Visual inspection for improved quality in manufacturing

The challenges of quality inspection in manufacturing

Manufacturing operations strive to deliver the highest quality during every stage of the production or assembly process. Over half of these quality checks involve visual confirmation to ensure the parts are in the correct locations, have the right shape or color or texture, and are free from any blemishes such as scratches, pinholes, foreign particles, etc. Automating these visual quality checks is very difficult because of the volume of inspections and product variety, and because defects may occur anywhere on the product and could be of any size.

This is where IBM Maximo® Visual Inspection, an industry-leading AI-powered computer vision platform, delivers its highest value. As part of the IBM Maximo Application Suite, this automated visual inspection solution can monitor assets 24/7 and detect defects in under a second, dramatically enhancing operations to save time and money.

Learning from defect images that are “OK” and “NG”

Based on deep learning used by Watson, IBM Maximo Visual Inspection is designed for clients to automate visual quality inspections. Images of normal and abnormal products from different stages of production can be submitted to the centralized “learning service.” The learning service will build analytical models to discern OK vs NG characteristics of parts, components and products that meet quality specifications (OK) and those that don’t (NG). Further, if there is a need to classify defects into different types to address potential root causes and fix the quality issues, IBM Maximo Visual Inspection can be trained to perform such tasks with a high level of confidence.

Machine learning for continuous improvement in defect recognition

Based on advanced deep neural networks, the models trained by IBM Maximo Visual Inspection can be deployed on pre-configured edge devices on the factory floor so that there can be very little decision latency during production. Models trained in the MVI training servers can be deployed on MVI Edge or MVI Mobile. MVI Edge can be connected to be multiple cameras at the inspection point and allows you to scale your solution to multiple plants. MVI Edge interacts with Maximo Visual Inspection to allow you to create inspections that either collect images to train models or use trained models to identify objects in images. You can create inspection rules that define whether the identified objects pass or fail inspections, and automatically create work orders.

Inspections are the central configuration components in MVI Edge. The solution can learn continuously by taking feedback from manual inspectors, who can review the automated classification and override it based on human judgment. The corrective information, along with the image from the production floor, is then included in the next training cycle for that analytical model, thereby improving its ability to discern in the future. MVI Edge continues the journey through the suite by integrating with Maximo Monitor. SMEs can send the MQTT alert messages about inspections to generate them into the Maximo Monitor data lake for exploration and analysis.

Reduce dependency on manual inspection

In December 2021, Ford Motor Company presented its prestigious IT Innovation Award to IBM Maximo Visual Inspection. Voted on by Ford’s technical community leaders, this honor is granted once a year to the technology they believe has delivered the greatest breakthrough innovation driving value for the company. By delivering AI-enabled automation, IBM Maximo Visual Inspection allowed Ford to experience measurable success in the reduction of deficits and to scale out rapidly to multiple facilities.

In an increasingly complex operations environment, organizations use IBM Maximo Visual Inspection to harness the power of data and AI to derive real-time predictive business insights and make intelligent decisions.

Add intelligent “eyes” to your operations with IBM Maximo Visual Inspection.
World-class observability with IBM Maximo`