Page 105 - Contributed Paper Session (CPS) - Volume 1
P. 105

CPS1167  Indrani B.
            variable Y is given by (1) and (2) respectively. The probability density function g(t)
            and the cumulative distribution function G(t) of T are given as
                               1  − 1  − 1
                         () =     −                                                 for y < ln 
                         1     
              () =                          ln − 2  ln − 1
                               1  − 2  − 2 (    −    )              (3)
                                          
                       () =     −                 for ln  <  < ∞
                        2
                     {         

            and

                                       − 1
                         () = 1 −  −                                                  for y < ln 
                         1
               () = {                    ln − 2  ln − 1
                                      − 2 (    −    )                   (4)
                        () = 1 −  −                for  ln  <  < ∞
                        2

                                   1
                                                  1
                        1
            Where  = ln   = ln  and σ = .
                             1, 2
                    1
                                        2
                                   
                        
                                                  

                Suppose a sample of  experimental units are placed on a simple step-stress
            life test at an initial stress level of s1 and the stress level is changed to   at a pre-
                                                                              2
            fixed  time   .  Then,  the  progressive  Type-II  censoring  is  implemented  in  this
            experimental setting in the following manner. At the stress-level   and at the time
                                                                        1
            of the first failure,   of the  − 1 surviving units are randomly removed from the
                               1
            experiment. At the time of the second failure, R2 of the  − 2 −   surviving units
                                                                         1
            are randomly removed from the experiment, and similarly the test continues until
            time . Let   be the random number of units that fail at stress level s1 and
                        1
            be the total number of the censored units at stress level    where   denotes the
                                                                           1
                                                                 1
            observed value of  . Then, after time   (at stress level  ), at the time of the ( +
                                                                                    1
                                                               2
                              1
            1)-th failure,  +1  of the −   −  (1) − 1 surviving units are randomly removed
                                        1
            from    the    experiment.    At   the    time    of   the    ( + 2)  -th
                                                                            1
            failure,   1 +2  of the −  − (1)    1 +1 −2  surviving  units  are  randomly  removed
                                  1
            from the experiment, and similarly the test continues at the stress level  . Let
                                                                                  2
                       be the total number of the censored units at stress level   for a fixed
                                                                            2
            value of , the total number of observations. Then, let   =   −   denotes
                                                                              1
                                                                    2
            the random number of units that fail at stress level  . With ,   (  =  1, 2,· · ·
                                                                          
                                                               2
            ,  − 1) fixed in advance, the test continues until the -th failure at which time
            all  the  remaining             surviving  units  are  removed.  Note  that  =
             +  +  (1)  +  (2)  where   denotes the observed value of  . If  = ⋯ =
              1
                                                                               1
                                         2
                  2
                                                                          2
              = 0,     = ⋯ =  = 0, then  =  which corresponds to the complete
               1     1 +1      
            sample  situation.  If   = ⋯ =  1  = 0,  1+1  = ⋯ =  −1  = 0 and  =  −
                                                                               
                                  1
             then it corresponds to the conventional Type-II right censoring scheme. Note
                                                              94 | I S I S   W S C   2 0 1 9
   100   101   102   103   104   105   106   107   108   109   110