Chip manufacturers are using artificial intelligence (AI) to improve the efficiency of the wafer factory operation, and the relevant efforts have begun to obtain returns.
A high-level high-level representation of MICRON TECHNOLOGY, chip manufacturers are using artificial intelligence (AI) to improve the efficiency of the wafer factory operation, and the relevant efforts have begun to obtain returns.
Buddy Nicoson, Vice President
Buddy Nicoson, a Vice President, who was responsible for the operation of Walay, said in the SemiCon Taiwan annual international semiconductor exhibition in Taipei, said that the wafer managers need to be more than those in the world due to mobile phones. Process technology for the continuous change of the manufacturing site synchronous deployment; all of these efforts are chip manufacturers to meet the yield and quality goals of the enterprise level as soon as possible.
NICOSON has more than 30 years of experience in the semiconductor industry. Before joining Mada in 2014, NICOM has helped Samsung, AMD and Cypress Semiconductor's Wafer; and in Magica is to help the company's first test, Round factory operation. He said in the speech: "AI has reached a speed of 25% faster than importing the demand-effective target, and the demand defect rate (DPM) target is 35%. "
Beautiful light has a wafer factory in the United States, Japan, Taiwan and Singapore, and its respective teams learn from each other; NICOSON explains, such as the experience of studying from big data from big data: "We hope to quickly spread through internal network." Ai, Miguang is transferring the wafer factory from the wafer factory to the remote control center; "You no longer need to send people to the scene as before," he pointed out: "With the dashboard settings the remote operation center, you can It has been a wide range of vision. "
NICOSON revealed that Magua will collect 5TB (Terabytes) every day; the company has 3,000 data capture equipment, the task assigned to different workstation teams and teams. "If there is a 3,000 units in your organization, 3,000 devices are not a lot;" he pointed out: "We are trying some very complicated things to translate our workers who are trying to change. We must develop A tangible interface tool, let us use it. "
Bill Wiseman, a senior partner of Mckipment, a management consultant, said that AI's potential application is a prediction of demand, and this task is to be more difficult to do. "When Apple published a new iPhone, you'd better know its best sales forecast, otherwise you will not keep up."
He pointed out: "If you are in addition to Apple, you have to supply other manufacturers' poor people, do you know who will win? Who do you ship? We have seen a lot of production over every node on the value chain And the scrapped; "and machine learning can provide better prediction for chip supply for market demand.
NICOSON has also heard similar saying that he said: "Now the production time is flow, every minute will change; if you lack the scheduler (Scheduler) that can real-time the environment where the factory is located, you will fall behind. And this It can be strengthened through virtualization. "
He believes that virtualization is a "big thing", complete virtualization strategy can display hidden loss and waste: "If you go out and talk to the on-site engineer, one of the setbacks they have encountered, that is, I will call" Sitting on the sky " (SILOED views) problem; they will go to a system to see some things, go to another system to see other things, but they are incomplete or incoherent. "
NICOSON said that unstructured data can be used to eliminate prejudice in the beautiful light: "Eventually you will see the hidden losses and waste of the past; now they become systematic signs (Systematic Signatures) (SYSTEMATIC SIGNATURES) ), And you can do some related things. "
Mckinsey's Wiseman also said that machine learning can tell the company's employees intend to leave; by forecast results, the company's management can act and stay before the employee walks. However, this also means that there is a new communication agreement between management and employees.
NICOSON pointed out that machine learning is a technology that is still in the initial development phase of chip manufacturers. One of its goals is the use of AI to identify defects on the silicon wafer in early process. "This is very similar to the face recognition." He pointed out: "This solution has two facing, geometric identification with three-dimensional optical (photometric) identification, which is very effective in the incoming process."
NICOSON also believes that the semiconductor industry is a bit embarrassing of the use of AI technology; this industry has just started value by AI technology.
Original link: https://www.eeboard.com/news/jingyuan/
Search for the panel network, pay attention, daily update development board, intelligent hardware, open source hardware, activity and other information can make you master. Recommended attention!
[WeChat scanning picture can be directly paid]
Technology early know:
After the old two, the boss, relied on ridicule skills to preheat MATE 10, Huawei this time is easy to play.
The new generation of domestic baseband and radio frequency integrated new generation of intellectual property rights, officially released
This 90 Chinese guy developed the world's first vertical take-off flight car
Sweep one, smarter! Xiaomi released a new rice scan robot
Also 10nm process, transistor density is twice that of Samsung and Tablet, Intel's 10nm process is invincible!
Our other product:
