Six benefits of OT/IT convergence, and how to tell which ones you will actually get

Most convergence business cases make the same six promises, and most of them can be defended. What the slides leave out is that each promise rests on something being true first. That missing line is where the projected benefit and the one the site actually sees drift apart.

This is part 5 of 9 of the “OT/IT convergence” series; it follows part 4, measuring maturity and continues with part 6, the organisational models.

The six benefits, and what each one depends on

The list comes from the Cigref working group on IT-OT convergence, which published its report in December 2019: compliance, profitability, digital continuity, security, business development and attractiveness. We take them in that order.

Regulatory compliance. Convergence supports data integrity, process control, traceability, pharmacovigilance and the fight against counterfeiting. In a regulated industry it is often the benefit that pays for the programme. It is rarely the biggest; it is simply the one nobody can decline. The catch is upstream. The data reaching the upper layers must have been collected in a way an inspector will accept, and moving untrustworthy data faster helps no one.

Profitability. Operating costs fall and resources are used better, but the savings are spread over hundreds of small operations: fewer trips to a cabinet to copy a file, fewer manual transcriptions, quicker diagnosis of a stopped line. They only show up in the accounts if someone is named to collect them. Without that owner the programme delivers the capability and never the saving.

Digital continuity. Better-managed industrial systems mean machines that are available more often and easier to reconfigure. This benefit depends more than any other on how the work is engineered. Central backups, standard workstation images and managed spares raise availability. A badly designed dependency on a central service lowers it, because the line can now stop for a reason that used to belong to someone else: an expired certificate, a domain controller being patched on a Sunday night.

Security. Convergence enlarges the attack surface by construction. It improves security only where visibility, segmentation and monitoring arrive with the new connections, not two budget cycles later.

Business development. Data treated as an asset can support new services. Someone outside engineering has to want it, though. Plants store years of process data that no commercial function has ever asked to see.

Attractiveness. This one is real and underrated. Engineers choose employers partly on whether the work is interesting, and a site with modern, integrated, documented systems recruits from a different pool than one still nursing HMIs on an unsupported operating system. Cigref also links it to working conditions and to the company’s environmental commitments.

What it looks like on a site

Three examples make this less abstract.

The first is real-time production data sent to IT systems such as a historian. It is the founding case: better analysis, better use of the data, better tuning of production. As covered earlier in this series, it is also the strongest data integrity argument for convergence, since every intermediary removed between the controller and the record is one less thing to prove.

The second is shared technology. Time synchronisation (NTP), directory services, backup, security tools, remote access, central databases: IT has mature answers for all of them. The benefit comes from adapting those answers to OT constraints before deployment. Rolling out the office version and discovering the constraints during the first shutdown costs more than it saves.

The third is common processes, incident management and technical support in particular. It is the cheapest of the three and often pays back the most. When IT and OT share one incident process, the first call goes to the right person. With two processes, the first hour of an outage is spent working out whose incident it is.

Notice the pattern. The technical benefits need engineering, the organisational ones need governance, and the benefits that look softest, common processes and attractiveness, often arrive first and cost least.

Writing the business case honestly

Attach the precondition to every benefit. A case that promises better availability without saying which central dependency it creates cannot be checked by anyone. Compare it with: “availability improves provided central backup comes with local restore and a tested fallback”. That sentence can be audited, and defended two years later when the finance director asks what happened to the number.

Next: the organisational and managerial models (Rondeau, Besson and Rowe), and what leadership has to look like for any of this to work.

References

• Cigref, Convergence IT-OT : un rapprochement fructueux des systèmes d’information et des systèmes industriels, December 2019: cigref.fr