Yokogawa Electric Releases Quality Prediction Inspection Tool

Yokogawa Electric Corporation Releases Equipment/Quality Convenient Predictive Detection Tool. A new AI-based function has been added to the OpreX data acquisition product lineup for building equipment and quality anomaly prediction and detection systems for GX-series, GP-series, and GM-series SMARTDAC+ paperless recorders and data collectors. With this software, even users who are not AI experts can build their own equipment and quality anomaly prediction and detection system for factories. The system helps improve productivity through early detection of equipment failures and quality degradation in the factory.

Equipment/Quality Predictive Detection Tool for SMARTDAC+

Development Background

Loggers and data collectors are used in production and development sites across a wide range of industries to collect, display, and record relevant voltage, current, temperature, flow, pressure, and other variables. As a leading company in this field, Yokogawa Electric has provided many of its customers with data consulting services and technologies such as machine learning to help them predict factory equipment and product quality problems, and to analyse and identify their causes.

In recent years, there has been a growing demand for AI-based solutions in order to improve productivity in factories. However, AI applications have a high threshold and require extensive expertise in specialised fields such as data science. To meet this demand, Yokogawa Electric has developed the Equipment/Quality Convenient Predictive Inspection Tool, an AI-based easy-to-use software application for loggers and data collectors commonly used in industry. No specialised AI expertise or consulting services are required to use this tool.

Features

1. No specialised knowledge required, AI can create predictive inspection models based on existing logging data

Predictive detection models are created by importing past data into the software and simply marking it as normal or abnormal, without relying on AI experts or consultants with knowledge of machine learning, algorithms, and more. Both data recorded by Yokogawa Electric products and data recorded by other companies' products can be used. By running simulations in advance, it is possible to understand how AI evaluates data.

Simulation using a predictive detection model

2. Using predictive inspection models, you can easily build equipment and quality predictive inspection systems

A system for predictive detection of equipment and quality anomalies can be built by loading the predictive detection model created by the software into SMARTDAC+ in the field. By monitoring health scores, the degree of equipment deterioration can be identified prior to failure. Health scores notify operators via alarms or emails when equipment requires maintenance, minimising the likelihood of unplanned failures affecting production activities.

Early detection of problems by monitoring changes in health scores

3. Cloud and offline versions available

Yokogawa offers both cloud and offline versions of its equipment/quality convenient predictive inspection tool. Customers can use either version to build equipment and quality anomaly prediction and detection systems. The cloud version is easier to access and does not require any installation on a PC.

Application examples

1. Managing temperature and pressure in tyre production (curing)

When processing tyre rubber, curing machines are used in the heating and pressurisation process, and pressure leaks due to packing deterioration can be a problem. Using the ‘Equipment/Quality Convenience Predictive Detection Tool’, the pressure value of the vulcanising machine is monitored to quantify changes in health scores for packing deterioration and to detect signs of pressure leakage in advance.

2.Managing Heat Treatment of Aerospace and Automotive Parts

During heat treatment of aerospace and automotive parts, product quality defects and furnace shutdowns can occur due to burner failure or inadequate sealing. With the Equipment/Quality

Convenience Predictive Detection Tool, you can catch signs of temperature abnormalities before an alarm occurs, avoid product loss, and predict when maintenance will be performed.