Zornitsa - it would depend on what you mean by some of your terms. All critical analysis should be 'valuable' and it depends what you mean by 'complex'. The complexity would depend on the task i.e. is it part of an integrative review and/or a systematic review etc? The best way to ensure a 'reliable' critical analysis, as you state, is to use a particular and systematic method/approach/framework. For me, that means a systematic literature search and critical review process.

Attached here are a few chapters that you might find useful. One is the initial critical search and review process - the other is a more detailed framework for reviewing single quantitative and qualitative studies. 

Thank you! I searched for a methodology which combines literature review with validation process toward subject-matter experts or a combination with some statistical methods. Because I find the critical analysis too dependant on the person who performs it and that's why I am searching for a way to prove the results from it :)

I will look at the attachments that you sent to me. Thank you again.

Dear @Zornitsa, @Dean has attached his excellent resource Searching and reviewing the research literature and Critically reviewing research studies by Zevia Schneider. 

I am free to add some more resources Introduction to the Scientific Method , The Scientific method, and RESEARCH METHODOLOGY!

Good luck!

http://teacher.nsrl.rochester.edu/phy_labs/appendixe/appendixe.html

http://sciencefairproject.virtualave.net/scientific_method.htm

http://arxiv.org/pdf/physics/0601009.pdf