The digitalization of production processes

blog-en
The digitalization of production processes

Read the article in Spanish on Digital Biz.

With today’s cutting-edge machines and robots, it’s easy to assume that production processes have already been digitalized. After all, machines appear to do most of our manufacturing and distribution work. But these machines only form a small part of these processes; there are many other components that can be improved by implementing new technologies. And in an ideal world, improving means reducing costs and boosting quality at the same time.

Digitalization through sensorization, communication and data processing technologies allows us to be more competitive while cutting costs and enhancing quality.

Cost reduction can be achieved by optimizing each step on the production chain. It’s difficult to make generalizations, as different businesses—from a construction materials factory to a cleaning service—have very different processes and tasks. However, while production chains do differ, we can still speak generally about them in order to better understand the concept of digitalization.

The first step is understanding what’s going on at any given moment. Optimizing processes requires having the necessary information to make decisions that allow for small variations—or drastic ones, as the case may be. This information is found in each and every step, whether because it involves a machine running at a certain number of revolutions per minute, or because a person begins a particular activity at a certain time.

Sensorization

This is why collecting data through sensors, known as sensorization, is the first step toward digitalization. Different sensors allow us to follow what is happening at every step of production process with distinct points of measurement and parameters, such as temperature, revolutions, presence control, pulse counts —essentially every situation and every point of measurement. Of course, there are different types of sensors for different machines, and different means of collecting data for different situations.

Once the data has been collected, we must be able to put it to use, whether on site, in a centralized location or through a combination of the two. On site means that in addition to installing a sensor, an information processor is also installed that is capable of making some type of decision. In contrast, to submit the information to a site for centralized treatment, a communication system must be installed in addition to the sensor.

Communications have evolved a lot in the past 20 years. With the advent of mobile communications, especially cellular data networks from GPRS to 5G, sending data from a sensor to a treatment center is becoming increasingly faster and more affordable. In addition, we’ve seen the beginning of short-range networks, communication systems with low battery consumption, open spectrum networks, etc. Undoubtedly, data communication can be performed from virtually anywhere, without the need for expensive wiring and with a significant reduction in the price of devices. All of these developments help digitalize businesses.

Working with the information of data

The treatment of data, whether on-site or centralized, allows us to carry out three basic functions: performing analyses, setting alarms and taking action. So far, so good. For example, if there’s a drop in the oil level, I can set off an alarm so that it’s replaced. But to perform a centralized treatment, we can make more complex adjustments that combine the information available from all of the points from which we’re receiving data. We can also look at historical records and trends.

For example, if I am aware that the oil level drops a certain percentage every day, I can then calculate how much time remains before the drop becomes critical. If I confirm that there are no other programmed activities that require further oil consumption, I can then decide whether to change the oil sooner or later.

In this example, we’ve reduced production costs because the need for replacing the oil is now less frequent, allowing maintenance staff to devote additional time to more urgent activities. It also allows us to order larger oil reserves from the supplier, at a greater discount. Of course, this is just one example, and we can imagine far more complex combinations.

Data Quality

It’s important to note that in this example, we’ve incorporated two types of data. We’ve not only sensorized**, but also used data from future planning, employee activities, provider prices, and stock level.** We’ve integrated the entire production process. We’ve digitalized it and made it smarter in order to reduce costs automatically.

Of course, the quality of the product remains the same, as we haven’t altered the components or changed providers. It is, however, still possible to go further in terms of digitalization and improve the quality of the products or services offered. To do so we’ll have to digitalize the quality control department, which is responsible for ensuring compliance with certain quality standards. That could involve improving durability in the case of a product or customer satisfaction in the case of a service.

Digitalizing quality control is achieved by incorporating more data into the analysis we carried out previously. We’ll both automate the tests we need and provide feedback to the production system in order to correct deviations. The automation of tests consists of using systems that, for each product or service (or randomly for a percentage of them), carry out tests that will, in turn, evolve as they obtain more or less data. These tests provide feedback for decision-making and can lead to insights for modifying production speeds, predicting duration, or informing the customer through the publication of useful information, among many other possibilities.

Some conclusions stem from inferences made by employees, who are undoubtedly another source of data for the system. Clearly, this will depend largely on the process, but in any case, it will be included in integration and digitalization efforts.

Ultimately, digitalizing production processes means having access to the largest amount of information possible, as frequently as possible, as well as communicating this data, analyzing it and making immediate decisions.