What is Smart Manufacturing?

Smart manufacturing is the digitization of manufacturing at all levels — from the optimal design of products to the optimization and streamlining of production processes. It also incorporates the management of manufacturing lines through predictive maintenance and preventive maintenance, which spots issues before they arise on manufacturing lines. Beyond that, smart manufacturing extends into the wider ecosystem of a manufacturing business, including sales and the industry supply chain.

Smart manufacturing combines hardware and software, ranging from smart sensors, Industrial Internet of Things (IIoT) devices, big data, cyber-physical systems, and robotics to artificial intelligence (AI) and machine learning (ML), real-time data analytics, simulation, cloud computing, and digital twins. New manufacturing methods, such as additive manufacturing (3D printing) have also become a key part of smart manufacturing as it enables more optimized products to be created.

The evolution of smart manufacturing has brought a new level of optimization and automation to many manufacturing industries — including automotive, aerospace, defense, healthcare, energy, and semiconductors — and is helping improve the efficiency, sustainability, and decision-making of manufacturing processes in these industries. The advanced technologies developed and matured throughout the digital transformation of the Fourth Industrial Revolution — that is, Industry 4.0 — has helped smart manufacturing become possible today.

Smart manufacturing revolves around real-time data and data-driven technologies that allow the manufacturing line to adapt to any changes based on business demand and needs. Through data collection from the many different sensors on the production line, it optimizes the different parameters based on the expected output and identifies any potential faults before they happen through learned and predictive behavior. The hardware-software combination of smart manufacturing helps keep manufacturing lines running smoothly and free from downtime and disruption. Many components are responsible for the capabilities that smart manufacturing offers.

IoT, IIoT, and Edge Computing

Many of the devices in a smart manufacturing environment are part of the Internet of Things (IoT) network, and IIoT is a core part because it connects physical and digital assets, allowing informed decisions. This includes smart sensors that are wirelessly connected to the digital network and uploading the data for the software to analyze. This allows the smart network to make automated control decisions based on the obtained data against the historical data. The IoT network also includes low-cost processors that perform some computing tasks locally — such as sensor fusion that gathers and processes the data from sensors — before the data collection is uploaded to the cloud to prevent lag in the IoT network. This is edge computing and has become a key part of smart manufacturing IIoT infrastructure.

Simulations and Digital Twins

Simulations and digital twins (virtual environments) are both used in smart manufacturing to create and optimize physical products digitally so all tolerances and parameters can be determined, validated, and tested before production starts. This makes manufacturing errors much less likely due to pre-optimized processes. Digital twins can also simulate an entire smart factory and all associated assets to ensure that everything runs smoothly and offer a way of continually optimizing the process once production has begun. As real-time physical data can be fed into the digital twin to run potential scenarios, that can then be applied to the physical assembly line.

Cloud Computing

Cloud connectivity and cloud computing allow IoT sensor data to be stored and then analyzed by ML algorithms that operate from off-site servers (where there is more space to house the data centers). Using the cloud enables all the analytics and automation to be performed wirelessly. Plus, in the era of 5G connectivity, there is a lot less latency and there are higher speeds, both of which enable smart manufacturing to be performed on larger scales.

AI and ML

AI and ML are a key part of the data analysis side of smart manufacturing as they identify patterns in the data that can optimize processes, reduce product defects, and spot manufacturing/equipment faults before they occur. AI and ML go alongside many other data analysis algorithms and form a key part of the “smart” in smart manufacturing. Additionally, AI and ML are used outside of these core areas, including in IoT edge computing microprocessors and smart factory robots (such as cobots and other autonomous robots).

Additive Manufacturing

Additive manufacturing has become a key part of smart manufacturing technology. While it’s not part of automation technology, 3D printing is a key technology on the design and production side. Additive manufacturing offers a way to reduce manufacturing times, create rapid prototypes, and develop parts with complex geometries and internal channels that often require expensive manufacturing steps.

SCADA

Supervisory crowd and data acquisition (SCADA) coordinates the flow of materials and inventory in a manufacturing environment in real time. SCADA performs real-time system monitoring and is directly connected to control systems to monitor the performance of manufacturing systems and provide alerts when something is wrong.

Making the transition from traditional manufacturing processes to smart manufacturing systems has a number of efficiency, productivity, quality, sustainability, and economic benefits, including:

• Improved yields: a 20%-40% reduction in defects, a 15%-30% improvement in first-pass yield, and up to a 50% reduction in waste and scraps in forming and machining processes.
• Reduced time: a 20%-30% reduction in product development cycles, up to a 50% reduction in development errors, and 40% faster time to production.
• Decreased costs: a 20%-40% reduction in research and development (R&D) and prototyping costs, a 10%-20% raw material cost savings, and 15% energy savings due to optimized processes.
• Increased sustainability: a 25% reduction in carbon emissions, 20%-30% water and resource savings in manufacturing, and 15% improvement in energy efficiency.
• Improved process and quality control: a reduction in process variability and undetected errors across the assembly line that helps reduce product recalls, increase product quality, and improve supply chain management.

Alongside the operational aspects of the assembly line, a big part of advancements in smart manufacturing is designing optimal products that are more sustainable, more efficient, quicker to manufacture, and cheaper to produce, all before manufacturing starts.

This can take a number of forms, but digital technologies and simulation capabilities are crucial. On one hand, modeling different manufacturing processes can help identify when any defects are added into a product and if they are likely to fail (and after how long). A virtual environment can be run to look at how these defects can be removed or where tolerances in the product are OK to accommodate a certain level of defects. Virtual environments also help identify the best manufacturing techniques for a product by identifying which approaches introduce too much stress into the material — something that can quickly lead to product failure.

Once the subprocesses have been modeled, reduced-order models can be integrated into the system to identify when any manufacturing equipment is starting to deviate and move out of tolerance. Both aspects help improve the safety and quality of products while improving the yield of assembly lines due to fewer product failures. This can be applied to traditional manufacturing lines and additive manufacturing processes because the heat of the melt pools can also be monitored in additive processes.

Another key part of product design today is sustainability. Designing with simulation in smart manufacturing helps make parts more recyclable by enabling the choice of raw materials and by improving energy efficiency through optimized processes. It often involves the balance of materials relative to their recyclability and the energy they cost to produce. For example, steel is easier to recycle than a lot of plastics, but it requires a lot more energy to produce. Similarly, batteries from electric vehicles (EVs) are much harder to recycle than a traditional engine, but an EV reduces localized greenhouse gas emissions. Additive manufacturing also plays into this because instead of performing disconnected processes at multiple facilities, it is sometimes more energy-, material-, and time-efficient to use additive manufacturing.

Simulation approaches, such as model-based systems engineering or model-based process engineering, can simulate many manufacturing processes. This includes extrusion, mixing, crushing and grinding, stamping, forming, casting, coating, welding, soldering, and additive manufacturing, as well as many more common manufacturing methods. Simulation looks at both the part and the process, and a number of Ansys tools are used in these simulations, including:

• Ansys LS-DYNA nonlinear dynamics structural simulation software: used to model manufacturing subprocesses, such as stamping, metal forming, or rolling.
• Ansys Fluent fluid simulation software: used to model liquid-based manufacturing processes, such as mixing.
• Ansys Rocky particle dynamics simulation software: used to model the particle dynamics inside fluids and often used with Fluent software for liquid-based subprocesses.
• The Ansys Additive Suite comprehensive additive manufacturing solution: used to model additively manufactured parts, including full virtual prints that look at the stress on the part, the layered structure of a part, the best orientation for a part relative to the laser, the presence of defects or inclusions in the part, the properties of the melt pool, and the optimization of operating parameters.
• The Ansys Twin Builder simulation-based digital twin platform and Ansys TwinAI AI-powered digital twin software: These are used to create a virtual digital twin environment that can simulate manufacturing processes and even a whole smart manufacturing facility to run test and optimization scenarios in real time.

Simulation is a big part of creating smart manufacturing facilities. If you’d like to discover how you can leverage simulation and digital twin technologies to improve smart manufacturing product design and operation, then get in touch with our technical team to find out more about our smart manufacturing solutions.

Related Resources