Maintenance 4.0: challenges of the Brazilian industry 

Are you familiar with the concept of Industry 4.0? And Maintenance 4.0? Have you heard about it?  

In this post, we will point out the opportunities and challenges of implementing Maintenance 4.0 from the perspective of the national industry. But, before we start, let’s better understand some concepts that are related to the Industry 4.0 journey. 

Thinking about the maturity of technology 4.0 in companies over time, the following steps can be highlighted as primary and subsequent items:   

• Data Digitalization; 
• Information Visualization;  
• Prediction; 
• Prescription. 

When we understand that these steps are needed to achieve Industry 4.0 maturity, it is possible to point out the challenges and opportunities particular to each of them. 

Data Digitalization 

The first step of digitalization consists in bringing to the digital systems perspective the maximum possible data referring to the productive process and its assets. Among which we can exemplify: process parameters (machinery setup), anomaly indicators (asset health), asset availability indicators, process quality control, among others. 

To enable the entry into this context of digital transformation of the company, it is essential to apply good monitoring and data collection system(s), using, for example, digital sensors that collect data from the process and from the operation of the assets, while at the same time considering the digitalization of indispensable data coming from the observation/analysis of human beings, such as the anomalies found in a sensitive inspection on the factory floor. 

Some challenges will certainly be present in this perspective of going digital:   

Organizational culture/resistance: we have already commented, in texts here on the blog, how resistant people and teams can be to new technologies and processes. In many places, the culture of paper and clipboard, or manual data collection, still reigns. New technologies will always face initial barriers, and organizational culture can be a decelerator of digitalization if not treated with attention. 

Qualified people: without people, there is no transformation. Bringing in experienced people and investing in in-company training can be ways to ensure that the transformation process is innovative and at the same time well-structured.   

Not knowing where and how to start: there is a plethora of technologies and options on the market. Gathering information about the main internal processes and identifying the most critical ones, the ones that demand more work or that can somehow bring the greatest results with its digitalization, can be a path. Prioritization is the key word here. 

Signal availability/connectivity: when it comes to automating the collection of relevant metrics for process health monitoring, connectivity in more remote regions can also be a challenge when deploying digitalization programs. It is important to involve the IT team from the beginning to ensure that there are no surprises in this context. 

Having overcome the challenges of entering this stage of digitalization, we enter the next stage of Industry 4.0 maturity, which is information visualization. 

Information visualization  

The information visualization stage is responsible for supporting the interpretation of the asset/process scenario, starting from “what happened” (data) to “what is happening” (information). 

In this step, you must work with the data obtained so that it is presented as clearly and easily as possible, to transform it into useful information for whoever or whatever is consuming it. 

The main challenge at this stage is data decentralization. It is common for companies to have several uncoupled data acquisition systems with individual and uncorrelated analyses, resulting in views centered on the data acquisition tool/method and not centered on the asset/process as it should be.   

Opportunities to circumvent this challenge are usually based on integrations between systems and monitoring disciplines, putting the asset/process as the center and, orbiting around it, monitoring tools that present the data through graphs, dashboards, and customized reports as clearly and intuitively as possible.    

In short, here one must work to make sense of the data, transforming it into centralized and pertinent information.  

Prediction 

Knowing “what has happened” and “what is happening” (Data Digitalization and Information Visualization, respectively), one can move on to “What will happen?”. It is at this third level of Industry 4.0 maturity that the concept of prediction appears. 

By applying good analytical models to the information collected, behaviors can be mapped and understood with respect to the functioning of the asset/process, thus, one can begin to predict what will happen to the asset. 

Most of the challenges for the prediction step are found in the lack of correlation between the applied analyses. At this stage, even with centralized data, there is a risk that the analysis does not consider the correlated fronts, resulting in an individual behavior compatible only with the input data of the analysis. Thus, each measurement procedure/tool generates a prediction of individual behavior as if they were independent of each other. However, it is known that there are numerous interdependencies between the processes/components and the correlation between the analyses is fundamental. 

In this perspective of problems in connecting these analyses, one can list as an “antidote” the application of technological/innovative concepts such as Digital Twins. This concept mirrors the reality of the equipment to a virtual environment, such that the real machine and the “digital twin” are in constant communication. Therefore, by simulating the asset’s behavior in a more complete and correlated way, one can predict with more assertiveness the condition of the assets/processes and, consequently, correspond to the prediction step. 

At this stage, it is also important to team up with quality partners who have the technical know-how to develop prediction models together. Get away from miracle solutions and generalist methods that promise to work for all types of assets and processes. 

Prescription 

Understanding that there is no point in predicting the behavior of an asset/process without taking action to prevent and/or mitigate the negative effects of this behavior, we have reached the fourth stage of maturity 4.0, the prescription stage. 

In the prescription stage the goal is to determine what action/reaction should be taken in response to an encountered/foreseen problem. This is where the true value of 4.0 applications lies, because the effects of failure and process deviations will be mitigated, and this can extend the life of assets and improve production processes. 

As a challenge to this step comes the need to know the particularities of the assets and processes. This knowledge is fundamental both for the creation of logical rules that deliver an action plan based on predicted behaviors, and for the validation of ML models that promise to generate reports and action plans about the conditions. 

It is important at this stage, even to overcome the challenges mentioned, to consider the opinion and experience of key people with years of experience in the area/process. These people will be relevant in the validation of automated detections, as well as in the assertiveness of the prescriptions made. The process of adjusting detection models according to the opinion and experience of these key people will ensure that the knowledge is retained in the organization, thus reducing any dependence on personnel. 

Finally, it is understood that the challenges and opportunities of Industry 4.0 and Maintenance 4.0 vary according to the moment/maturity of each company within this context.   

Dynamox has a complete solution that can help in all the stages mentioned throughout this text, from the digitalization process (with sensors and gateways for automated data collection + app for sensitive inspection registration), to information visualization (with web platform and data analysis dashboards), to the prediction and prescription stages (with automated detection application in assets, as well as predictability tools and the possibility of recommended actions).  

Did you enjoy this article? Keep browsing the blog and read “7 benefits of the Dynamox Solution in maintenance”.