What is the difference between business intelligence, simulation and data mining?

Russell Acaph, the father of system thinking, has a sentence with the following theme:

You do not need to anticipate what you can control it.

No sanity predicts room temperature, but controls it.

Now in the organization's space, if we divide business variables into two categories of internal and external variables, the common sense of peace prescribes that internal factors are controlled and the outlook of external factors is predicted.

With this introduction, it can be said that business intelligence is a set of meters and gauges that eventually show the result of senior management decisions in the past at the management level and the result of the actions of the headquarters at the operational level.

In fact, according to Dr. Deming's beautiful interpretation, such analyzes are like driving with a rear-view mirror

Ie seeing the past and translating them based on key performance indicators

On the other hand, the indicators, even the best ones, can not describe all aspects of the business, and it is more dangerous to look at the past and with the time lag ....
Just like a bathroom faucet, when you want to adjust the hot and cold water, based on your current sense of temperature, you manipulate the tap without letting the change happen.
So far, we are looking at the past

Now, if we want to know the uncertainty of the future that is affected by external factors, here are two approaches:

The first approach is simulation, and this is for the time when we have neither enough data nor a large part of the uncertainty.
Here, based on sample data, probable future situations are obtained
For this, we need to move forward with the "model-driven" approach. That is, we break the components of the problem into smaller variables and estimate their individual and interactive behavior and then simulate the behavior of the whole system.
Research approaches to business operations and dynamics are analyzes

The second approach is to see the future of data mining and data-optimization

Here, for the sake of understanding the future, we do not need to know all the variables of the system, but assume that the effect and effect of each known and unknown variable, in any case, shows itself in the final variable (in the so-called label data mining), so it is sufficient Estimate and cease

The goodness of this approach is that it requires us to know every single element of the system, but it needs a lot of data.

Because these days, data mining and learning of the popular machine is due to two reasons:
1- We want to know the future

We have a lot of data

This is the difference between business intelligence, simulation and data mining.