'AI for Architectural Design - Can it redefine industry productivity?' Technology Forum was held

"AI for Architectural Design - Can it redefine industry productivity?" Technology Forum held

On November 11, 2023, the Zhengzhou sub-forum of the China Computer Society Young Computer Scientists and Engineers Forum held a technical forum titled "AI for Architectural Design-Can it redefine industry productivity?" in Zhengzhou. The forum was hosted by the Computer Federation of China and hosted by CCF YOCSEF Zhengzhou. Xu Guoyu from Henan University of Finance and Economics and He Lei from Inspur Information Henan District served as executive chairs. More than 50 experts and scholars from the fields of architecture and computer gathered together to participate in the conference

In the thinking session of this forum, participants had intense discussions and reflections on the application of artificial intelligence in the field of architectural design and technical bottlenecks.

Thinking Topic 1: Compared with vertical fields such as finance and industry, what is the reason why AI has been slower to implement in the architectural design industry?

Xia Bing, Secretary General of CCF Zhengzhou Branch, believes that there are three reasons. First, the construction industry itself iterates very slowly, while artificial intelligence, on the contrary, develops rapidly, with new technology iterations every three months; second, computer practitioners lack domain knowledge and find it difficult to participate in construction software development; third, there is a lack of a large number of With the accumulation of raw data, the traditional construction industry cannot collect various information into data that can be used for AI training.

Zhang Jiangbo, Assistant President of China Construction E-Commerce Co., Ltd., believes that the construction industry used to regard digitalization as an increment rather than as the core. The current decline in the construction industry has made AI a very good track, and companies must carry out reforms.

People from the construction industry put forward different views. Some experts pointed out that the reason why the construction industry does not use large models is that it has not yet brought value and failed to truly empower the industry. Ma Zhengxiang, a professor at Zhengzhou Aviation Industry Management College, believes that it is unrealistic to adopt AI design for the entire building, because the characteristics of the building are personalized and need to meet the unique needs of each person. In addition, the construction industry still lacks standardized processes and products, and data collection is also very difficult.

Thinking Question 2: In the field of architectural design, what are the main technical difficulties faced by the application of artificial intelligence? Is it data, model or other aspects? What's the solution?

In response to this problem, Dr. Ding Songyang, deputy general manager of Third Dimension, believes that there are still many technical bottlenecks in AI large model technology that need to be solved. First of all, data collection and processing is a difficulty. The data of each design institute is private and not shared. Without data, large models cannot be effectively trained; secondly, there are many local regulations in the construction industry, and how to integrate this knowledge into the model is also a difficulty; thirdly , it is also necessary to consider how to integrate the mechanism model and the data model to achieve more accurate and explainable predictions and decisions. These issues need to be solved by academia and industry.

People in the architectural field generally believe that for the application of AI in the field of architectural design, data and algorithms are not a problem, nor is technology. The main reason is human and institutional. Liu Runqian, chief planner of the Henan Urban and Rural Architectural Design Institute, believes that it is normal for the construction industry to be resistant to AI. On the one hand, design institutes are generally market-oriented, and the first consideration is cost, not technology; on the other hand, during the design process, enterprises It requires constant negotiation and compromise with Party A, and it is difficult to complete this work using large models.

Thinking Topic 3: What technological innovations and changes will AI large models drive in the architectural design industry?

Some people in the construction industry believe that the first to be eliminated are "tool workers" who can only use tools, because their work has no technical content and is the easiest to replace. At the same time, regarding large models, there is a general sense of crisis in the construction industry. It is only a matter of time before some traditional construction industries and professions are subverted.

However, Inspur Information AI application architect Zhu Hong and other computer experts believe that designers’ widespread use of tools is a good thing. Designers can use tools such as large models to assist creative design, improve design efficiency, and free themselves from tedious and repetitive work. Therefore, they believe that tools should be used as much as possible, and large models will not disrupt the construction industry, but give designers more capabilities

Finally, Zhang Lei, chairman of the CCF YOCSEF Zhengzhou sub-forum, concluded: This forum is a cross-border exchange between the construction and computer industries. It is normal for everyone to have differences in opinions. We hope that this forum can bring useful exploration and inspiration to the application of large models in architectural design. (Zhang Lei)

The above is the detailed content of 'AI for Architectural Design - Can it redefine industry productivity?' Technology Forum was held. For more information, please follow other related articles on the PHP Chinese website!