Back to blog

The Central Theme for the Future of Mechanical Engineering

August 23, 2019

Author: Andreas Nigg is CTO, Product Owner & Data Scientist at the IIoT provider Senseforce and lecturer of signal theory and systems theory at the Voralberg University of Applied Sciences.

IIoT builds a bridge to transparency and data-driven decisions. Even SMEs can get on board the data train - as long as they’re able to initiate the right processes.

IIoT has the potential to permanently revolutionize mechanical engineering. The leap into the world of data must be done in all industries at once in order to avoid losing market. While large companies are pushing ahead with developments in this area and are already enjoying the profits, medium-sized companies are faced with this question: How can the journey from physical product to IoT integration be efficiently started and brought to completion? Only a short time ago there were uncertainties about the potential challenges of new technologies, but now standards for machine data digitization are taking shape.

Ethernet-Based Protocols as a Foundation

European SMEs are perfectly suited for IIoT scenarios. They enjoy a high level of automation and, correspondingly, a large existing data set. In addition, up to 98 percent of existing equipment manufacturers area already using Ethernet-based communications protocols to connect to their PLCs. While Profinet continues to dominate the landscape for historical reasons, OPC UA has established itself as the de facto standard for IIoT in Europe, with increasing importance in Asia and the Americas. It can be assumed that the majority of next-generation machines will rely on OPC UA, which will lead to a unifying standard for data acquisition technology. The data are usually stored in a centralized cloud platform in order to meet the high requirements for storage technology and computing power needed for analysis. It is appropriate to install an edge gateway between the OPC UA server and the cloud that collects the data from the OPC UA, cleans it up if necessary, and securely transfers it to the cloud. Here, the best choice for the transmission protocol is MQTT due to its low implementation complexity and its extremely bandwidth-efficient message architecture. AMQP is also considered to be a similarly capable alternative, offered by all larger IIoT providers as well as Senseforce.

Relevant Tasks for Edge Components

In addition to simply recording and transferring the data, the edge component is also able to:

  • Ensure data normalization: In order to achieve the best possible data analysis and ensure the maximum value of the product, high-quality and thus normalized data is necessary across the entire machine fleet. Feature extraction, i.e., the extraction of relevant information already at edge level, pre-aggregations, and even simple unit conversions at the edge level have proven to be successful measures.

  • Ensure message integrity: Network quality in many machine locations is suboptimal. Connection breakdowns and narrow bandwidths are still a reality. The edge gateway must therefore be able to locally and securely buffer several weeks’ worth of data to be transmitted.

  • Ensure security: Reports of millions of unsecured Internet-connected IoT devices have renewed the debate about security in the IIoT field. There must exist a robust security protocol from the machine itself to the cloud platform user in order to prevent against attacks. State-of-the-art encryption, authentication, and authorization as well as methods to identify and close security holes are only part of the measures to be implemented. Complete solutions like Senseforce work with expertly tested solutions to ensure the highest security standards.

Data Analysis and Management Architectures

In many cases, however, it is no longer necessary to be concerned with the question of which data is collected. Because of the high level of automation and regulation in modern machines, a wealth of data is already available. The use of environmental data such as pressures, temperatures, and wind speeds as well as machine-specific data such as forces, currents, voltages, and axis speeds now allows for the implementation of various use cases, from engineering to service, purchases, and sales. A comprehensive data analysis and management solution is necessary for each scenario. Such a solution must also be able to handle the correspondingly large amount of IIoT data, requiring scalable technology and software architectures. Senseforce uses a microservice architecture to flexibly add or remove analysis and management functionality as needed. A data abstraction layer connects the IIoT-optimized, highly-scalable, cluster-ready database to the microservice application. Extensive user management and group/role concepts integrated directly into the platform allow for internal rollout to various departments as well as end user integration and, therefore, the development of new business models. The feature set includes a graphical formula editor, R and Pything scripting engines, statistics and machine learning tools, graphical data visualization tools, reports, and rule-based notifications.

Human Machine Interface Reevaluation

Aside from the clean and high-quality implementation of the IIoT application, the platform user must always be kept in focus. Although the topics of security, stability, and performance are approaching initial solutions, it is becoming increasingly clear that IIoT can only be successful when it is actually adopted and used by employees throughout the company. Technologies like Senseforce therefore rely on a modern user interface and high abstraction of technical details in order to achieve the highest possible number of end users. The most important element in IIoT will not be the technical details. It will be the employee, as creator and recipient of data-based added value.

The compact, modular IO system




Cube is the unequaled modular fieldbus system for perfect decentralized installation concepts. Thanks to its flexibility, the best solution can be found for every application - from IP20 to IP69K.

Cube solutions are the most cost-effective in every phase of a machine’s life cycle.