An article to understand the smart factory (smart factory/Smart Factory) - what is a smart factory?

The next generation of smart factories will use cyber-physical systems for data collection and analysis to perform tasks more efficiently and create more valuable products.

With the intelligent development of mobile phones, automobiles and water, the intelligentization of industry has become within reach. You may have heard the term smart factory, or the Industrial Internet of Things (IIoT). It is not difficult to infer that this new generation of factories will introduce new technologies and do better in all aspects. So how do they accomplish this goal?

A cyber-physical system (CPS) is the combination of connectivity and tools that make a smart factory "smart". By collecting and analyzing data, they can perform tasks more efficiently. This information can be used to create better products and more efficient technologies.

urgent data needs

The first step to making factories smart is centralizing data. Any successful business should pay close attention to the numbers. In smart factories, there are corresponding systems for data collection and centralization. Mostly, this is a network of wireless IIoT sensors and devices that are constantly collecting and storing vast amounts of data. This data can be anything from the timing of a specific robot to environmental conditions across a factory.

It is impossible to have a smart factory without a good set of aggregated data and the right tools for maintenance. However, having data is only the most basic component of a smart factory.

level one

In fact, the smart factory model treats available data as the first of four levels. At this level, data is captured, but in a way that is difficult to analyze holistically. Separate processes or systems keep the data they collect, and you can't really get a bird's-eye view of a plant's processes.

level two

The second level is accessible data, which means the data has been pulled from silos or completely different corners of the company. This data goes into some kind of central reporting structure that employees can use to make informed decisions. At this point, the type of data collected may be standardized to facilitate analysis.

make informed decisions

However, having a lot of data is not enough, you need to know what to do with it. This is where the smart factory model starts to shine. As more data is aggregated, it becomes possible to create models of processes and sub-processes to inform the overall task of the factory. Network-aware CPSes use the Internet of Things (IoT) to communicate what they are doing and the results of their tasks. As a result, the plant itself becomes aware of its own successes and failures, which are defined by those in management roles using key performance indicators (KPIs).

Level three

At the third level of the four-tier smart factory model, smart factory managers implement advanced data analysis techniques powered by artificial intelligence (AI) and big data analytics. These techniques can sift through vast amounts of data—even more than a team of human statisticians can reasonably analyze—and perform pattern detection and predictive model creation.

Factory managers and production line employees can use these insights to improve factory processes and make more informed decisions.

grade four

At the highest level of the smart factory — layer four — the factory’s use of the data itself has been automated. Factory processes can effectively identify what went wrong and come up with new ideas to produce better results—sometimes without any human intervention.

Machine Learning and Execution

When something goes wrong, operators are self-aware enough to stop production or make changes. For example, in the automated component inspection process used by major global manufacturers, defective parts are immediately discarded rather than submitted to paid QA testers. This saves a lot of time and money.

Part inspection works by using a model that tells CPSes exactly what a finished part should look like, measure and perform. Robotic CPSes can then test the workpiece and make a decision while recording the inspection results. These results are cataloged and possibly even added to the dataset of "bad" examples, improving part inspection algorithms.

Taking the automobile factory as an example, further explain the working method of the fourth-level smart factory. Imagine an operation that maximizes the number of cars produced each day: A factory's CPSes require a repair every 100 cars, which requires you to take some or all of the factory's vehicles offline for processing. Smart factories have a feature that notifies administrators when maintenance is required, and even the maintenance process itself.

This operation uses a lot of electricity, but it allows plant managers to save on electricity costs by leveraging the capabilities of a smart factory. The factory is context-aware, which means that by defining the cost of electricity usage as a KPI, you can configure it to use electricity most efficiently. You can run maintenance later in the day, off-peak, or slow down production to reduce workload when electricity is expensive.

Thanks to the interconnectedness of smart factory systems, factories can make recommendations to managers and even take actions themselves.

Does this mean that factory workers can say goodbye to their jobs? Rather, and more likely, jobs will be created in other sectors as well, so there may be a shift in the direction of these efforts, which means the number of jobs may not be the same. will decrease.

Author: Megan Nichols

This article is from a translation, if you want to reprint, please obtain authorization from this site first.