Introduction to Itoh Shokai's Solution Business

In the manufacturing field, where continuous improvement in productivity and quality assurance is required, efforts utilizing AI are accelerating to quickly detect equipment malfunctions or failures and to prevent issues such as the occurrence of defective products. Various sensors and devices can be installed on production equipment, allowing for the collection of time-series data on measurement values and control parameters. The anomaly detection solution Impulse's "Anomaly Detection App" uses such time-series data to detect unknown anomalies by leveraging AI technology, rather than relying on traditional threshold-based systems. The "Cause Analysis App" can analyze the factors leading to the occurrence of anomalies. In this webinar, after introducing examples of time-series data analysis using AI technology, we will pick two specific cases from those examples and explain the creation and analysis methods of the anomaly detection model using actual screens and data, along with demonstrations. - Detection of early signs of plant equipment failure - Analysis of defect factors in the injection molding process of automotive parts manufacturing

In the manufacturing field, where constant improvement in productivity and quality assurance is required, efforts utilizing AI are accelerating to quickly detect equipment malfunctions or failures and to prevent issues such as the occurrence of defective products. Various sensors and devices can be installed on production equipment, allowing for the acquisition of time-series data from measurement values and control parameters. The anomaly detection solution Impulse's "Anomaly Detection App" uses such time-series data to detect unknown anomalies by leveraging AI technology, rather than relying on traditional threshold-based systems. The "Cause Analysis App" can analyze the factors leading to the occurrence of anomalies. In this webinar, we will introduce case studies of time-series data analysis utilizing AI technology, and from those, we will pick up the following two cases to explain the creation and analysis methods of anomaly detection models using actual screens and data, along with demonstrations. - Detection of signs of failure in plant equipment - Analysis of defect factors in the injection molding process of automotive parts manufacturing

In the manufacturing field, where constant improvement in productivity and quality assurance is required, efforts utilizing AI are accelerating to quickly detect equipment malfunctions or failures and to prevent issues such as the occurrence of defective products. Various sensors and devices can be installed on production equipment, allowing for the collection of time-series data on measurement values and control parameters. The anomaly detection solution Impulse's "Anomaly Detection App" uses such time-series data to detect unknown anomalies by leveraging AI technology, rather than relying on traditional threshold-based systems. The "Cause Analysis App" can analyze the factors leading to the occurrence of anomalies. In this webinar, after introducing examples of time-series data analysis utilizing AI technology, we will pick two specific cases from those examples and explain the creation and analysis methods of anomaly detection models using actual screens and data, along with demonstrations. - Detection of early signs of plant equipment failures - Analysis of defect factors in the injection molding process of automotive parts manufacturing

In the manufacturing field, the process of visual inspection plays a very important role in preventing the outflow of defective products and improving processes to ensure that defects do not occur. However, visual inspections conducted by humans are difficult to streamline, and even when attempting to introduce AI visual inspection devices, there are numerous challenges that often prevent full-scale implementation on the shop floor. (Issues with human visual inspections) - It is difficult to secure personnel, and it requires significant labor costs and time. - There is variability among inspectors. (Issues with visual inspection devices) - In inspections of metal processing surfaces, such as pressing and forging, there are frequent occurrences of false positives due to significant visual variability. - Operators struggle to effectively use the tools on-site, making it difficult to adjust rules and models after the introduction of automated inspections. - It takes time to explore the optimal imaging conditions for automated judgment. The abnormal detection solution Impulse's "Visual Inspection App" is chosen by customers who find that existing efforts and systems do not meet their needs. In this webinar, engineers who have been involved in actual projects will introduce the challenges surrounding visual inspections and their solutions.