Dipankar Acharya

In Design of Experiments, we vary multiple Inputs (X's) at a time, in order to find the best combination of inputs to deliver the desired output (Y's).
 
During the same, we may come across some parameters which are uncontrollable. These are called Noise factors.
 
E.g. for an experiment on making the the right Dish, we can vary and control the quantities of constituents, temperature of cooking, time of cooking etc. But we cannot control factors like the mood of the Chef, the skills by which he mixes the dish etc. Mostly these human factors become noise and can affect the DOE results.
 
Noise factors are of 2 types : Known and Unknown
 
To address the same we can use 2 methods :
 
a) Known Noise Factors : Use Blocking (A block is a group of homogenous experimental observations).
 
b) Unknown Noise Factors : Use Randomization (make the Run order separate from Standard Order)

In Project management, after identification of activities, we analyze them if any relationships exist between those. If relationships exist, it is better to highlight them in Project Schedule Network Diagrams to maintain the project schedule better.
 
There are four (4) types of dependency relationships. Taking an example of building a house on a piece of land :
 
1) Finish to Start : Predecessor must finish before Successor can start. (Example : Land must be purchased before house building can start)
 
2) Start to Start : Predecessor must start before Successor can start. (Example : Land digging must start before House foundation can be laid)
 
3) Finish to Finish : Predecessor must finish before Successor can finish. (Example : Laying Cement must be complete before final Painting can be completed)
 
4) Start to Finish : Predecessor must start before Successor can finish. (Example : Whitewashing must start before final painting can be completed)
 

The correlation coefficient merely shows the relationship (positive or negative) between two variables - and the extent of the same - via the value of the coefficient (-1 to +1).
 
It does not express the relationships between the 2 variables or can't predict the value of the dependent value based on the changing value of the independent one.
 
The Regression analysis expresses the same via an equation, analysis of R square (how much of the variation of the output is explained by the equation) and also can predict the "Y" based on changing values of the "X's".
 
Further MLR or Logistic Regression can predict relationships between Multiple factors as well as multiple types of (Discreet and Continuous) variables. 
 
Thus Correlation coefficient does not give the complete picture and we need to do Regression as well.