BCG-生成式人工智能在未来工厂中的作用（英）_市场营销策划_重点报告202301202_doc.docx

《BCG-生成式人工智能在未来工厂中的作用（英）_市场营销策划_重点报告202301202_doc.docx》由会员分享，可在线阅读，更多相关《BCG-生成式人工智能在未来工厂中的作用（英）_市场营销策划_重点报告202301202_doc.docx（15页珍藏版）》请在第一文库网上搜索。

1、Generative APs Role in the Factory of the FutureDECEMBER 08,2023By Daniel Kupper3 Kristian Kuhlmann3 Monika Saunders3 John Knapp5 Kai-Frederic Seitz5JuIian Englberger5 Tilman Buchne 5 and Martin KleinhansREADING TIME: 15 MINGenerative Al is one of today,s hottest business topics, with companies expl

2、oring its potential applications and benefits across industries and functions, including manufacturing. But despite the recent buzz, manufacturers should recognize that simply applying tools like ChatGPT on their own will not revolutionize factory operations.Instead of replacing traditional Al, GenA

3、I offers complementary use cases in the areas of assistance, recommendations, and autonomy that pave the way to the factory of the future. It does so through its capacity to generate content, such as text and images, tailored to specific tasks or inquiries. (See uHow GenAI Works/5)Howgenai works-To

4、discuss the applications of GenAI5 it is essential to first define how it differs from “classical machine learning (ML). Classical ML algorithms discern patterns within observed data, enabling them to generalize these insights to new, previously unseen data. For instance, an ML model might be traine

5、d using specific text fragments-such as operator incident reports in which machine breakdown descriptions are classified into specific root causes such as “end of tooling life or operator error. Based on this training, the model can process previously unseen text fragments of incident reports and ju

6、dge what caused the incident. The basis for such models may be deep neural networks, support vector machines, or other methods.GenAI takes this approach further. Beyond merely classifying existing text, it can generate new text based on specified criteriasuch as operator instructions that outline a

7、process to resolve a particular root cause of a machine breakdown. Although the progression from classical ML to GenAI might seem incremental, it poses a fundamental technical challenge. In classical ML5 the model merely needs sufficient training to confidently categorize a text fragment. In contras

8、t, GenAI must construct a text fragment from individual words and letters, ensuring that it is grammatically correct, comprehensible, and accurately represents the process.The number of potential outputs from GenAI is virtually limitless. Considering that there are roughly 170,000 English in current

9、 use, a mere five-word text has more than 140 septillion potential combinations. On the other hand, only a fraction of them would be grammatically correct and understandable. Among those, an even smaller fraction would accurately describe a given process to fix the root cause of a machine breakdown.

10、Consequently, the margin for error in GenAI models is incredibly narrow, necessitating extremely precise models. To attain this precision, GenAI must use foundational modelsj, instead of being trained only on context-specific data. Foundational models are trained on extensive datasets, such as all a

11、vailable text or images online, and are subsequently fine-tuned for specific applications. These models can be large language models (such as OpeAs GPT-4 or Amazon Q) or image or speech models. They seem to gain an understanding of reality from the extensive datasets. However3 foundational models ar

12、e observational learners that do not apply logic or reason as humans do. This means that there is no guarantee of plausible or accurate results from GenALIn our operator instruction example, the foundational model first learns what constitutes comprehensible and accurate text, with process descripti

13、ons being a small subset. Next, the model is fine-tuned by learning what operator instructions look like and how they correlate with given machine breakdown root causes. However, there is no assurance that the model will create correct or high-quality operator instructions. Ergonomics illustrates th

14、e problem. Because the GenAI model lacks insight into the process that fixes the root cause and the people who are the operators, it might overlook potential limitations, such as infeasible movements or inaccessible spaces. As a result, a quality assessment is always required to ensure that the reco

15、mmended remediation is practical from an ergonomic perspective.Gen As greater technical complexity elevates the importance of establishing a robust technological foundation to harness its capabilities effectively. With a number of archetypes possible, manufacturers must understand the factors that d

16、etermine an optimal choice. They can apply this knowledge to integrate GenAI into factory operations, considering the value-adding applications, change- and people-related initiatives, and required technological infrastructure.Manufacturers Are Prioritizing GenAI for its Disruptive PotentialBCG recently surveyed manufacturers to understand their perspective on technology developments. (See uAbout the Survey.)Regardless of their affinity for digital te