Four things to remember about DCS migration

Whether it’s a refinery, chemical plant, or other process-related facility, the primary control system must operate efficiently, safely, and economically. If it doesn’t, perhaps it’s time for an upgrade.

Jack Smith
Oil refineries and other process plants rely on control systems that must operate at peak performance. Courtesy: Emerson

Oil refineries and other process plants rely on control systems that must operate at peak performance. Courtesy: Emerson

Oil refineries and other process plants rely on control systems that must operate at peak performance. Courtesy: EmersonPlant and operations managers don't wake up one day and realize their distributed control systems (DCSs) and programmable logic controllers (PLCs) may be approaching obsolescence and a plan of action should be formulated. Unfortunately, the realization that a process plant's control platform is becoming ineffective typically comes too late, speaking from practical and economical perspectives.

When that realization comes, it's time to do some type of control system upgrade. Whether piece by piece replacement over time or a full rip-and-replace migration is in order, there are four things to remember about DCS migration:

The necessity of migration

John Rudolph, vice president of Lifecycle Solutions and Services at Honeywell Process Solutions, wrote in an article that appeared in the July 2016 issue of Control Engineering titled "Surviving a control system migration project":

"In its 2015 report, 'Distributed Control Systems Worldwide Outlook,' the ARC Advisory Group estimated that $65 billion worth of installed process automation systems in the world today are nearing the end of their useful lifecycles, which, in many cases, can exceed 25 years. Many of these systems—as much as $12 billion worth—are some of the original distributed control systems (DCSs) installed in the late 1970s. Ironically, many manufacturers treat their business systems and email servers very differently than their process control systems. Companies make a concerted effort to keep IT infrastructure current. The same level of emphasis is not yet common practice for plant automation."

Even though a legacy control system may still work well after 30 years or even longer doesn't mean that it is operating efficiently, reliably, safely, or cost-effectively.

Jeff Morton, a sales manager at Cross Company Integrated Systems Group, wrote in an article that appeared in the January 2017 issue of Control Engineering titled "Six action items for an aging DCS/PLC": "Budgets are always a constraint and capital expenditures may be tight, especially for full distributed control system (DCS) or programmable logic controller (PLC) replacements, which are often multi-million dollar investments.

"That investment, though, is nothing compared to the cost if the production facility went down and could not recover for a week or more. What's even more concerning than that is there are many facilities that are unprepared for system failure or obsolescence, which can have catastrophic consequences. Operations, facility, or engineering managers need to begin preparing for migration of your control system from a legacy system to current and supported architectures." 

DCS migration justification

According to Rudolph, when it comes to keeping automation technology up to date, proactive is the new normal. Companies that migrate to a newer, more effective control system gain a key advantage over competitors that wait for assets to reach end of life. The "doing nothing" option is no longer viable.

In a Control Engineering article titled "Distributed control system upgrades for process control systems," Aneel Shahzad Baig, senior project manager at Intech Process Automation, wrote that reasons to upgrade a DCS include:

Emerson's Laurie Ben, director of global modernization business development, and Aaron Crews, director of global modernization solutions, offer additional reasons for a control system modernization project. Their article, "Using automation modernization for business success," appeared in the April 2017 issue of AppliedAutomation.

Plan the migration path

Although the initial decision to upgrade and justifying the migration project are necessary, the actual project planning is probably the most point to consider so far. There are so many decisions to which the classic engineering answer is applicable: "It depends."

Rudolph suggests the following upgrade possibilities for a legacy DCS: Upgrade possibilities for a legacy DCS include:

Typical migration alternatives, according to Rudolph, can include:

Ben and Crews suggest three behaviors for achieving success in a DCS modernization:

  1. Begin with the end in mind
  2. Actively manage project risk
  3. Use a forward-engineering philosophy.



When planning any kind of DCS migration, one must not leave out the training aspect. "A DCS migration, particularly one in which a new platform is introduced, requires much thought and can unleash a variety of problems if not executed well. Training is a significant issue and should be approached carefully," wrote Peter Welander, contributing content specialist for Control Engineering, in an article titled "DCS migrations: Opportunity for improvement, or operational disaster" in the July 2016 issue of the magazine. 

Implement the migration plan

There are many things to consider when faced with an aging DCS, but they can be summed up into these four things to remember: necessity, justification, plan, and implementation. By the same token, there are many challenges to consider as well.

"If properly planned and implemented, control system migrations enable industrial organizations to migrate legacy control platforms at their own pace, allowing new controllers to be added at any time and integrated with existing equipment," Rudolph said. "A well-executed strategy to address technology obsolescence delivers significant operational and business benefits through seamless integration of new and existing automation assets." 

Jack Smith is content manager, CFE Media, Control Engineering, jsmith(at)


Key concepts

Consider this

How old is the distributed control system (DCS) running your critical process? If it is approaching 30 years old, do you have a migration plan in place?

ONLINE extra

See related articles below, offering more information distributed control system (DCS) migration.

Related News



Visit Our Sites

Contact Us


Close Home
click me