• By Robert Glaser
  • System Integration
Industrial devices—from robots to field devices, controller programs to drive systems—all have their own programming languages and data formats. Software version-control systems can make maintaining these assets easier

By Robert Glaser

The Industrial Internet of Things (IIoT) is set to bring about radical changes to the work environments of today's industrial maintenance departments. IIoT smart devices come with their own software and intelligence that needs to be maintained. But how often have you-or should you-ask whether the software version running on a device is the same as the latest released version on the server? Only with the right kind of data management system-giving you more certainty, safety, and security-is it possible to answer this question.

One important aspect of data management is version control. Version control refers to a system in which the changes made to a file or to several files are logged over time. This helps to centrally manage upload, download, and compare processes.

The term "version control" originated from software engineering. Software developers use the term to meet their goals, to safeguard and optimize their work, and to make the process of developing software more flexible. Version control is slowly winning acceptance with industrial automation departments. This is partly due to increasing costs and the time constraints placed on maintenance departments. These departments have a critical mission to keep production environments running optimally-minimizing downtime and recovery time.

Maintenance departments are fully aware of the challenges they face. Changes made to devices can lead to errors, which in turn can lead to downtime or defective products. For this reason, it is necessary to monitor, track, safeguard, and compare changes made to projects and to device program logic. The need for these processes is further reinforced by the fact that the number of different processors, devices, data, and possible sources of error continue to grow.

The version control process enables an organization to monitor who changed what, where, when, and why, helping to more accurately pinpoint the source of error when any problems arise. When required, version control can also help to quickly find and restore a previous, error-free version.

A detailed comparison of the online version (running on the device) and the offline version (running on the server) makes it possible to display the results of a comparison graphically, so that personnel are not forced to fly blind when it comes to operating a device.

Version control versus backup

The terms "version control" and "backup" are not synonymous. They are two distinct tools that have the greatest benefit-with regard to organizing data with clarity and traceability-when both are used together.

Users often mix up the backup of a device with the backup of project files. In this discussion, we define "backup" as "extracting the software running on devices." "Version" means "saving project data in centralized data storage with user management."

On their own, neither centralized data backups nor version control are enough to ensure 100 percent certainty, safety, and security. It is only possible to reliably determine if the latest released version on the server is the same as the online version if regular automatic backups have been carried out. With regular backups, it is possible to compare the software version running on a device with the last version checked into the server. This, in turn, allows changes to be detected and analyzed.

Conversely, it does not make sense to automate the process of creating versions using the data taken from a backup. This is because not all backups are the same. Depending on the device type, a subsequent use of the backup in the respective engineering system is not reasonable, as often no symbols and comments are included. Sometimes a subsequent use in the engineering system is not possible at all, because the backup consists only of compiled data.

It is therefore imperative when choosing a data management system to consider the type of data backups being produced. When it comes to maintaining high-quality data management and enabling the greatest possible production capacity, restorable backups always need to be available.

A manufacturer-independent approach

In recent years, the need for version control has also become apparent to device manufacturers. Nowadays many manufacturers offer version-control solutions. However, they are normally only compatible with their own devices. This means that they can only be effectively used in plants that do not have a wide variety of devices. However, in today's world, a production facility comprised of only homogenous devices is becoming more the exception than the norm.

As the automated sector continues to grow, so too do the number of different devices. This growth goes hand in hand with the increasing complexity of production facilities, which contain myriad different robots, field devices, controller programs, and drive systems, all of which have their own program language and data formats.

A future-proof version-control system is both manufacturer independent and transparent. It is capable of working with a full range of automation devices and systems commonly used today and is not tied to any single manufacturer. As newer products become more established, action needs to be taken to ensure that the necessary comparators are made available to users. (A test comparator helps to automate the comparison between the actual and the expected result produced by the software.)

Comparators are essential when it comes to version control, as it is only possible to detect and display differences between versions with their help. Version-control comparators allow organizations to make text-based, graphical, and tabular version comparisons that are easily accessible and readable, and backed up by technical implementation.

A location-independent approach

With the globalization of manufacturing plants, it is no longer rare for production to be spread across many locations and time zones. Therefore, it is highly important that version-control systems are location independent. A comprehensive server allows backup data taken from different devices and locations to be synchronized and centrally managed. It also allows differences between versions to be detected quickly and easily.

A version-control system also must account for external factors and third-party users. In times of lean production and lean maintenance, it is imperative that a version-control system be capable of detecting, monitoring, comparing, and analyzing changes to devices made by external system integrators and original equipment manufacturers. A version-control system capable of coordinating external personnel according to a variety of requirements has an added benefit: the documentation-of-change reasons allow for 100 percent clarity and traceability.

Because a good version-control system can monitor, detect, compare, and analyze changes to devices made by anyone anywhere, it can support a move to smart maintenance. Smart maintenance combines machine learning with predictive analytics. This allows maintenance work to be transformed from corrective to predictive.

To predict maintenance needs, a system must be able to recognize indications of malfunctions or defects taken from a large quantity of different types of data. And that data must be accurate. If data is corrupted or is missing, the source of error needs to be quickly detected, and normal operations need to be restored as quickly as possible. With version-control software, maintenance staff can rely on the accuracy of the data, and then produce well-defined and clear strategies to predict when maintenance action is required.

Steps to a successful implementation

With a central server and an unlimited number of installed clients, some version-control solutions allow users to work offline and check in new versions when necessary. Automatic synchronization helps to prevent unauthorized access and automatically documents who changed what, where, when, and why.

The industrial plant's information technology (IT) departments should be involved in any implementation of device version-control software to help ensure everything is properly prepared upfront. Depending on the level of integration during installation-such as integrating only a few typical device types or installing a plantwide solution-implementation of a system can take as little as one day or as long as several weeks.

When implementing across multiple facilities, including global implementations, it is recommended to roll out the system in phases to make sure all is running properly and in sync.

Nonhomogenous production facilities require many different project planning tools and editors that require multiple shifts. This results in large production and maintenance departments. Good version-control systems can integrate tried and tested editors and project structures while supporting users with a menu-driven tutorial and automated backups. This results in a high degree of usability by a wide range of staff and should require minimal training.

Implementing a system for device software backup and version control will logically result in adjustments to the IT infrastructure. However, it is a worthwhile investment-not only due to the reduced costs that it brings about, but also because of the way it helps to reduce risks to production facilities. A secure and safeguarded production facility means fewer errors, reduced downtime, reliable delivery times, and satisfied customers and partners. These benefits make the investment in a version control and backup system truly worth it.

The version control and data backup functions of a data management system are a valuable aid when automating simple but time-consuming tasks that would have been carried out manually. It thus helps to free up maintenance staff for other important tasks. Regularly carrying out comparisons allows authorized changes to be monitored. It also allows unauthorized changes to be quickly detected, which in turn allows users to quickly find and restore a previous error-free version. Thus, although the IIoT is set to bring about radical changes to the work environments of today's maintenance departments, the aim of those departments to maintain a well-secured and safeguarded production facility will not change. Production processes will simply become more efficient.

Reader Feedback

We want to hear from you! Please send us your comments and questions about this topic to InTechmagazine@isa.org.

Like This Article?

Subscribe Now!

About The Authors

Robert Glaser is the CEO of AUVESY Inc. Before this role, he worked for AUVESY Germany for more than 10 years, since the company’s founding in 2007. He was most recently director of research and development for the company. Glaser holds a master’s degree in computer science.