Analyses of undergraduate degree programs in chemical engineering generally come from two perspectives, and with predictable results. Educators conclude that the basic principles are instilled into the students, most find gainful employment at a respectable salary, and everything is just fine. Employers conclude that the graduates know lots of theory but nothing about practice, the result being somewhere between a gap and a chasm that the employer must bridge at a substantial expense.
What's needed is an analysis from the perspective of the student, with emphasis on financial aspects. Industry assesses the worth of ventures using tools such as discounted cash flow (DCF) to compute net present value (NPV). Why not assess the worth of an undergraduate degree in chemical engineering using these same tools?
The situation today has changed little from when I left academia in 1979 after 13 years on the LSU faculty. I was on the chemical engineering faculty, electrical engineering faculty and for five years was chairman of the Computer Science Dept. Obviously, I was having some difficulty holding a steady job.
Dr. Jesse Coates was the father-figure of chemical engineering at LSU—his father started the Audubon Sugar School in New Orleans, which eventually evolved into the Chemical Engineering Dept. I recall Dr. Coates telling me, “Son, you have two choices—you can either teach or go to work.” So I resigned a tenured, full professorship to go to work. I left with considerable sadness—LSU had been very good to me.
But in addition to computing DCFs, we need some straight talk on certain aspects of a chemical engineering education.
Engineering faculty members are under intense pressure to secure research grants and contracts. These are largely from the federal government, and the reason is quite simple—for every $1 from industry, the same effort will get $10 (or more) from Uncle Sam. The occasional grant from industry is inconsequential.
To be considered for promotion and tenure, a tenure-track faculty member must author papers for peer-reviewed journals. Termed scholarly publications, most are highly theoretical, often with no obvious relationship to anything useful to industry. An entry-level faculty member is given around five years to meet the requirements. The term “publish or perish” is quite accurate, and contributes to the pressure to secure grants and contracts that lead to scholarly publications.
How does undergraduate teaching figure into promotion? At most universities, excellent undergraduate teaching will not get one promoted, but terrible undergraduate teaching can prevent one from being promoted. A consequence is that non-tenured faculty members avoid courses involving engineering practice. These tend to be time-consuming, and few faculty members have sufficient meaningful industrial experience to teach engineering practice. Many departments must hire adjunct professors to teach such courses.
Now for an uncomfortable fact: the reputation of a chemical engineering department is in no way related to the quality of its undergraduate instruction program. The reputation of a department reflects the collective reputations of its faculty members. How does a faculty member develop a reputation? By publications resulting from research endeavors, most of which rely on funding from grants and contracts.
While the emphasis on research has the potential to detract from the undergraduate instruction program, another aspect is of major importance to graduating students. The reputation of the chemical engineering department influences decisions by prospective employers. On average, graduates of the more esteemed chemical engineering departments will receive better job opportunities. If a faculty member isn't developing a professional reputation (through publishing or whatever), that faculty member is shortchanging the students and should be terminated at the first opportunity.
Academic institutions are no longer ivory towers. It’s a tough business.
Funding (or lack of it)
Financial stresses within academic institutions are severe. Engineering programs, and chemical engineering in particular, are expensive. Costs continue to increase, with salaries comprising 80% or more of most budgets.
Most chemical engineering students attend state-supported universities. The governor and all legislators are committed to a world-class status for every educational program in every university in their state. But when crunch time comes on the budget, reducing financial support for higher education seems irresistible. To date, universities have offset these reductions by increasing student fees. The ease of getting student loans helped make this the path of least resistance, but this approach is not sustainable.
A crude way to view the efficiency of a chemical engineering undergraduate program is the number of undergraduate degrees per tenure-track faculty member per year. Ways to increase this efficiency include the following:
- Increase class sizes. In the 1960s, class sizes of 20 were common. Today, engineering class sizes are approaching 100 at some universities.
- Use instructors who are less expensive than tenure-track faculty. Teaching assistants (mostly graduate students) are the least expensive; adjunct professors (mostly retirees from industry) are next.
- Increase the use of computer-assisted instruction (CAI). In this regard, one has to distinguish education from training, an appropriate observation is: you train monkeys; you educate people. CAI has been proven to be effective for training, but chemical engineering is definitely an education.
- Increase the course load for tenure-track faculty. Probably one should first propose to reduce their salary, and then offer this as a compromise.
All of these have one thing in common: reducing the face-to-face time between tenure-track faculty members and undergraduate students. If this is carried to the extreme, undergraduate students will earn their degrees without ever seeing a senior faculty member. Students deserve better.
Eliminating tenure is often a component of simplistic solutions. Those in industry aren't guaranteed a job for life, so why should university faculty members be treated differently? For a simple reason—a person explaining to a class of students why the governor is a jerk should not have to fear retribution. No one from industry is likely to make such a presentation, nor is a chemical engineering professor. But a professor of political science just might. Of course, the next question is why senior faculty members are afforded this protection but junior (non-tenured) faculty members are not.
Engineers are probably not as immune as they think. Shortly after Hurricane Katrina, Dr. Ivor van Heerden, a research geologist at LSU, pointed out that the improper construction of the levees led to their failure and the flooding of New Orleans, a view clearly critical of the U.S. Army Corps of Engineers. The implications were huge—the flooding of New Orleans is generally viewed as a natural disaster, but improper levee construction makes it a man-made disaster. Van Heerden’s view seems to have prevailed.
Van Heerden called it as he saw it–that’s academic freedom. However, the interim dean of engineering did not renew his contract, a personnel action that eventually cost LSU about $1 million. However, the larger loss is that van Heerden is no longer a member of the LSU faculty.
Trust me on this one–interim deans do not make such decisions on their own. Tenure was designed to protect faculty from politicians; is the new role of tenure to protect faculty from university administrators?
Some still cling to the desire to create a separate curriculum in control engineering. Given the fiscal realities, that's not going to happen. A new curriculum inevitably increases costs, a fact not lost on college administrators. An alternative is to create a control engineering specialty within chemical engineering, but the course requirements for chemical engineering don't permit sufficient flexibility.
Another problem with new curricula is that by the time academia responds, the need for that specialization has passed. In the 1960s, a number of universities, including LSU, introduced a curriculum in nuclear engineering. The future for nuclear technology was bright, or at least most thought so. Where are these programs today? Very few standalone nuclear engineering programs remain; the one at LSU morphed into minor in mechanical engineering.
Have we missed the window for control engineering? Probably. Into the 1980s, most process companies had internal groups, specializing in developing process control applications in their production facilities. Today is very different:
- Internal process control groups have been downsized if not eliminated.
- Companies now outsource most, if not all, work pertaining to their process controls.
- Few major chemical, petrochemical and refining companies view their know-how in process controls as providing an edge over their competitors.
For the few control specialists required by the process companies, hiring a chemical engineer and transforming him/her into a control engineer is quite viable (some would argue preferable). It is well-known that chemical engineers make the best control engineers, so why change anything? Let’s focus on improving chemical engineering, not creating a new degree program.
Process control course
At the undergraduate level, most chemical engineering departments offer one elective course on process control. This has been a very popular course, but the popularity seems to be dropping, probably for the same reasons cited previously regarding the need for a curriculum in control engineering.
The content of the process control course, as generally taught, exhibits perhaps the widest gap between what's taught and what's practiced in industry. Much of this pertains to linear systems theory and LaPlace transforms. A partial fraction expansion might be good for exam questions, but is it ever used in practice?
Some say reduce linear systems theory content; I say eliminate it entirely. Process types just don't get s, even if only used for notational convenience. The contents of a true process control course would include the following:
- Focus on developing piping and instrumentation diagrams (P&ID) using steady-state relationships and steady-state simulation models as the primary source of the required process understanding. None of the key aspects have anything to do with dynamics. These include limitations on operations (constraints); nonlinearities from equipment such as heat transfer; steady-state interaction between process variables; and propagation of variance from one variable to another.
- Do everything in the time domain. Don't even mention LaPlace.
- Explain PID in the time domain, including parallel/series, reset windup, initialization/tracking, reset feedback, remote setpoint (cascade), etc.
- Present time constants and transportation lags, including their impact on loop performance and how the PID tuning coefficients are related to these parameters.
- Cover control valves and pumps with variable speed drives. Converting inherent characteristics to installed characteristics is a good exercise in fluid dynamics.
And there are some topics to avoid:
- Examples from batch are OK, but it's not the responsibility of the process control course to introduce students to batch processes.
- Leave optimization, model predictive control, etc., to advanced degree programs.
- Leave safety to other courses, but emphasize that the process controls should never take an action that would elicit a reaction from the safety system.
- Avoid systems (bits and bytes) topics such as PLC programming, DCS configuration, graphic displays, etc. Present process topics only.
- The problem is most departments have no one who could teach such a course.
