Page 270 - Well Logging and Formation Evaluation
P. 270

260               Well Logging and Formation Evaluation

            The VOI is clearly given by VOI = (DEMV - Z). Now consider the
          effect of reliability. This means in effect that sometimes the parameter C
          will be found to be true even though the field is in state 2, and vice versa.
          The reliability is expressed via:

            R = ( P  C S )
                     1

          i.e., the probability that C is found to be true when the field is indeed
          in state 1. For simplicity we will also assume that this is the same as
          P(C¢/S 2), i.e., the probability that C is found to be false when the field is
          indeed in state 2. The introduction of R clearly leads us to have to con-
          sider the additional NPV scenarios:

          1. NPV(S 1 and C¢): the NPV realized when the field is thought to be in
             state 2 but is actually in state 1.
          2. NPV(S 2 and C): the NPV realized when the field is thought to be in
             state 1 but is actually in state 2.
            We will now calculate DEMV, introducing these additional scenarios:

                                                               )
                                      )
                      Z
            DEMV =- +     P() [ *  P(C S 1 *  NPV(S 1  and )+  P(C S 1 *
                                                            ¢
                                                    C
                            S 1
                                                   )
                                C
                                                ¢
                            and ¢)]+  P() [ *  P(C S 2 *
                     NPV(S 1           S 2
                                           )
                                C
                            and ¢)+  P(C S 2 *       and )]-  EMV   (A4.10)
                                                         C
                     NPV(S 2                  NPV(S 2
            Using our definition of R:
                            S 1 *
             DEMV =- +    P() [R*   NPV(S 1  and )+ (1  - R *
                                               C
                                                        )
                       Z
                                                           C
                                C
                                        S 2 *
                             and ¢)]+  P() [R*         and ¢)+
                     NPV(S 1                    NPV(S 2
                          )
                                        C
                     (1 - R *  NPV(S 2  and )]-  EMV                (A4.11)
            Another way to look at this is that we have four possible combinations
          of C, C¢, S 1 , and S 2 , each having a probability and NPV associated with
          it. The EMV is therefore:
            DEMV =   P(S 1  and )*  NPV(S 1  and )
                             C
                                            C
                     +  P(S 2  and )*  NPV(S 2  and )
                               C
                                              C
                     +  P(S 1  and ¢)*  NPC(S 1  and ¢)
                              C
                                              C
                     +  P(S 2  and ¢)*  NPV(S 2  and ¢)
                               C
                                              C
                     - Z  -  EMV                                    (A4.12)
   265   266   267   268   269   270   271   272   273   274   275