Macnica supports the simultaneous progress of Yokogawa Electric Corporation 's predictive AI project, successfully achieving ``long-term support prediction'' and ``inventory prediction'' on a global basis - Achieving high-precision prediction in a short period of time to accelerate DX -

Macnica (Headquarters: Yokohama City, Kanagawa Prefecture, President: Kazumasa Hara, hereafter Macnica), a total Services & Solutions provider in semiconductors, networks, cybersecurity, and AI/IoT, has recently announced that Yokogawa Electric Corporation Corporation (Headquarters : Musashino City, Tokyo, President: Yokogawa Electric Corporation Corporation by introducing H2O and Luminoso products.

■ Background
Yokogawa Electric Corporation, which provides instrumentation and control systems to various production sites, is the first step in providing solutions that support customers' DX (Digital Transformation). We are promoting DX with the policy of deploying the know-how acquired outside the company. The company, which is completely in-house and is working on DX promotion, has challenged sales forecasting by analyzing data from the core system, but it cannot be said that the forecast data is highly reliable, and it takes a certain amount of time and effort to utilize the data. was Therefore, in order to efficiently predict “what will happen in the future” with higher accuracy and utilize it in operations in the business department, H2O.ai’s machine learning automation platform “Driverless AI” and Luminoso’s natural language analysis AI “ Luminoso Daylight” has been introduced and a full-fledged predictive AI project has started.

■ Succeeded in predicting long-term support Some regions succeeded in predicting with an accuracy of 80% or more
Yokogawa Electric Corporation provides long-term plant warranties to ensure that our products are used stably, and our support policy is to respond quickly in the event of a problem. In this project, "Luminoso Daylight" will be used to conduct natural language analysis of multilingual inquiries from global customers to create an environment in which the meaning can be immediately understood and answered, and "Driverless AI" will be used. , by predicting issues that are likely to become problems and how long it will take to respond, we have realized smoother customer support. We have succeeded in predicting with an accuracy of 80% or more in the area with the highest accuracy.

■ Product inventory and even the number of parts associated with it can be predicted
Depending on the country, there are bases that must always have inventory due to business practices. In this project, we used "Driverless AI" to predict how much inventory we could have and how much we could sell in what period of time. We have achieved up to the prediction of In addition, in a project related to a product called a gas chromatograph, by using "Driverless AI", we were able to complete rough estimate work in 2-3 days instead of 10 days in the past.

■Reasons for the success of using predictive AI
Yokogawa Electric Corporation 's highly accurate and smooth progress in achieving highly accurate predictions using predictive AI, which is said to be difficult to increase accuracy even by 1%, and proceeding with as many as 10 projects simultaneously in one year. This is because Driverless AI itself has high prediction accuracy and can easily automate and speed up the creation of machine learning predictive models. Additionally, since the company had been working on data analysis using BI tools for some time, having an environment where they could judge whether the answers given by AI were correct was also a major success factor. The company has also praised Macnica 's support, which always considers the customer together from consultation to implementation, as one of the factors contributing to its success.

We have received the following comments regarding our efforts toward this support.

・H2O.ai Co., Ltd. Senior Vice President APAC Leader Abhinandan Kumar
"Through our sales contract for H2O.ai (product name: Driverless AI), Macnica and I have been providing solutions to support our customers' digital transformation and AI efforts in the Japanese market. We are very pleased to announce the results of the use of H2O.ai at Yokogawa Electric Corporation, one of Japan's leading manufacturers, as a public case study. H2O.ai is an automatic machine learning platform that condenses the world's highest level of know-how and rapidly realizes business efficiency improvements through AI that could not be achieved without excellent data scientists until now. We sincerely hope that Yokogawa Electric Corporation will use our solutions to further improve business efficiency and that Macnica 's AI business will continue to expand in the future.

Yoshinobu Kumamoto, Country Manager, Luminoso Japan, LLC
We are very pleased that our product played a role in the predictive AI project conducted by Yokogawa Electric Corporation for "natural language analysis of global customer inquiries". Luminoso" was designed based on a unique design concept to answer the question, "Can AI 'understand' language like humans? Luminoso is a text data analysis, classification, and retrieval tool with unique features developed based on our original design concept. We will continue to collaborate with Macnica, which provides support from the customer's point of view, to provide products that easily realize a problem that many companies are facing: the analysis and classification of large volumes of natural language documents.

Macnica will continue to leverage its advanced technology and support capabilities to support customers' DX promotion.

[Click here for details of the case]
https://www.macnica.co.jp/business/ai/manufacturers/h2oai/case_02.html

*Company names and product names mentioned in this text are trademarks or registered trademarks of Macnica and each company.
*The information published in the news release (including product price, specifications, etc.) is current as of the date of announcement. Please note that the information may be subject to change without prior notice.