I am reading a paper in which a nonhomogeneous Poisson process is constructed with intensity function $$ \lambda(t) = \left\{ \begin{array}{ll} 0.42 &\text{ for } t \in [0, 121), \\ 0.35 &\text{ for } t \in [121, 244), \\ 0.40 &\text{ for } t \in [244, 365). \\ \end{array} \right. $$ Then we are told the expectation (i.e., what I assume is the mean value function) of the process is given by $$ \Lambda(t) = \left\{ \begin{array}{ll} 0.38t &\text{ for } t \in [0, 121), \\ 0.35t + 45.98 &\text{ for } t \in [121, 244), \\ 0.40t + 85.4 &\text{ for } t \in [244, 365). \\ \end{array} \right. $$ Where did these values come from? My understanding is that, for $t \in [0, 121)$, we should have that $$ \Lambda(t) = \int_{0}^t0.42ds = 0.42t; $$ and for $t \in [121, 244)$, we should have that $$ \Lambda(t) = \int_{121}^t 0.35ds = 0.35t - 42.35; $$ etc.
What am I missing here? (Note that——for whatever reason——I cannot find a single example online where a similar computation is done.)