Throughout university, I had the impression that performing complex calculations, describing systems perfectly, and arriving at numerical answers with extreme accuracy was the job of an engineer. When I transitioned to the workplace, I had to completely change my mindset. I quickly realized that real-world engineering is not a textbook problem that can be solved with exactness. Trying to be as accurate as we were taught in school often takes a lot of time and/or money, and usually is not possible due to the inherent error and uncertainty in the tools used to collect raw data. I then realized what makes engineers valuable—their ability to apply systematic thinking to solve challenges with limited data.
Of course, there are many important tasks where an engineer must take the time to be as accurate as possible and pay attention to significant digits in each step of the calculations. In my experience, however, value is not realized by producing precise numerical answers to a problem. Value is added by providing sound opinions and judgments about which action should be taken.
Proficient engineers balance time, budget, and technological constraints to formulate solutions with a risk that is as low as reasonably possible in the absence of complete or extremely accurate data. It takes experience and exposure to develop the skills to build relevant models that describe and forecast a system or solve a problem given incomplete or imprecise data.
How Can Young Engineers Gain Valuable Estimation Skills?
The time we spend in school and training provides us with a technical skillset for building solutions. However, it may take years of work experience to develop an intuition for what level of engineering is required for a given problem. A good place to start in developing this intuition is the Pareto principle, which states that roughly 80% of the effects come from 20% of the causes. In other words, the real challenge is to collect, analyze, and discern the key parameters that will noticeably affect how a process is built or operated. This enables an engineer to build a solution to the appropriate level of detail. Learning why senior engineers decide to ignore certain parameters, provide quick estimates for others, and perhaps spend time to determine accurate values for some is invaluable to young engineers. Finding this balance for new problems is a mix between science and art—there are no universal rules one can easily describe or teach.
I’ll admit that this philosophy is something I am still getting comfortable with, but there are some strategies I have used to help implement it at work. Understanding the application of my outputs or calculations is one of the most helpful strategies I use.
Knowing the level of precision needed by the end user allows an engineer to decide how accurate the calculations should be. For example, an engineer may be asked to find the pressure at a specific point in an oil and gas process, but the person inquiring is really only interested in knowing whether the pressure is higher or lower than the maximum operating pressure. In this example, understanding how the pressure would be used allows the engineer to quickly find the essential information for the particular issue, rather than calculating the exact pressure to three decimal places. It is amazing how often simply eyeballing some trends on a graph leads to the same recommendations as a time-consuming numerical analysis. Just like eyeballing ingredient measurements when cooking, eyeballing engineering data gets easier with experience, and is often as accurate as needed.
I have also found it very helpful to ask my coworkers about the inherent accuracy of the data we use to regularly justify our decisions. Much of the instrumentation used in the oil and gas industry is subject to harsh operating conditions and produces noisy readings. Having an idea of the inherent error in the raw data helps guide the best approach to analyze it. For example, one may decide to use the most recent value for measurements known to have high accuracy, whereas one could use average values or a regression to analyze measurements with more uncertainty.
Consider the Scale of the Problem
Another consideration when deciding on the appropriate level of detail is the scale of the problem. For example, an engineer may be able to use more simplifications when designing a process with a flow rate of 45,000 kg/hr, compared to a 100 kg/hr process. The large capacity of most oil and gas processes causes many variables important for small-scale processes to become negligible.
Young professionals should have the mindset that the most important part of the problem-solving process is actually determining which problems need to be solved to add value to the business. Working in industry has taught me that the first step in solving a complex problem should be to break it down to see if there is a dominant sub-problem that allows one to neglect other subordinate components of the issue.
Young professionals should challenge senior engineers and professors and ask them about the common simplifications that can be made for industrial applications. Case studies should be taught in every engineering class, where real-life processes are analyzed and justification for why existing specs do not match the theoretically optimum specs are discussed. For example, choosing what diameter of pipe to use in a process is constrained by the available sizes of pumps and valves as well as the standard pipe sizes offered by the manufacturer. As a result, simply comparing a few possible scenarios will yield the same recommendation as performing a tedious calculation to optimize the pipe diameter and picking the realistic option closest to the theoretically optimum.
Employment may be very different from studying engineering at university, which cannot only cause frustration among new engineers, but impairs their ability to immediately add value to the business. A common thread between academia and industry is the value of time. The upfront knowledge of the precision required to generate an adequate answer will maximize the value generated (in dollars or grades) for the time invested. In most cases, the required precision for a problem is initially unclear. An upfront understanding of the audience, or users of the output, provides context for an engineer to judge what the appropriate level is.
It is important to develop an intuition for knowing when estimates are appropriate. This takes time, so young professionals may need to solicit the experience of others better-versed.
The ability to use estimates instead of detailed calculations is one of the most valuable time management skills for engineers. New engineering graduates should have the confidence to make recommendations based on estimates if they are able to enumerate the main sources of error in their methodology. In the absence of recognition of the limitations of data or project requirements, reaching out to coworkers or supervisors for support in determining the appropriate level of analysis before beginning a new task is imperative. Both new engineers and students should learn from this common workplace experience and carry with them a practical awareness, not a fear, of uncertainty.
Amanda Calleberg is a reservoir engineering intern at ConocoPhillips Canada, and a student of chemical and petroleum engineering at the University of Calgary. She is the student liaison for the SPE Calgary Section’s Young Professionals Committee and the director of club relations for the University of Calgary Petroleum and Energy Society. Calleberg is a team lead for the SPE Drillbotics competition and a participant of the SPE PetroBowl tournament. She has conducted university research on the numerical simulation of methane combustion in porous media, as well as modeling downhole drilling parameters. Her other involvements include the University of Calgary Scholars Academy program, Unconventional Engineers club, and Dance Montage performances.