Why have automation projects sometimes failed to deliver their expected return?
Over the years, a relatively high percentage of the capital investment made in process automation technology has failed to deliver the return that was promised when management chose to invest in it. This has, in some cases, given process automation a “bad name” and discouraged owners and managers from making further expenditures.
I believe that investment in process automation not only makes business sense, but is essential in enabling an industrial site be competitive in the world marketplace.
Examining automation projects that have delivered less than stellar financial results, it seems that there are a set of common mistakes made in their planning and implementation. My list is as follows:
- Controls installed without considering process and/or mechanical issues
- Control implemented on poor measurement and/or end device foundation
- “Panel Board” control strategy cut over to computerized system
- Automation strategies not robust to allow for process upsets or measurement interruption
- “Advanced Control” dependent on special, unreliable, or hard to maintain measurements
- Proprietary hardware technologies and/or “black box” control implementations used
- Wrong control tool used for an application
- No “Owner” of automation technology on each process in plant/mill
- Poor or no documentation of installed control and automation systems
- Inadequate training of operations personnel
- Control or automation technology implemented without operations SOP changes
- Training of maintenance personnel in use of new tools not completed
- Maintenance SOPs not addressed and changed
I believe that each of the above issues can be mitigated or eliminated from automation projects through proper planning and implementation, such that the capital investment made delivers significant return to the business.
Comments, other issues, thoughts on how to ensure financial success?
- Login or register to post comments
- 3175 reads
- Permalink
- 2 comments
Great Analysis
This is a really good summary of some of the reasons measurement and automation applications fail to achieve the desired ROI. I think I have seen most of these at customer accounts I have visited over the past 11 years. And George Buckbee's comments are right on target too.
I might summarize your list with this theme: a lack of effective collaboration between management and engineering. A successful process automation (or measurement) project needs to achieve the goals of both stakeholders. Too often a management goal (like cost reduction) sabotages the engineering goals via underinvestment in technology or skilled integration services. Similarly, an engineer running open-loop might create a fantastic application that falls short of management expectations becuase it either doesn't accomplish management's stated objectives. Tragically these systems sometimes fail because no one thinks to implement a metric to assess the project's effectiveness in accomplishing management's objectives.
Technology investments need to be made based on business justification - not for their own sake. Managers and engineers should collaborate closely at every stage of the project to ensure that real value is being created.
Sincerely,
Jeff Buterbaugh, Ph.D.
Business Development Manager
National Instruments
@measure @nivision
You get what you measure...
Peter Martin's book, "Bottom-Line Automation" is a must-read on this topic. Engineers often fail to measure and report their results in business terms.
As a result, engineers can easily drive toward the wrong metrics. For example, take a panel or DCS replacement project. If the project team focuses on complete and accurate replacement of the previous DCS, they may well minimize project cost and schedule. However, they may have missed a HUGE opportunity to improve production, and reduce energy costs. (Through new control strategies, tuning, and similar improvements).
I have personally seen this scenario many times. As an engineering community, we owe it to our companies to identify process and control improvements, rather than simply replace old technology with new.
This all starts long before the project is funded...Make sure the project goals, objectives, and funding include the ability to improve the plant. If these goals are not defined up front, it will be an uphill battle to add even small budget and schedule items later. If you don't include process and control improvements, you are just churning hardware.
After funding, the quest for business benefit needs to continue all the way through project close-out. This means that the project team cannot fold up and go home after the new DCS is in place. They must stick around and document the results...again in business terms.
Since you got me started, one more point to make here: Probably one ofthe biggest opportunity areas for the automation engineering community is COMMUNICATION. We need to communicate our thoughts, ideas, and especially, our business results. We need to actively spread the word to management. If we don't tell them how much money we saved, how will they know? Why would they agree to fund our next project? Why would they see a need for process controls at all?
George Buckbee, P.E.
V.P. Product Development at ExperTune.com
George Buckbee, P.E.