Big Data Applied To The Industry 4.0 Concept
The term Big Data was created in 1997 to designate a broad volume of disordered and rapidly growing virtual information. Today, this expression refers to a set of macro data or big data from multiple sources produced in a recent period of time, and allows a reading of reality in real time, provided that such information is submitted to adequate treatment. The Industry 4.0 concept was used for the first time in 2012, in Germany, and designated a new phase of the manufacturing segment, where manufacturing is computerized. This is the fourth industrial revolution, with a wide system of advanced technologies, such as artificial intelligence, robotics, internet of things and cloud computing, which together change operational mechanics, creating business models and reducing costs.
According to a survey carried out by the Brazilian Agency for Industrial Development (ABDI), the cost reduction in Brazilian industry with the migration to the 4.0 concept will be approximately R$ 73 billion per year. Within this amount, the decrease with repairs can reach R$ 35 billion. The gains in production efficiency can be R$ 31 billion, and the other R$ 7 billion are related to the reduction of energy costs.
However, this migration takes some effort. Industry 4.0 is fed back by Big Data on a daily basis. We are talking about a lot of information. Google estimates that humanity creates 5 exabytes of data every 48 hours. It’s the same volume of knowledge created by all the peoples of the world from antiquity to the year 2003. Just to remind you, an exabyte is a thousand petabytes. And a petabyte is a thousand terabytes. The legendary library of Alexandria is a grain of sand in this immensity. But how to work with so much information?
Well, we won’t collect all the data produced around the world. It is necessary to delimit a cut within this virtual universe. Something that occurs naturally in daily operations. By the way, when we talk about Big Data we necessarily go through the five “Vs”, defined by Information Technology theorists:
- Volume, which translates into the large data capacity that such technology can store;
- Velocity, which represents how quickly the hardware and software can work in collecting and processing information;
- Veracity, which demonstrates the power of the algorithm to filter the correct data that will be used in the analysis;
- Variety, which refers to the various sources of the records, both internal and external;
- and Valor, which represents the ability that Big Data has to add value to the daily life of the corporate world.
So we can say that each industry will build a delimitation on the first V, being the volume of Big Data. What will harm the processing speed, what is not true, what does not add value, what does not concern the industry activity, (already connected in phase 4.0, of course) is automatically eliminated by the IT professional, responsible for the implementation of this operational routine in the company. Still, it’s a lot of information, from different sources.
According to the Oncase website, “ the main sources of data that support the operation of Big Data in Industry 4.0 are:
- Social Data : is the data collected from user interactions on social networks, searches on Google and actions with other digital channels of the company. It is mainly through them that it is possible to develop the customer journey map, capable of tracing the consumption patterns and behavioral profiles of the target audience;
- Enterprise Data : these are inputs made evaluatable by the company at all times, such as data on human, financial and productive resources, as well as other records. They are fundamental to align the company’s operational capacity with existing demands ”.
Practically, we can give the following example: imagine that a car factory is planning to paint a certain model black in greater quantity, but the sales department registers a large volume of orders in red. In addition, several netizens ask if the car is available in red. Plans are quickly modified before production in black starts. This all happened in minutes, avoiding the creation of unnecessary inventory or, at the very least, the cost of rework.
But how to deploy Big Data technology in an industry 4.0? The website A Voz da Indústria says that this process goes through 3 phases:
- “Phase 1 – Information Architecture — The first action is to define which are the most relevant data and what are the company’s objectives with the use of this information. With the definition of the data that must be collected in this planning, it is necessary to decide how and how they should be determined, from the production line to the installed sensor and the machines, in order to arrive at a well-structured database. Therefore, in this phase, the information architecture is created, the organization of the parts of the same system so that it is comprehensible. This action will avoid a “patchwork quilt” and create a data standard, with a single language to integrate information and its exchanges between all industry sectors. The actions of planning, architecting and defining where the company wants to go with the data will optimize the use of Big Data.internet of things (IoT) or open protocols for communication between machines (Modbus or OPCs) — tools that enhance the integration of equipment and, consequently, data. Currently, it is also possible to install magnetic or presence sensors, for example, in processes or equipment that do not have their own way of accounting for production. As a metalworking industry, where welding processes did not have mechanisms for counting the parts produced and magnetic sensors were used to automatically detect and count them. Another possibility is to use data from machines, which often already have some kind of counter. Among this information should be the KPIs (Key Performance Indicator)productivity, quality and production costs. With these issues well aligned, the company will have security and clarity about the data analysis that will be performed in the next phase.
- Phase 2 — Visualization of data dissemination — One of the challenges is to make the information reach the right people and at the right time for more assertive decision-making. Companies usually use some more traditional models such as big screens or dashboards to display this data, which may be associated with alarms or alerts for the team. Another point that must be defined is the time that the data will be consolidated. One possibility is to collect them in longer periods, and not just visualize them in real time, consolidating this data in weeks or months. In this way, it is possible to check productivity over a period of time to assess problems and propose solutions to optimize production. At this stage, it is important to disseminate data to those who will act in the company and to stop production when a critical defect is identified before causing losses with rework or the use of raw materials. Or determine a review of the entire production process, after analyzing the data set over a longer period and in line with industry KPIs.
- Phase 3 – Use of data to generate scenarios and decision-making — In this last phase, after correctly extracting the data that was clearly visualized and disseminated in the two previous phases, it is possible to carry out strategic simulations before intervening in production. The use of real data in simulations allows the implementation of the digital twin concept in the industry. These simulations can provide the best parameters and conditions for reliable and efficient production. Simulations can use technologies such as data analytics, business intelligence or even artificial intelligenceto analyze historical data, highlighting trends, patterns, or correlation between data. The main objective with the use of these innovations is to direct human capital to more strategic situations and put the machine in operational actions. Therefore, after these simulations, the data can be used to make more assertive and low-risk decisions, avoiding energy waste or the need for a new production process. From this moment on, the necessary actions will be established to increase productivity and product quality.”
According to the Portal da Indústria website, “the incorporation of Advanced Robotics, Machine-Machine Connection Systems, the Internet of Things and Sensors and Actuators used in production equipment makes it possible for machines to “talk” throughout industrial operations. This can allow the generation of information and the connection of the various stages of the value chain, from the development of new products, projects, production, to after-sales.” It is in this scenario that Big Data information will be used. All this technology still enables an average growth of 22% in productivity. Does it work? Yes. It is worth it? For sure.
Did you like the text? At Águia Sistemas, we are national leaders in Intralogistics, and we can help your company to always find the best solution in this field. Keep in touch!
(The text above was written with information from the websites, Oncase, A Voz da Indústria, Portal da Indústria).