Preparedness for careers in industry
No one can fault an employer from wanting a new employee to be productive immediately. This is generally the case with trades such as pipefitters and electricians. However, this isn't the case with chemical engineers straight out of the university. The complaint isn't a lack of knowledge, but an inability to apply this knowledge to the real-world problems that arise in industry. Most learn how to do this on the job, preferably with good mentoring.
Industrial employers wish this were done prior to graduation. Most also recognize the obstacles:
- Few faculty members have sufficient engineering practice orientation to teach such material.
- A four-year curriculum doesn't provide time for this. A five-year curriculum might, but potentially the result would be to teach more theory.
- An increasing number of graduates find employment other than with the traditional employers. For example, chemical engineering is a good background for pursuing a career in medicine.
Most in industry are resigned to the situation as it currently exists, but this does not keep them from complaining (whining may be better use of the King’s English).
Even when I was at LSU, a five-year undergraduate program was being discussed. And some industrial recruiters thought this was an excellent idea. But Dr. Coates could end this discussion abruptly. He was fond of DCF analyses, and was doing these prior to the age of electronic calculators and computers. But then this was a man who could design a seven-effect sugar evaporator in his head.
Dr. Coates applied the DCF analysis to the five-year chemical engineering curriculum and found it to be economically unsound, given the differential between the starting salaries for five-year graduates vs. four-year graduates. And he seemed to take delight in pointing this out to proponents from industry, followed by stating that when industry paid an adequate differential for a five-year graduate, LSU would be delighted to offer a five-year program. None took him up on that. As a couple of Shell Oil guys once told me, you will be amazed at our flexibility when we see the price.
The gap or chasm between what a student knows on graduation and what is required for a career in industry must somehow be filled. However, the student must not be saddled with the costs. Industry must step up and pay these costs. We can negotiate how they pay, but not if they pay.
The accepted requirement for a four-year degree is 128 course hours, which translates to 16 hours per semester. But every student knows that the demands of one three-hour course can be very different from another three-hour course. Making sure that a three-hour course is not actually a five-hour course is not as easy as it seems. As the base of knowledge expands, it's very tempting to add material to a course and not increase the hours credited for the course.
My four-year degree in chemical engineering comprised 144 hours. By the time I completed graduate school, it was down to 128. But such reductions can be smoke and mirrors. One change was to make calculus the first math course credited toward the degree. Previously, algebra and trig courses were credited toward the degree. This accounted for six hours of the reduction. The knowledge required for the degree in chemical engineering did not change; the only change was in how to keep score. You can’t just say that 128 hours are required for the degree, so it is a four-year degree program.
At least on paper, most chemical engineering undergraduate programs are four-year programs. But how many students complete their degree in four years? This question is not easy to answer. A student with a part-time job may take a reduced course load, which will extend the time to graduation. But of the truly full-time students, how many graduate in four years? Or max four and a half years? If the answer is very few, a five-year program is masquerading as a four-year program. If so, there is a serious truth-in-marketing problem.
Analysis from the student’s perspective
Today, who is analyzing the economic value to the student of an undergraduate degree in chemical engineering? Most get a good paying job, so it's casually assumed that the undergraduate degree is a prudent financial investment.
The two major reasons for pursuing any discipline are to obtain a good job upon graduation, and for the love of the chosen discipline. While I was at LSU, the mathematics department was highly theoretical. To them, the only reason to study mathematics was a love of mathematics. But to their credit, they were very upfront about job prospects on graduation: you will get a job not because you are a mathematician; you will get a job in spite of being a mathematician. From their perspective, DCFs are irrelevant.
A common component of the sales pitch for pursuing a degree in chemical engineering is the prospect of a good job on graduation. Don’t we need a DCF to back up this claim? Too bad Dr. Coates is no longer with us. You can't expect a prospective undergraduate student to compute the DCF (although given the capability of the students attracted to chemical engineering, most could do the math). If chemical engineering is to continue to attract the top students, someone on the chemical engineering faculty needs to be computing DCFs.
The DCFs would clearly show some unpleasant realities:
- Each increase in university fees decreases the NPV of the degree.
- Borrowing money by a student lowers the NPV. Some are so deeply in debt that they have to start work as soon as possible.
- Running a five-year program disguised as a four-year program is a financial disaster to a student.
- Cutting back on the course load because of a part-time job significantly lowers the NPV.
Anyone encouraging high school students to pursue a degree in chemical engineering should be very interested in the results. In this respect, do we have a truth-in-marketing problem?
Homepage image courtesy of Stuart Miles at FreeDigitalPhotos.net