Stops working at 5am

Question

question

Karen Measells asked Feb 4, '19 Karen Measells commented Feb 6, '19

Stops working at 5am

hep-baseline-2nd-floor-procedure-area.fsm

Hi again,

So I've reviewed all of my distributions, and none should be returning a very high or negative value. I deleted the patients who were causing problems at midnight on the first night (PCI 16 & 19), but now the model just quits at 5am with no error message. I've checked that it's set to run until 5:23:59. I also checked all of my decision points to see if I accidentally programmed a loop, but I can't find anything. Can you take a look (again!)

Thanks :)

Karen

Software Version:

FlexSim HC 5.1.0

stop

hep-baseline-2nd-floor-procedure-area.fsm (5.1 MiB)

Answer 1 · 2019-02-05T20:05:26Z

Matthew Gillespie answered Feb 5, '19 Karen Measells commented Feb 6, '19

The Custom Arrivals table on PatientArrivals2 does not have the correct structure. To fix it, open the table and hit the up button next to the number of labels.

The table is most likely incorrect because you copied and pasted the table from somewhere else. If you do this, add an extra row or label and then delete it so that the table editor can make sure the table structure is correct. Alternatively you can import an arrivals table from excel using the excel importer.

· 8

Karen Measells commented · Feb 05, 2019 at 08:10 PM

Ok, thanks, I'll try that. FYI, Sam told me to delete that column

0 ·

Matthew Gillespie ♦♦ Karen Measells commented · Feb 05, 2019 at 08:16 PM

The Custom Arrivals table keeps an extra, hidden "Scheduled" column that keeps track of whether the arrival has been scheduled yet. Without that column the arrivals object loses track of which arrivals it's already scheduled and just makes an infinite number of arrivals (an infinite loop). @Sam Stubbs must not have known what that column was for.

0 ·

Sam Stubbs ♦ Karen Measells commented · Feb 06, 2019 at 09:07 AM

Ah that makes sense! Sorry for leading you astray!

0 ·

Karen Measells Sam Stubbs ♦ commented · Feb 06, 2019 at 12:49 PM

No worries! One of many errors....

0 ·

Karen Measells commented · Feb 05, 2019 at 08:18 PM

I just fixed the column, but now I"m getting a clock overflow error at 8:15am. Does this mean one of my distributions is returning an abnormally large number? Is there a way to check to see which PCI it is?

Thanks!

0 ·

Matthew Gillespie ♦♦ Karen Measells commented · Feb 05, 2019 at 08:21 PM

The easiest way to check is to run the model to 8:14, open the event list, and then click the step button until you see the error. That way you can see which event is causing the error.

0 ·

Matthew Gillespie ♦♦ Karen Measells commented · Feb 05, 2019 at 08:50 PM

PCI 17 is using lognormal(0, 126.7, 0.4, 86) for its Post Time label. That distribution returns enormous numbers.

0 ·

Karen Measells Matthew Gillespie ♦♦ commented · Feb 06, 2019 at 12:51 PM

Matthew-

Thanks for all of this troubleshooting help. My file was corrupted, so I had to recover an older version, complete with mistakes that we'd already fixed!

Karen

0 ·

question