Based on a semi-Markov process J(t), t greater than or equal to 0, a r
eward process Z(t), t greater than or equal to 0, is introduced where
it is assumed that the reward function, rho(k, x) is nonlinear; if the
reward function is linear, i.e. rho(k, x) = kx, the reward process Z(
t), t greater than or equal to 0, becomes the classical one, which has
been considered by many authors. An explicit formula for E(Z(t)) is g
iven in terms of the moments of the sojourn time distribution at t, wh
en the reward function is a polynomial.