smart manufacturing (SM)
Smart manufacturing (SM) is a technology-driven approach that utilizes Internet-connected machinery to monitor the production process. The goal of SM is to identify opportunities for automating operations and use data analytics to improve manufacturing performance.
SM is a specific application of the Industrial Internet of Things (IIoT). Deployments involve embedding sensors in manufacturing machines to collect data on their operational status and performance. In the past, that information typically was kept in local databases on individual devices and used only to assess the cause of equipment failures after they occurred.
Now, by analyzing the data streaming off an entire factory's worth of machines, or even across multiple facilities, manufacturing engineers and data analysts can look for signs that particular parts may fail, enabling preventive maintenance to avoid unplanned downtime on devices.
Manufacturers can also analyze trends in the data to try to spot steps in their processes where production slows down or is inefficient in their use of materials. In addition, data scientists and other analysts can use the data to run simulations of different processes in an effort to identify the most efficient ways of doing things.
As smart manufacturing becomes more common and more machines become networked through the Internet of Things, they will be better able to communicate with each other, potentially supporting greater levels of automation.
For example, SM systems might be able to automatically order more raw materials as the supplies, allocate other equipment to production jobs as needed to complete orders and prepare distribution networks once orders are completed.
A lack of standards and interoperability are the biggest challenges holding back greater adoption of smart manufacturing. Technical standards for sensor data have yet to be broadly adopted, which inhibits different kinds of machines from sharing data and communicating with each other effectively.
In the United States, the National Institute of Standards and Technology (NIST) is investigating opportunities to develop and promote standards with various industry stakeholders, including technology companies and manufacturers. The process is ongoing. Other challenges include the cost of implementing sensors broadly and the complexity of developing predictive models.
History/background
It's been nearly 260 years since the beginning of the original Industrial Revolution, thought to have started around 1760. In the United States, the latest iteration of this process, the fourth industrial revolution, has been called "smart manufacturing," while in Europe it's known as "Industry 4.0."
The first industrial revolution was characterized by steam power and the power loom; the assembly line was introduced during the second industrial revolution; and automation and data-enhanced automation came along in the 1970s during the third industrial revolution.
This fourth industrial revolution is characterized by a range of interconnected automated systems that are fusing the physical, digital and biological worlds.
Related technologies
In addition to the Internet of Things, there are a number of technologies that will help enable smart manufacturing, including:
- Artificial intelligence (AI)/machine learning – enables automatic decision-making based on the reams of data that manufacturing companies collect. AI/machine learning can analyze all this data and make intelligent decisions based on the inputted information.
- Drones and driverless vehicles – can increase productivity by reducing the number of workers needed to do rote tasks, such as moving vehicles across a facility.
- Blockchain – blockchain's benefits, including immutability, traceability and disintermediation, can provide a fast and efficient way to record and store data.
- Edge computing – edge computing helps manufacturers turn massive amounts of machine-generated data into actionable data to gain insights to improve decision-making. To accomplish this, it uses resources connected to a network, such as alarms or temperature sensors, enabling data analytics to happen at the data source.
- Predictive analytics – companies can analyze the use huge amounts of data they collect from all their data sources to anticipate problems and improve forecasting.
- Digital twins – companies can use digital twins to model their processes, networks and machines in a virtual environment, then use them to predict problems before they happen as well as boost efficiency and productivity.
Pros and cons of smart manufacturing
Smart manufacturing offers a number of benefits, including improved efficiency, increased productivity and long-term cost savings. In a smart factory, productivity is continuously enhanced. If a machine is slowing down production, for example, the data will highlight it, and the artificial intelligence systems will work to resolve the issue. These extremely adaptable systems enable greater flexibility.
In terms of efficiency, one of the main savings comes from the reduction in production downtime. Modern machines are often equipped with remote sensors and diagnostics to alert operators to problems as they happen. Predictive AI technology can highlight problems before they occur and take steps to mitigate the financial costs. A well-designed smart factory includes automation as well as human-machine collaboration, features that enable operational efficiency.
A big downside to smart manufacturing is the upfront cost of implementation. As such, many small to midsize companies won't be able to afford the considerable expense of the technology, particularly if they adopt a short-term philosophy.
However, since savings over the long term will outweigh the startup costs, organizations have to plan for the future even if they can't implement smart factories immediately.
Another disadvantage is that the technology is very complex, which means that systems that are poorly designed or not adequate for a particular operation could cut into profits.
How SM differs from traditional manufacturing approaches
Traditional manufacturing methods, developed during the age of mass production, focus on economy of scale and machine utilization. The thinking was that if a machine was idle, it was losing money, so companies kept them running continuously.
To achieve customer satisfaction, traditional manufacturing companies keep large inventories on hand so they can fulfill potential orders. Consequently, these companies have to keep their machines running with specific setups for as long as they can to reduce the costs of making the parts.
This is known as batch-and-queue processing – a mass production approach to operations where the parts are processed and moved to the next process, whether they're needed or not, and wait in a line (queue).
However, this approach isn't very efficient for several reasons, including:
- A longer machine set-up time means more lost production time, because nothing is produced while a machine is down.
- The quality of the product suffers because if parts in a batch aren't made correctly, no one will likely notice the problem until the next operation. This means the work has to be done again, which is expensive and ties up valuable resources.
Smart manufacturing, on the other hand, is a collaborative, fully-integrated manufacturing system that responds in real-time to meet changing the conditions and demands in the factory, in the supply network, and in the needs of the customers.
The goal of smart manufacturing is to optimize the manufacturing process using a technology-driven approach that utilizes Internet-connected machinery to monitor the production process. Smart manufacturing enables organizations to identify opportunities for automating operations and use data analytics to improve manufacturing performance.
