Do experiments, observation, simulation, or survey to solve the problem scientifically.

There are many books about this tool that we do not need to explain any more detail.

In principle, experiment means we want to know the response based on one or several combinations of factors. We can also know what factors influence other factors through experiment. In experiments, we can control the subject under very limited treatment.

Observation means we do not have much control about the subject. We just observe the behavior and guess some theoretical background to explain that behavior. If our theoretical background is correct, then the prediction of the theory will match with the result of the observation.

Simulation means we want to imitate part of the real world using several models and the interaction of those models.

Survey means we want to know the collective behavior or attitude or perception or opinion of people through systematic questions that we usually put into questionnaires.