Is Simulation an Art or Science?

Using “Process Simulation” tools is the most fancied and a dream activity amongst practising process engineers. It is treated as the whitest of the white-collar jobs. The brightest engineers are selected for carrying out process simulations.

Undoubtedly, the modern commercial steady-state process simulation suites can synthesize every process existing under the sun. The results can be presented with quasi chemical accuracy, too, and at lightning speed using the power of computer processors.

Interestingly, the simulation softwares can produce multiple answers to the same problem, leaving analysis and selection of correct/appropriate solutions at the mercy of interpretation by the user. A glycol dehydration column with the same specs and capacity may have 10 theoretical stages against 20 obtained by another user. These are not unreal situations. Then is simulation an art or science?

I think it’s a scientific art. The art is based on sound scientific principles. Dag-drop and create a model with whatever is available or given input data, sit back- relax and press the run button. Rest assured that the results will be artistically beautiful but variable. So are the modern-day simulators not smart enough? The simulators are certainly SMART, but the prerequisite is that the user should be equally SMART. I sometimes wonder if the speed, accuracy and precision that can be extracted from the simulation model are the boon or bane.

Essential to note the below principles:

  1. First Principle: Mass balance theory states that ‘Garbage In’ equals to ‘Garbage Out.’ The simulation results are highly dependent on the quality of input data fed. The input data includes configuration of problem statement, choice of equations, algorithms, tolerances, physical properties estimation methods, tear stream data, sequencing, initial estimates, etc.
  2. Second Principle: Converging a flowsheet is only the beginning of the simulation. Obtaining a convergence is not the ultimate objective of the simulation. The converged simulation can act as a reasonable initial estimate for subsequent runs. It is imperative to validate the results and verify the result trends under several conditions is in accordance with fundamentals and scientific theories.
  3. Third Principle: Thermodynamics is a universal king that dominates the simulation world. It has a role in everything that affects any chemical engineering calculation. Physical property estimation, multiphase equilibria calculation, or predicting is commonly known as water boiling point at atmospheric pressure (Simple? Try getting the exact answer of 100 Deg C with your simulator to see for yourself).

When the simulation is commanded to run, the screen flashes several colours jumping from one unit operation to another – green-red-blue. Don’t be colour blind. Don’t follow the simulation results blindly. It’s just a tool, not magic. Simulation tools are as powerful as the user. You need to extract this power by applying some more of your brainpower!

Read more at

Should you rely on your simulation results? (digitalrefining.com)

Author

  • Atul Choudhari with 28 years of engineering experience is the Chief Technology Officer at Tata Consulting Engineers (TCE). In this role, Atul’s primary involvement is in technology development, Innovation management and knowledge management. He has contributed several technical papers addressing a wide range of issues related to process design engineering both at national & international journals and conferences.

Atul Choudhari

Atul Choudhari with 28 years of engineering experience is the Chief Technology Officer at Tata Consulting Engineers (TCE). In this role, Atul’s primary involvement is in technology development, Innovation management and knowledge management. He has contributed several technical papers addressing a wide range of issues related to process design engineering both at national & international journals and conferences.

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *