> I have two subsystems of which one has no. of failures per demand and another has a failure rate per hour. Now how to compute the system failure? < Interesting question but answer requires more sharply defining what is meant by "... how to compute the system failure?". Demand failure can be based upon unavailability (function of failure rate and repair rate) or latent failure probability (function of failure rate and maintenance/calibration interval). In other words, no direct conversion between failures/demand and failures/unit-time. Need to know problem and desired safety parameter.
I do not believe that there is any way to directly convert these. I believe that you will need another number, Demands per Time. Once you have consistent units, then the relationships between the devices will determine how you calculate the system failure rate. Bill Mostia ================================================ William(Bill) L. Mostia, Jr. PE Independent I &E Consultant WLM Engineering Co. P.O. Box 1129 Kemah, TX 77565 [email protected] 281-334-3169
