HANOI, Vietnam–(BUSINESS WIRE)–Nov 6, 2022–

With the rise of data-centric MLOps, leading IT companies are adopting this approach to create better real-time inspection solutions in manufacturing. An interview with FPT Software’s Director of AI, Dr. Phong Nguyen, explored the potential of new technologies.

This press release is multimedia. See the full version here: https://www.businesswire.com/news/home/20221106005111/en/

Dr. Phong Nguyen (fourth from left) with FPT Software team at Hannover Messe – the world’s largest exhibition for industrial machinery and manufacturing technology, September 2022. (Photo: Business Wire)

Artificial intelligence (AI) applications are seen as the next step in the manufacturing revolution. In a report just released on November 3, ResearchAndMarkets.com calculated that the value of AI in manufacturing is $2.3 billion in 2022 and is expected to reach $16.3 billion by 2027; growing at a CAGR of 47.9% from 2022 to 2027.

In manufacturing processes, spotlights are turning to AI in quality inspection as technology and devices improve dramatically. In addition to high-resolution cameras to uncover tiny defects, manufacturers can choose to apply the new data-centric AI approach, which proves to be much more effective than rules-based inspection. For example, rule-based inspections require human experts to define a lot of rules and spend a lot of time maintaining and adding rules. An AI-based approach uses machine learning to teach machines to effortlessly recognize faulty features from examples.

Driven by Landing AI Founder and CEO, Dr. Andrew Ng, data-centric and focused on optimizing quality data to refine and shorten the time it takes to train AI, instead of spending time , computing resources and efforts to improve state-of-the-art models.

One of the benefits of data-centric AI is that it can increase accuracy through the ability to find new defects based on learned data. For FPT Software’s Director of AI, Dr. Phong Nguyen, it also significantly reduces the time needed to train the AI. According to LandingAI in one particular case, LandingLens – the data-centric AI platform specializing in visual inspection – took only 6 minutes to train and get the AI ​​up and running, while the machines based on periods took longer than 6 months.

“The platform paves the way for the application of data-centric AI solutions in factory quality inspection, as visual is the most common inspection approach. the data can not only be applied to visual AI, but also to the inspection of sound and IoT sensors,” said Dr. Phong.

Earlier this year, FPT Software shook hands with Landing AI in a strategic collaboration to distribute LandingLens to more than 1,000 IT company customers, focusing on the manufacturing industry. The collaboration means that while LandingAI uses collective data from real world problems to improve its platform, FPT Software provides a complete set of end-to-end AI solutions to its customers.

“With experience in vertical industries and technical know-how, our role is to evaluate each operating system, create algorithms based on the particular problems faced by the customer, collect and optimize data for train the AI ​​model and then deploy that model in the factory,” said Dr. Phong. “This way, we can help our partner develop and refine their platform, as well as provide a solution comprehensive to our customers”, he concluded.

In addition to LandingLens, FPT Software is also developing two additional data-centric solutions for sound and IoT sensor inspection, forming a comprehensive solution set for its customer base. These AI solutions are trained based on the data the company has collected over the years as they improve their models to solve various problems faced by factories.

“With the support of our customers, we can collect extremely valuable manufacturing data for research and train AI to learn from real-world problems and provide solutions,” said Dr. Phong.

In addition to manufacturing, data-centric AI can be used in many other ways within a business. According to Dr. Phong, FPT Software uses this approach to improve employee happiness, increase productivity through natural language code synthesis for people who don’t know how to code to be able to understand, and even predict time series to optimize the offer. for their supply chain.

“In a short time, the approach has been successfully applied in multiple ways. It can give a company a lot more insight and create long-term results,” Dr. Phong concluded.

Show source version on businesswire.com:https://www.businesswire.com/news/home/20221106005111/en/

CONTACT: Media contact

Mai Duong (Ms.)

FPT software

Professional manager

Email: [email protected] Duong (Ms)

Corporate Branding Manager

Email: [email protected] Website:https://www.fpt-software.com/newsroom/



SOURCE: FPT Software

Copyright BusinessWire 2022.

PUBLISHED: 11/06/2022 11:22 PM / DISK: 11/06/2022 11:22 PM