How to ‘Optimize’ Human Emotions?


As my first post in OR-Complete, I would like to write about an interesting question: Is there any way or approach to quantify human emotions? Well, I encountered with this question at or-exchange and I think it is interesting to talk about.

First of all, why we need to measure emotions? In human decision making systems, it is vital. From another point of view, in service sector, you need to know about human reaction. For instance, if you are operating a private hospital, you need to keep customer pleasure not less than a certain level. Another instance may be advertising. Although I have absolutely no knowledge about ads, i guess, predicting the customer behavior should be very important. So, it is clear that sometimes we need to measure or at least estimate human emotions.

So, how do we do it? In high-tech research, artificial intelligence based studies can be more complicated. I don’t know about them, but in recent years, imitating human decision making has growing importance. In business, there are ‘soft’ versions for tracking human behavior. The most common form is surveys. There are lots of surveys. Believe me, there are. Even in small enterprises there are “suggestion” boxes, which may help our measurement purpose. I think, data from these boxes may misguide us, because people are more eager to write in unpleasant cases. However, if you are able to collect surveys “uniformly” than it could help. In our university, we have evaluate our professors towards to end of each term. It is better if you are interested with student satisfaction. Although you have more fair results now, it may not be a good idea to rate professor performance just considering these evaluations.

Let turn back to original question. In OR studies, can emotions used as objective function? Can we maximize or minimize feelings? Partially, maybe. Even having difficulty about estimating one cost parameter, one can found difficult to estimate emotions. Even if there is a successful estimation (linear regression :) ) people tend to have unstable reaction even you could create same environment.

For those interested, there is a study titled “Modelling Human Decision Making” of Stewart Robinson, Warwick Business School. What do you think about it? Is it worth to study for modeling human emotions? Or do you think it is too complex and unnecessary to spend time for it?

Sertalp Bilal Çay

PhD Candidate and Teaching Assistant in Industrial and Systems Engineering Department at Lehigh University. Researcher on Conic Optimization, Inventory Theory, Supply Chain Management and Simulation. Blog: