabilities, industry customization and ecosystem integration, to jointly support the platform that provides digital solutions for regional enterprises. Currently, there are 8 platforms in the country. There are also pan-industrial solutions driven by projects and bases."

Leifeng.com learned that the industrial intelligence foundation is based on Tencent's basic cloud at the bottom, and above it are basic service products, big data capabilities, AI capabilities, engines, computer language recognition, and natural language processing. Upward are application service products, including Tencent Cloud artificial intelligence service platform, intelligent dialogue platform, data asset management platform, etc. Upward will be a variety of one-way applications, such as face authentication, text recognition, face recognition and image recognition, etc. Above are industry solutions, such as smart industry, smart transportation, etc.

Finally, Zhao Jianchun analyzed Tencent’s specific practical cases in the field of industrial AI, such as the Huaxing Optoelectronics panel ADC (automatic defect detection) project, the factory (large assembly) labor efficiency optimization project, the equipment predictive maintenance project, the energy industry knowledge graph project and many other cases.

It is understood that there is a consensus on industrial digital transformation that the future world is an intelligent and interconnected world. Similarly, industrial enterprises are no exception. Whether it is intelligent factories, products, or services, they are all inevitable trends. However, in this process, data-driven enterprise intelligence will be a huge challenge.

Qin Cheng, CTO of PTC China, shared “Digital Integration: Industrial Internet Reconstructs Digital Enterprises”. He said:

"What is digitalization? It is based on informatization and combines all new technologies to apply to all value chains of enterprises. It is not about people adapting to huge systems, but about the data in all systems adapting to you, finding you and empowering you according to your work scenarios and roles. This is the value of digitalization."

In addition, he mentioned that the industrial Internet platform is just a technology that ultimately empowers all aspects of an enterprise. All data from the underlying system forms an invisible and continuous "0101" river. Marketers will look at the marketing status, designers will look at the design status, and manufacturers will look at the manufacturing status. Everyone obtains information differently. The value of the industrial Internet platform lies not in the equipment, but in the empowerment after the interconnection of the entire value chain and all factors of the enterprise.

As a member of the national team of "double cross" industrial Internet platform, UFIDA Network has been doing things in corporate office buildings, from financial software to management software. Later, their management software was used to manage all affairs in office buildings. With the advent of the Internet of Things, cloud computing, and big data, UFIDA's industrial Internet has extended beyond office buildings to factories, workshops, warehouses, and to the industrial chain.

Yang Baogang, senior vice president of UFIDA Network, said: "UFIDA's understanding of the industrial Internet also comes from digitalization, networking, and intelligence. We believe that intelligence is the ultimate goal of the development of the manufacturing industry, but intelligence must be based on digitalization. Without digitalization, there is no information about the equipment. All it has are mechanical signals from the past. Traditional physical signals cannot be connected, transmitted, or processed by computers."

Afterwards, Yang Baogang gave an in-depth interpretation of UFIDA's Intelligent Industrial Brain, and he emphasized:

"The main scenarios of UFIDA Industrial Brain include batching, batching management of factories, fuel batching, steel batching, etc. We use operations research methods to solve batching methods and decisions. We also optimize scheduling, find the optimal solution for various constraints of multiple products and multiple production lines. In addition, we use machine recognition, image recognition and other related content combined with some algorithms provided by deep learning."

Leifeng.com learned that UFIDA Jingzhi Industrial Brain mainly consists of intelligent algorithms with operations research as the core, models and algorithms made with image recognition and related recognition technologies with deep learning as the core, and predictive models with process mechanisms as the core.

Different companies will take different paths in their digital transformation. Some choose to start by building a platform, while others choose to start with a vertical solution-oriented solution accelerator.

In November 2018, Saiyi Information and Huawei Cloud jointly launched the Saiyi Industrial Internet Platform, which released its capabilities accumulated in the manufacturing industry over the years and became the "third pole" of the industrial Internet field. At this conference, Cai Shenglong, vice president of Saiyi Information and general manager of the industrial Internet subsidiary, gave an in-depth interpretation of "How to realize digital asset management under the transformation of manufacturing model".

Cai Shenglong said: "The transformation direction of smart manufacturing enterprises, we talk about digital transformation every day, and there are various types of digital transformation. Different enterprises, different industries, and different stages of transformation have different starting points. Some are engaged in business model transformation, some are engaged in R&D model transformation, and some are engaged in service model transformation."

Talking about how to realize the digitalization of asset management and the framework thinking of asset management, Leifeng.com (official account: Leifeng.com) learned that what asset management manages, how to manage, and how to gradually improve the asset or equipment management level are all issues for many manufacturing companies that are undergoing manufacturing model transformation. Its maturity model comprehensively considers asset value and business impact, enterprise asset management awareness (from cost to investment), and combines four aspects of graded application scenarios, benefits, shortcomings, and required technical support, and divides it into five maturity levels: passive maintenance, planned maintenance, condition-based maintenance, predictive maintenance, and specification-based maintenance.

The industry predicts that the industrial Internet will be a track that will continue to expand and become more prosperous. At present, many companies in different fields and regions in China are accelerating the digital transformation of industry. Especially after experiencing the baptism of the epidemic, the industrial Internet has officially become the "super outlet" of the industry.

According to statistics from the Industrial Internet Alliance, there are hundreds of industrial Internet platforms of various types in China, and more than 50 platforms have a certain regional or industry influence. In addition, the situation in each branch and each sub-field of big data platforms, industrial AI, industrial software, etc. is even more complicated.

"After several years of fierce competition, the Industrial Internet has become an important path and infrastructure for achieving digital transformation for enterprises," said Liu Qiang, senior vice president of PTC and president of Greater China.

"In 2020, the industrial Internet has once again become a hot topic as one of the important development directions of the 'new infrastructure'. Driven by both policies and the market, the industrial Internet will usher in unprecedented development opportunities."

After a key year of implementation, as an important cornerstone of the Fourth Industrial Revolution, the Industrial Internet, from local pilot projects to full popularization, what changes will it bring to traditional manufacturing models, production organization methods and industrial forms?