H Steel Inspection

Case Study: Digitally transforming steel inspection

Oct. 31, 2022
Digital-transformation enablers, such as AI, allow continuous improvement to the customer’s operations, without additional knowledge and effort on their part.

By Daniel Young, Senior Manager at Toshiba America, Inc.

Toshiba has been selling MetalSpector steel-inspection equipment to several major steel companies for several decades. Recently, as part of Toshiba’s Digital Transformation strategy, Toshiba redesigned MetalSpector into a steel-grading evaluation service, intended to reduce steel inspector’s workloads while assuring the data integrity of the steel-grading process.

The result of this redesign uncovered a new set of services focused on how the inspection equipment is used by the customer, rather than focusing solely on improvements to how the inspection equipment works.

The redesign of the steel-inspection system was initially focused on improving how the system worked. This is a typical outcome for a company that is focused on hardware sales. However, as part of the company’s new focus on O&M services, the customer environment was also observed as a way of understanding how the QA inspectors used the steel-inspection equipment.

As a grade-determination service, that AI-inference model develops its own analysis skills, automatically, with each steel sample it analyzes. It also adapts to the analysis from veteran human inspectors, who have the final judgement decision. As the human inspector performs his normal measurement analysis, he is automatically training the AI model, with no additional burden on the inspector.

As a training service, the AI model can provide real-world image samples to junior inspectors and teach them to measure for non-metallic inclusions. Since the veteran inspectors continually train the AI model, they are also providing new training material for the junior inspectors. Thus, the burden on veteran inspectors is reduced, and junior inspectors gain experience from real-world data samples. The value of this new service offering was immediately recognized by the customer, as it reduced the number of veteran inspectors needed for training purposes, while simultaneously increasing the skill levels of junior inspectors.

From a hardware-sales business model perspective, the redesigned system solves the customer’s pain points via automation, with a strategy of minimizing the human role in the process. However, this is not the customer’s objective, as human inspectors are a critical resource and highly valued. The customer’s objective is to better enable inspectors to perform their critical role of quality assurance.

Gartner describes digital humanism as a means of redefining the way people’s goals can be achieved. Contrasted with automation, digital humanism focuses on minimizing the complexities and difficulties a human user has when using a system. Instead of automating system processes and tasks for the purpose of eliminating human operation, the goal becomes instead to empower human users to do more with the system, potentially unveiling new uses and providing new value that was otherwise unachievable.

It is important to note that this AI modeling is not intended to eliminate the human inspector from the process. In fact, it is critical to retain skilled inspectors for the steel manufacturers competitive advantage. Regulations also prevent the sole use of AI to perform metallic-inclusion inspection, so until such regulations are updated, human inspection is necessary.

The use of AI modeling is therefore intended to offload many mundane tasks and allow inspectors to focus on the non-repetitive analysis steps. While automation is a key process improvement, it is not the end goal. Supporting the steel inspector’s ability to identify issues with the steel-production process is the end goal; this involves both machine automation as well as improved human involvement. By combining automation with digital humanism, the focus on providing customer value can be maintained.

The successful deployment of MetalSpector helped validate the managed-services approach at the business level. Since then, new image-inspection services for other manufacturing and infrastructure applications have been developed, opening new business opportunities. For example, Toshiba has created a new managed service for the pharmaceutical industry, to measure surface defects in PTP packaging sheets. Similar managed services are also being developed for the inspection of film materials, paper, and non-woven cloth. Because these use cases also involve sample inspection, many of the image-analysis components developed for MetalSpector can be applied to these inspection cases, too.

The migration from hardware sales to As-A-Service sales is a challenging effort for many companies. Our fundamental change was from a reactive product-development design to proactive service offerings. Digital-transformation enablers, such as AI, allow continuous improvement to the customer’s operations, without additional knowledge and effort on their part. It also uncovers new use cases that help solve additional customer problems, which can lead to the creation of new services, providing new value to customers.

*This is a condensed version of a case study that appeared in the November 2021 edition of the Industry IoT Consortium (IIC)’s Journal of Innovation. View the full case study here.