Data-driven production optimization

FitMech [EN]
5 min readMar 2, 2021

--

What is data-driven manufacturing? What benefits can you achieve thanks to it? What challenges can you face? We invite you to the world of data-driven manufacturing.

The results of the report “Insights-Driven Businesses Set The Pace For Global Growth”, prepared by Forrester experts, show that customer-oriented companies that systematically use data throughout their organization and implement it in order to build a competitive advantage using software, grow by an average of over 30 percent per year.

At the same time, according to the McKinsey Global Institute, data-driven companies are 23 times more likely to attract new customers, six times more likely to retain them later, and 19 times more likely to make a profit as a result. So there is something to fight for!

Data-driven manufacturing in industry

Data-driven manufacturing is an approach to production that assumes making decisions based on facts, not guesses or gut feelings. It uses data from monitored machines, from operators, obtained in the supply chain and from other analytical sources, on the basis of which solutions are implemented that translate into lower costs or increased operational efficiency.

Thanks to new technologies, an increasing number of manufacturers are able to collect and process data about their activities, leading to a sharp increase in the use of production analysis. However, not all analyzes are the same. To use operational data efficiently, accurate information must first be collected, which can later be efficiently processed and presented in a way that is easy for the user to use.

In this case, the method of collecting data is of key importance. Collecting them manually requires not only a lot of time, but is also associated with a high probability of making a mistake. On the other hand, collecting data directly from machines via a PLC produces a very accurate and objective data collection. Both solutions in this case can be considered “data-based”, but the latter is much more effective, and thanks to the speed of their acquisition and their greater accuracy, it allows you to make better decisions.

The greatest benefits

The development of production based on data brings a number of benefits to industrial enterprises, such as:

  • Greater visibility: From the data gathered throughout the organization, both shop floor and office specialists can better determine the performance of the enterprise. The data provides them with insight into not only individual results, but also operations as a whole. Such possibilities help decision-makers to identify weak points, such as: worse working shifts, repeated machine downtime or other production bottlenecks.
  • Artificial Intelligence and Machine Learning: With large datasets, manufacturers are able to run machine learning algorithms to help solve complex problems and apply advanced practices such as predictive maintenance. Without access to data and the possibility of their processing, even simplified forms of decision-making based on data would not be possible.
  • Automation: There are two categories of automation that can be used in data-driven strategies. The first is automatic data collection. If the enterprise is equipped with the appropriate data collection devices and software to process it, the process no longer requires manual work. The second element is the use of data for automatic decision making. As manufacturers develop, their analytical pathway moves from descriptive to predictive analysis. This means that they first use the data to understand what has happened or is happening at the moment, but eventually mature to understand what might be, so they can react well in advance.
  • Savings: Data also provides manufacturers with the information they need to improve production processes and minimize waste. Without hard data, it is difficult to accurately measure the improvement in production and be sure that the changes brought real savings for the company.

Possible challenges

The Internet of Things (IoT) and its application to big data and analytics have led to a new generation of manufacturing. It includes using data to lower costs through sales and operations planning, dramatically increased productivity, optimization of the supply and distribution chain, and new types of after-sales services. However, this presents many challenges that companies must face in order to drive data-driven production. These include, among others:

  • integration with other systems — many systems and devices have problems with effective integration and communication with each other. They can result from separate operating systems in different departments or from older hardware. Therefore, aggregating data in different systems can be a big challenge and reduce the value of the collected data. This problem is helped by the solutions of the IoT platform, which connects many levels of systems and makes old equipment available online.
  • security threats — the IIoT security problem arose for two main reasons. First, the more devices are connected to the network, the greater the chance of creating security holes. Second, security at the computer level was not dealt with before, which means that no standards or protocols have been developed, so manufacturing companies should also invest in security solutions to mitigate this risk.
  • data storage — the more devices and systems are connected in a production plant, the more data will be collected. Seemingly, this is a big advantage, as the user can gain a wide insight into the collected data. However, their dynamic growth also requires space for their collection and processing, which can be a large and costly challenge, especially if the user has to store data locally, which speaks for the use of cloud computing.
  • Inaccurate or incomplete data — Incomplete production data can have a huge impact on situational awareness and, consequently, decision making, especially in critical projects where data is the basis for success. It also means a great deal of work, a waste of time and effort to complete incomplete records or to make sure that they are authentic and correct.

Use the potential of your data

To reap the benefits of optimizing your production processes, you need to create a data-driven strategy that should involve three basic steps:

1. Data Capture and Monitoring: Real-time data collection is the basis for efficient data use. This is made possible by FitMech sensors mounted on machines that measure vibrations and accelerations generated during work, providing information on efficiency, work cycles, downtime, occupancy, efficiency and downtime in real time.

2. Data categorization and visualization: the collected data must be processed and analyzed. The data from FitMech sensors is then sent to the computing cloud, which processes it, identifying machine status, work cycles and downtime. Then, in the administrator panel, the causes of irregularities are marked, and the online panel presents the results of analysis of data from sensors in the form of live view and reports.

3. Analysis and business activities: the collected data is presented in the form of transparent reports, automatically adjusted to the needs of various production monitoring models. This makes it easier to make decisions and create strategies. After collecting and analyzing machine data, it becomes possible to develop practical strategies to optimize production efficiency and then implement them.

--

--