SRC Forum - Message Replies

Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum (src_forum@alionscience.com )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Original Message:

Posted by: IC
Date posted: Mon Mar 5 16:07:21 US/Eastern 2007
Subject: Reliability Apportionment
Message:
I am looking for information in relation to apportionment and the techniques used for calculating it. I have calculated the apportionment for a System with 6 sub-systems (Using the Equal Apportionment Technique from MIL-HDBK-338). NOW I am trying to calculate the apportionment for the sub systems with the new information that each sub system has now been "weighted" (i.e sub sys 1 will be responsible for 20% fo all system failures and Sub sys 2 47% etc etc). Can anybody tell me how I will calculate the new reliability apportionment based on the "weighted" factors or let me know of any techniques/equations that I can use.

Thanks for the help.

Reply:

Subject: Reliability Apportionment
Reply Posted by: John Cloarec
Date Posted: Tue Mar 6 6:26:38 US/Eastern 2007
Message:
The Equal Apportionment Technique is not very sophisticated, that's why you need another method for subsystem apportionment. In the same document MIL-HDBK-338 you can find a more sophisticated and useful method : Feasibility of Objectives Technique which use 4 parameters and a set of simple formulas. You have no limitation in the decomposition of your system. You can put your model in a excel file and redo the calculation any times you want to assess the impact of some changes in the value of each parameters. This method is interesting because there is a great analogy with reliability prediction as per MIL-HDBK-217. The only limitation is that the system configuration is assumed to be serial (no redundancy). I've conducted a survey for EdF in France. The papers I have are all in French ... What we found is that the best compromise for : - reliability allocation is the Feasibility of Objectives Technique method, - maintainability allocation is the method described by Arsenault & Roberts in R&M of Electronic System, Ed. Pitman. If you can read French, I can send you a scan of my papers. Regards. John

Reply:

Subject: Reliability Apportionment
Reply Posted by: IC
Date Posted: Tue Mar 6 6:27:43 US/Eastern 2007
Message:
Thanks John, Unfortunately I do not read or speak French. Also thanks for the info, however this techniques does not fit with what I am trying to do as I am only in the requirements definition phase of the project and am only trying to define a sub system reliability requirement based on the system requirement of 0.8. I have 6 subsystems and when using the Equal apportionment technique it gives a reliability of 0.9635, however based on experience I know that each subsystem effects the systems differently, hence the % weighted values (System 1-20%; Sys 2; 47%; Sys 3 8%; Sys 4 8%; sys 5 2%; Sys 6 15%)for system failure. I am just trying to define what the apportioned reliability should be. Is the calculation just a simple proportional calculation or is there some other method of calculating this high level number. Thanks and Regards

Reply:

Subject: Reliability apportionment with side constraints
Reply Posted by: Larry George
Date Posted: Tue Mar 6 6:28:47 US/Eastern 2007
Message:
95.68% 90.35% 98.05% 98.05% 99.58% 96.67% is the least squares solution that I get from a quick and dirty analysis. Check it. The product is 80%. Consider using the workbook linked to www.fieldreliability.com/RalloMan.htm with your side constraints. Alternatively, plug in the parts costs and system failure costs and see what solution cost-based allocation leads to.

Reply:

Subject: Allocations
Reply Posted by: BWD
Organization: System Reliability Center
Date Posted: Fri Mar 9 13:18:14 US/Eastern 2007
Message:
The three allocation techniques that most common are; Equal Allocation, AGREE Allocation, and Feasibility of Objectives Allocation. I believe that the feasibility technique is the best as it weights the design for intricacy, state of the art, operating time and environment.