Problem with hourly arrival

In my model I use different patient arrivals, one of them is the "aankomsthelikopter".
In this patient arrival I use the hourly arrival table like in the attachment.
With this table it will not 'give' patients the complete runtime, it will stop at a random time.

If I add a 1 on a random place in the hourly arrival table it will 'give' patient the whole time, but with this adjustment my patient throughput is not right.
How can I fix this so it will give patients the whole time without adding a 1 in the table.

I'm assuming you are running the model for multiple weeks, and you're discovering that for some weeks there is not a single helicopter arriving, and that when a zero arrival week occurs, the model stop, correct?

Based on the distributions you're using to define the arrivals within each hour of a given week, it is not a surprise that some weeks do not have a single arrival. Well, this is a problem with the way the Hourly Arrival table works. It depends upon at least one arrival within the week to create the event for the start of the next week.

I suggest you use the Custom Arrivals Table. This is the ultimate arrival schedule. You have total freedom to specify whatever time period you want your arrivals to occur within (you're not forced to an hour period). Each row in the table can specify a different time period, different number of arrivals and a different PCI. The time periods defined in the various rows of the table can overlap and do not need to be organized chronologically! Because you get to specify the time interval between repetitive start times for the table (doesn't need to be a weekly time interval), you have total flexibility in defining your arrivals, and you will not run into the problem you're experiencing now with small number distributions possibly stopping the model when no arrivals occur within a set repeat interval. You also won't need to carry so many decimal numbers in the definition of your distributions hopefully.

By the way, why are you using poisson(0)? It's my understanding that this will always return 0, so you might as well just type 0.

Incidentally if you ever use a non-discrete distribution to define the number of patients arriving within an hour of the Hourly Arrival table, keep in mind that rounding does NOT occur. Instead, the value returned by the continuous distribution will simply be truncated down to the nearest integer. You may need to round and cap at zero like this:

maxof(0, round( normal(5,1), 0))

Thank you Cliff King.
The Custom Arrival Table fixed my problem!