A software update can close a security gap and still disrupt a production line. Compatibility testing, planned downtime, working backups, and a recovery plan help engineering professionals manage that risk. 

Consider a hypothetical packaging plant installing a security update on its control computer. The installation finishes, but the operator’s screen can no longer exchange information with the controller running the conveyor. The team keeps the line stopped while engineering professionals investigate. 

The update has become a production problem. Preventing that situation starts well before anyone clicks “install.” 

Why a Factory Software Update Needs Extra Care 

Factories use operational technology (OT): equipment and software that monitor or control physical processes. A programmable logic controller (PLC), for example, is an industrial computer that can control machinery such as a conveyor. 

A change to this software can affect how equipment operates. In its Guide to Operational Technology Security, the US National Institute of Standards and Technology (NIST) recommends testing updates where possible, scheduling them during maintenance windows, and preparing a recovery plan. These precautions help protect production and safety. 

Check That the Equipment Still Works Together 

Compatibility testing checks whether the updated software still works with the other parts of the system. That means checking the actual equipment models and software versions used at the plant. 

Manufacturer testing is a useful starting point. Rockwell Automation’s patch qualification guidance explains that it tests particular combinations of Microsoft operating systems and industrial software.

It cannot cover every combination and advises verification on a system outside production, or while the facility is inactive. 

For our packaging example, a separate test setup would need to show that the operator’s screen still communicates with the controller. The agreed checks should also cover relevant alarms and operating functions. An installation message saying “successful” cannot answer those questions. 

Leave Time for More Than the Installation 

A maintenance window is an agreed period when equipment can be taken out of normal service for work. It gives production and maintenance teams time to coordinate the interruption. 

The schedule needs to cover stopping the process safely, installing the update, restarting equipment, and checking that it works. It also needs room to recover if a problem appears.

This is a practical application of NIST’s approach to patching as preventive maintenance, with planned actions and timeframes. 

For example, an update expected to install in 20 minutes could still require a much longer window. Engineering professionals need estimates for the whole job, including recovery. Operators, technical staff, and any required vendor support also need to be available when the work happens. 

Make Sure the Backup Can Restore a Working System 

A factory backup may need much more than copies of documents. NIST’s OT Backup Quick Start Guide identifies items such as controller programs, equipment settings, operator-screen files, software licenses, and the tools needed to restore them. 

The backup should reflect the system’s current, working setup. Otherwise, restoring an old copy could bring back settings that no longer match the equipment. Copies also need protection from unauthorized changes or destruction, with separate storage locations considered in the backup strategy. 

Engineering professionals then need to test restoration on equipment outside production.

Can they load the files? Do they have compatible hardware and the right software? Does the restored system function correctly? 

For the packaging plant, finding a backup folder would be only the starting point. A useful test would show that the saved files can rebuild a setup in which the screen and controller work together.

Decide What Happens If the Update Fails 

A recovery plan sets out how the team will return equipment to service if the update causes trouble. Before work starts, the team should agree who makes the recovery decision, what failure would trigger it, and which checks must pass before production resumes. 

The recovery method also needs testing. Some updates cannot simply be uninstalled. The DHS recommended practice for control-system patch management, hosted by CISA, calls for testing patch removal and preparing for recovery when removal is unsuccessful. It also recommends measuring how long restoration actually takes. 

In our example, the plan could set a deadline for resolving the communication fault.

If that deadline passes, the designated engineering professional would begin the tested recovery procedure, leaving enough time to restore the system and check it before the planned restart. 

That decision should follow the agreed plan rather than depend on hurried troubleshooting as the next production shift approaches. 

Manage the Security Risk While Waiting 

Careful scheduling still needs to account for the danger of leaving a security weakness exposed. NIST advises a documented process for deciding when to patch and what temporary protections to use when patching is delayed. 

Depending on the system and threat, this could include restricting unnecessary remote access or adding network protections. Those changes also need assessment for their effect on operations. An urgent threat may require an earlier maintenance window. The decision should account for both the attack risk and the consequences of interrupting production. 

For engineering professionals and technicians interested in this area, the Engineering Institute of Technology (EIT) offers a Professional Certificate of Competency in Industrial Control Systems Cybersecurity, covering operational technology, industrial communication, threats, and incident response. 

References 

NIST. Guide to Operational Technology Security, SP 800-82 Rev. 3, September 2023. Sections 5.2.5.2 and 6.2.11. 

Rockwell Automation. Microsoft Patch Qualifications FAQs. Questions 2 and 9. 

NIST. Guide to Enterprise Patch Management Planning, SP 800-40 Rev. 4, April 2022. 

NIST. OT Backup Quick Start Guide, SP 1339, June 2026. 

DHS. Recommended Practice for Patch Management of Control Systems, December 2008. Hosted by CISA. Sections 2.1.4 and 2.1.6. 

This article was published November 18th, 2026 and the content is current as at the date of publication.

Back to news
Engineering Institute of Technology