Constantly increasing advances in technology and continuous process improvement techniques have helped ensure the efficient operation of the global supply chain, transforming raw materials into products that make their way to physical stores and e-commerce warehouses. However, we have seen how an external disturbance such as a pandemic or a flood can break the supply chain, throwing businesses and consumers into temporary chaos. These disturbances are not necessarily due to a lack of product but more to a lack of data. Where are the products in the supply chain at any given moment? How can they reach their destination when needed? When will there be spikes in consumer demand and how much more product will be needed to meet that demand?
Quality is another important aspect of manufacturing. Whether you’re talking about components on a high-speed production line or levels in packaging machines, every aspect of the manufacturing industry focuses on quality discovery and quality assurance. It is difficult to maintain high quality, even for experienced operators.
Manufacturers need to run their business from the windshield, not the rearview mirror. The computer vision platform provides the ability to “see” and interpret in near real-time by collecting and analyzing data at the edge at the right speed and velocity to make time-critical and agile business decisions.
This forward-looking ability provided by computer vision enables manufacturing organizations to better meet the top five business needs:
1. How can we improve manufacturing personnel and facility safety?
For individuals, cameras look for the use of personal protective equipment (PPE), such as hard hats and safety goggles, and then the system sends alerts to the manager if PPE is not being worn or tracks metrics used by the safety officer to determine if training is required. Geofencing creates a virtual border around equipment that is hot to touch or an area within reach of the robot; A loud alarm or flashing light is triggered when someone or something gets too close.
Repetitive motions is another safety issue that has long been anathema to manufacturing processes. Multiple repetitions of the same movement, as when assembling a circuit board, can cause injuries to workers. The use of cameras to study and improve workflow often results in fewer injuries to employees, promoting an overall safer workplace.
Thermal cameras are also increasingly being used in manufacturing to monitor things where sensors are not practical or to inspect things that are difficult for people to reach on site. For example, you can train the camera on the bearings of a conveyor system to watch the change in temperature, which will eventually produce a temperature graph that shows a heat signature over time that may indicate a future failure. Thermal imaging is also useful if visibility is required in low light, such as securing the perimeter in dimly lit areas.
2. How can we increase manufacturing operational efficiencies today and what can we expect in the future?
Operational efficiency usually comes down to the overall effectiveness of the equipment. If you’re not producing a quality product, you have to either get rid of it or rework. Increased production is required to make up for out-of-specification products to meet the product demand plan.
Consider a product line that uses automation technology, such as Programmable Logic Controllers (PLCs). The cameras can be used to capture real-time streaming video images and thermal images of the production line and equipment.
This highly scalable computer vision platform will enable you to integrate and aggregate data, and then analyze the resulting data set to determine the cause of a product that is out of specification. Taking this one step further – with a computer vision platform that takes advantage of a standardized approach – analysis occurs in near real time as there will be no need to return large amounts of data to a central storage location. This approach uses edge models to analyze the data, sending the results only to the central location for training and model optimization (Fig. 1). After that, only the updated form is sent to the edge. It’s a much more efficient way to determine if a piece of equipment or a piece of equipment is running a little slower or a little hotter versus doing physical inspections more accurately and quickly.
Figure 1. Transition to real-time situational awareness and insights. An example of a computer vision workflow where data capture, organization, movement, preprocessing, security archiving, data training and inference provide real-time insights at the edge.
Other uses of computer vision include being able to distinguish colors in the quality control process (are potato chips a light golden color or brown and overcooked?) and finding defects in hard-to-see parts within large assemblies. Thermal imaging can measure levels of product in a tank, whether the product is crude oil in the field or soup in a factory, as well as the temperature of those tanks.
Looking to the future, predictive maintenance will be enhanced by computer vision technology, especially for equipment that needs to be in constant motion, reducing the time frame for replacements because scheduling can be done in advance and reducing overall maintenance costs. Computer vision will be able to capture more features from images and video, and may well replace many of the sensors and transmitters in use today. Also expect to see entire factories built virtually, creating a digital twin with running processes that can be optimized before implementation.
3. How can we positively influence the employee or “people” experience?
Expect major improvements in the future to workday operations that currently require manual participation and analysis. When an employee enters a factory, computer vision may recognize that person, record the time the shift begins (no time-card punch needed) and automatically open doors to restricted areas. This is also related to safety. In the event of an emergency, plant managers and first responders can get instant statistics of the number of people in different locations. Augmented reality (AR) will allow more flexibility in the types of skill sets required for specific jobs by pairing remote experts with lower-skilled workers who wear cameras and interact via augmented reality.
4. How can we influence sustainability in manufacturing?
One of the desirable outcomes of process improvement is to reduce manufacturing consumption in terms of energy, chemical use, air quality and raw materials. Does the engine need to run 24/7? Probably not, and a computer vision platform can look for patterns to determine when that motor should run and when it can idle, affecting maintenance efforts and energy use. The same principle applies to the use of chemicals, where optimum use also means reducing over and under doses. And the use of quality computer vision production reduces scrap that cannot be reformulated and ends up in landfills.
Cameras are also an effective way to monitor stack emissions as part of an effort to reduce sulfur oxide (SOx) and nitrogen oxide (NOx) emissions, and to check the color of the water, as in a tank, to determine the presence of pollutants or algae.
5. How can we influence manufacturing revenue?
Business owners look at ROI, net present value, and other metrics to justify the financial statements for undertaking a project or introducing a new product line. Computer vision is used in manufacturing to increase revenue, reduce costs, and increase worker safety. For example, maximizing the quality of the first pass increases yield or reduces scrap, which saves money in raw materials. To further the sustainability discussion, reducing the use of energy or chemicals also lowers costs.
Although some aspects of computer vision have been used in manufacturing for years, real-time data analytics has the greatest impact. With more factories embracing computer vision for quality assurance, plant visibility, equipment monitoring and worker safety, it is possible to envision a future global supply chain that can better respond to the variables of external disturbances and continue to operate smoothly under almost any condition.
For an overview of computer vision and its impact on manufacturing, see Delivering game-changing production quality with computer vision.
Learn more about how computer vision has positively impacted other industries:
Intel® Technologies Move Analytics Forward
Data analytics is the key to unlocking the most value you can extract from data across your organization. To create a productive and cost-effective analytics strategy that delivers results, you need high-performance hardware that is optimized to work with the software you’re using.
Modern data analytics encompasses a range of technologies, from custom analytics platforms and databases to deep learning and artificial intelligence (AI). Are you just getting started with analytics? Ready to develop your analytics strategy or improve the quality of your data? There is always room for growth, and Intel is ready to help. With a deep ecosystem of analytics technologies and partners, Intel accelerates the efforts of data scientists, analysts, and developers in every industry. Find out more about advanced Intel analytics.