Ask Russ Rhinehart - Why use phenomenological models based on first principles over data-based modeling options?

Russ' Response:

I would broadly classify models as empirical or phenomenological.

Empirical models are developed by matching flexible mathematical functions to data.  The functions might be neural networks, or classic statistical power series for regression.  These approaches are used in Big Data, Machine Learning, and Artificial Intelligence.  They might be termed data-based, or regression, or even model-free models.  Many folks are reporting benefits from such data-based, empirical models.  

By contrast, in phenomenological models, the mathematical functions are grounded in a mechanistic relation.  These models could be termed rigorous, first-principles, mechanistic, or cause-and-effect.  First-principles models would be the same sort as those used in undergraduate instruction in process units, revealing basic principles, but not attempting to be rigorous about details of secondary importance.  

A first-principles model does not include phenomena that would be of secondary relevance.  Such might include the thermal influences on density, friction factor dependence on Reynolds number, the volume in a pump volute or that in an expander-contractor, or the 1.85 exponent value that better match pressure loss phenomena than the squared exponent of the Bernoulli ideal flow relations.  One could progressively complexify the relation seeking to develop a rigorous model, and violate the K.I.S.S. principle.  I don’t recommend doing that, until you know that it is necessary.

An extended version of this pod with details and examples can be accessed under the Modeling/Simulation menu on my web site, www.r3eda.com.  

Which method do you prefer?

I prefer the use of first-principles models. But expediency might justify empirical modeling.

In subsequent Pods, I’ll introduce how to create your own first-principles models, how to simulate environmental vagaries, how to calibrate and validate models, and how to use the models to evaluate the various economic indicators of transient events.  I hope to visit with you later.  Meanwhile, visit my web site www.r3eda.com to access information about modeling, control, optimization, and statistical analysis.