Up From the Ashes

Greg McMillan and Stan Weiner bring their wits and more than 66 years of process control experience to bear on your questions, comments, and problems. Write to them at [email protected].

Stan: What happened to process control at Monsanto after I retired?

Greg: Engineering Technology (ET), which at one time had 75 of the premier brains in process modeling and control, declined to 1/3 its size through retirement incentives. Nearly all of them were PhDs who were leaders in their respective areas of expertise. Being one of the few exceptions, I was often called Dr. McMillan, at which point I suggested Nurse McMillan would be more appropriate. Man, I learned a lot from these guys, especially my Hall of Fame compatriots Terry Tolliver and Vernon Trevathan. In 1991, a reorganization meteorite wiped us out. ET was extinct. I was homeless. I didn’t even know what had happened until I was informed by Vernon that he had created a small process control improvement team and that process control specialists Terry Tolliver and Glenn Mertz and process modeling specialists Henry Chien and Bill Hooper were in it with me. We had funding from the manufacturing directors to find and implement process control improvements.

Stan: How did Vernon get the ammunition to make this happen in a time of great downsizing?

Greg: We visited 12 of the best companies in the chemical industry to see what they had accomplished in process control, and we found a large number of varied opportunities in process control to increase profits and revitalize manufacturing. My edited report to the Chemical Manufacturing Association on the benchmarking results and my presentation on the advanced control part of this report to Control’s 2006 Automation Exchange are posted at www.controlglobal.com/articles/2008/.html

Stan: What could you do in a multi-billion dollar company with such a small team?

Greg: By 1991, DCS had been installed in all 110 production units, but in spite of enormous benefits, the real power of the DCS remained untapped. We had 140 experienced engineers ready, willing and able to do something more than just maintenance and small projects.

Stan: How did you get the priority for these engineers to work on process control improvement?

Greg: I tried an opportunity-sizing process on an agricultural plant in Iowa, computing the practical and theoretical gaps in performance for production of an herbicide. Glenn Mertz, who had an extraordinary interest in accounting for an engineer, formalized and truly exploited the sizing process. He found the best demonstrated on-stream time, production rate and yield, and after evaluation of cost sheets with the production department leaders, set this as the practical goal.

Stan: How did you get key plant personnel to agree?

Greg: We knew it was suicidal to go into a meeting and straightaway start to proclaim what wonderful things the plant should do. Glenn and another member of the PCI team studied the opportunity sizing with a business representative, operating supervisor and process technologist and reached an agreement on the value of the gaps and documented the opportunity sizing.

Stan: How did you get to the bucket of gold at the end of the rainbow?

Greg: The most experienced process engineer reviewed the process in detail, concentrating on those areas that potentially affected the gaps. I can’t emphasize how technically and psychologically important it is to have the most experienced process person do nearly all the talking and analyze trend charts of key process variables. The audience of configuration engineers, instrument engineers, maintenance engineers and technicians, analyzer specialists, operators and chemists interjected only as needed. This audience and the PCI team concentrated on being good listeners. If you had an idea, you would ask exploratory questions, but saved solutions for one to two days of brainstorming. The opportunity assessment process culminated in an opportunity summary where each idea was listed. The time, cost and skills for implementation were approximated and used to set priorities. Finally, names were assigned. “Quick hits” were items that had significant value that could be done in a few weeks, which by definition meant that no hardware or equipment was needed. Little or no new capital was required and $75M per year in ongoing benefits was delivered from the PCI program.

Stan: What was the most frequent “quick hit”?

Greg: In this vein, we have a correct reply to our January puzzler from Tien Li Chia of ControlSoft.

Tien: Some of the biggest opportunities in process control improvement would have to be

(a) Reduction of energy costs based on tighter control of the system;
(b) Overall automation improvement while keeping labor costs down;
(c) Reduced wear and tear on the equipment and increased up-time and on-spec product through optimized PID loops tuned to fit the system;
(d) Improved process control through real-time control loop performance monitoring. Process degradation happens gradually; operating conditions load and production rates also change. An online, real-time control loop performance monitoring tool can improve the performance, because both loop and process performance will be monitored and evaluated continuously in real time, and any performance degradation will be detected as it occurs, and responsible personnel notified. Problems can be corrected immediately, keeping the process at optimal operating conditions at all times.

Greg: Since loop performance hinges on tuning, the first and last thing that should be done regardless of the perceived opportunities is tuning the loop. Sometimes that’s enough to reduce cycling and allow you to get closer to the optimum.

Next month we will explore some other opportunities realized and overlooked.

In the mean time we conclude with bits of conversation Glenn and I imagined one evening after more than a few beers. Keying off a highlight of the “Newlywed Game,” we envisioned a meeting where I had to explain PCI to Bob. Since we had several Bobs (CEO, president and director), we felt relatively safe in documenting the conversation.