3-5 Evaggelikis Scholis, 14231 Nea Ionia, Greece

10 Ways AI Is Improving Manufacturing In 2020

Originally posted on forbes.

  • Machinery Maintenance and Quality are the leading AI transformation projects in manufacturing operations today, according to Capgemini.
  • Caterpillar’s Marine Division is saving $400K per ship per year after machine learning analyzed data on how often hulls should be cleaned for maximum efficiency.
  • The BMW Group uses AI to evaluate component images in ongoing production lines to spot deviations from the standard in real-time.

Perceiving the pandemics’ hard reset as a chance to grow stronger, more resilient, and resourceful dominates manufacturers’ mindsets who continue to double down on analytics and AI-driven pilots.

Combining human experience, insight, and AI techniques, they’re discovering new ways to differentiate themselves while driving down costs and protecting margins. And they’re all up for the challenge of continuing to grow in tough economic times. They’re not alone in accepting that challenge. Boston Consulting Group‘s recent study The Rise of the AI-Powered Company in the Postcrisis World found that in the four previous global economic downturns, 14% of companies were able to increase both sales growth and profit margins as the following graphic shows:

AI Is Core To Manufacturing’s Real-Time Future  

Real-time monitoring provides many benefits, including troubleshooting production bottlenecks, tracking scrap rates, meeting customer delivery dates, and more. It’s an excellent source of contextually relevant data that can be used for training machine learning models. Supervised and unsupervised machine learning algorithms can interpret multiple production shifts’ real-time data in seconds and discover previously unknown processes, products, and workflow patterns.

The following are ten ways AI is enhancing manufacturing in 2020 based on Capgemini’s recently published Scaling AI in Manufacturing Operations: A Practitioners Perspective study and interviews with manufacturers over the last four months:

  • 29% of AI implementations in manufacturing are for maintaining machinery and production assets. Capgemini’s research team found that predicting when machines/equipment are likely to fail and recommending optimal times to conduct maintenance (condition-based maintenance) is the most popular use case of AI in manufacturing today. General Motors analyzes images from cameras mounted on assembly robots, to spot signs and indications of failing robotic components with the help of its supplier. In a pilot test of the system, it detected 72 instances of component failure across 7,000 robots, identifying the problem before it could result in unplanned outages. The following graphic from the study illustrates how AI is used for intelligent maintenance in manufacturing:
  • General Motors’ Dreamcatcher system is based on Autodesk’s generative design algorithm that relies on machine learning techniques to factor in design constraints and provides an optimized product design.  Having constraint-optimizing logic within a CAD design environment helps GM attain the goal of rapid prototyping. Designers provide a definitions of the functional requirements, materials, manufacturing methods and other constraints. GM and AutoDesk have customized Dreamcatcher to optimize for weight and other key product criterion essential for the parts being designed to succeed with additive manufacturing. The solution was recently tested with the prototyping of a seatbelt bracket part, which resulted in a single-piece design that is 40% lighter and 20% stronger than the original eight component design.. Please see the Harvard Business Review case analysis, Project Dreamcatcher: Can Generative Design Accelerate Additive Manufacturing? for additional information.
  • Nokia has introduced a video application that uses machine learning to alert an assembly operator if there are inconsistencies in the production process. Nokia launched the video application that uses machine learning to monitor an assembly line process in one of its factories in Oulu, Finland. It alerts the operator of inconsistencies in the process so that issues can be corrected in real time. Please see the article Nokia claims first “real-world” 5G smart factory trial with Telia and Intel for additional details.
  • Analyzing images in real-time to complete product quality inspections in the automotive and consumer products industries also helps manufacturers stay in compliance with stringent regulatory requirements. High-resolution cameras continue to drop in price while AI-based image recognition software and technologies continue to improve. These two factors and more are leading to greater adoption of real-time in-line inspection. Audi is a leader in adopting these technologies, having installed an image recognition system based on deep learning at its Ingolstadt press shop. The following graphic illustrates how real-time product quality inspection workflows work:
  • Improving demand forecast accuracy is showing solid results across multiple industries with Consumer Packaged Goods manufacturers leading all others. Danone Group is a French multinational food-products manufacturer who is using a machine learning system to improve its demand forecast accuracy today. They’re using machine learning to improve planning coordination across marketing, sales, account management, supply chain, and finance, leading to more accurate forecasts. Using machine learning, Danone can meet demand from product promotions and achieve its target service levels for channel or store-level inventories. The system led to a 20% reduction in forecast error, a 30% reduction in lost sales, a 30% reduction in product obsolescence, and a 50% reduction in demand planners’ workload. The following is an overview of how machine learning-based demand planning and forecasting systems are being designed today:
  • Thales SA, a leading supplier of electronic systems to a wide spectrum of industries, is using machine learning to predict preventative maintenance for high-speed rail lines throughout Europe. The company collects historical and current data on thousands of sensors, train parts, and the current state of subsystems across European transcontinental rail systems. Drawing on the data, it has developed an AI algorithm to predict potential problems and identify when specific parts need to be replaced, attaining a high level of reliability in the process. The following is a dashboard from their TIRIS  Big Data Analytics tool for Predictive Maintenance, supporting the rail industry to achieve a zero unplanned shutdown approach.
  • The BMW Group uses AI to evaluate component images from its production line, allowing it to spot, in real-time, deviations from quality standards.  In the final inspection area at the BMW Group’s Dingolfing plant, an AI application compares the vehicle order data with a live image of the model designation of the newly produced car. Model designations and other identification plates, such as “xDrive” for four-wheel-drive vehicles, as well as all generally approved combinations, are stored in the image database. If the live image and order data don’t correspond, for example, if a designation is missing, the final inspection team receives a notification. For more details on how BMW is innovating with AI across their many production centers, please see the article, Fast, efficient, reliable: Artificial intelligence in BMW Group Production. The BMW Group is also sharing its AI algorithms used in production on an open-source platform. Additional details on their open-sourcing of their AI algorithms can be found here. The following is an example of the AI-based image matching technology BMW uses today:
  • Schneider Electric created a predictive IoT analytics solution based on Microsoft Azure Machine Learning service and Azure IoT Edge to improve worker safety, reduce costs, and achieve sustainability goals. Schneider Electric data scientists use data from the oil field to build the models that predict when and where maintenance is needed. Data scientists use automated machine learning capabilities to intelligently select the optimal machine learning models and automatically tune machine model hyperparameters to save time and improve efficiency. When the company deployed the Azure Machine Learning service-based solution, it helped operators increase efficiency by 10 to 20% in just two days. Please see the article, Schneider Electric minimizes costs and worker risk with Azure Machine Learning service predictive maintenance for additional details. The following is a diagram illustrating the IoT Edge Analytics implantation, which includes the Azure Machine Learning Service:
  • Nissan is piloting the use of AI to design new models in real-time, hoping to reduce time-to-market for the next-generation model series. Nissan calls the program DriveSpark, and it’s been in existence for four years. Nissan designers are using the DriveSpark system to create entirely new models that comply with the latest industry compliance and regulatory requirements. They’ve also used AI to extend the lifecycles of existing models as well. For more information, see the article, DriveSpark, “Nissan’s Idea: Let An Artificial Intelligence Design Our Cars,” September 2016.
  • Canon has invented an advanced Asset Defect Recognition system that brings new levels of quality control to its manufacturing centers. Canon is combining human expertise, insight, and AI techniques, including machine learning, computer vision, and predictive modeling, to help drive increased accuracy and efficiency of testing high-precision parts of machinery. Manufacturers of precision parts in industries such as Automotive & Transportation, Aerospace & Defense, Oil & Gas, and Construction often require a rigorous post-assembly inspection process. Canon’s manufacturing components are closely examined using industrial radiography (X-ray images) and images to verify the integrity of each part and its internal structure. With computer vision and machine learning the Assisted Defect Recognition technology system can Intelligently analyze images of inspected parts, automatically identify potential defects, even those that may be missed by the human eye and learn and improve accuracy of the technology, based on human acceptance or corrections of the results. There are additional details of Canon’s technology initiatives here on the Artificial Intelligence/Machine Learning area of their site.

Source: forbes

Related Posts