Random Variation in a Model

For the NHPP Rate option under the Data Driven option of a Numeric Source block, the following inputs are required as shown in Figure B.5:

Data Name: The location of the SAS data set or JMP table that contains the data to be used to estimate a cumulative intensity (rate) function.
X: Name of the column in the data set that specifies the subinterval cutoff points $x_0,x_1,...,x_ m$ so that the NHPP has an intensity function that is piecewise constant on each subinterval $(x_0,x_1],(x_1,x_2],...,(x_{m-1},x_ m]$ . The subintervals do not necessarily have equal widths. The NHPP is defined on the time interval $(0,S]$ so that $x_0=0$ , $x_ m=S$ , and m is the number of subintervals. The time units must be consistent with the data. For example, if the interval of interest is from 1:00 p.m. to 5:30 p.m., then the interval $(x_0,x_ m]$ is $(0,4.5]$ if the data are in hours or $(0,270]$ if the data are in minutes.
Rates: Name of the column in the data set where each value $r_1,r_2,...,r_ m$ is the estimated rate of arrivals over each subinterval. Specifically, $r_1$ is the rate of arrivals over the subinterval $(x_0,x_1]$ . The length of the Rates column should be one less than the length of the X column. Furthermore, the units of the rates $r_1,r_2,...,r_ m$ must be consistent with the rest of the simulation model. For example, if the units used in the model are minutes, then the rates $r_1,r_2,...,r_ m$ must be specfied in number of arrivals per minute.

Figure B.5: NHPP Rate Option for the Numeric Source Block