Advertisement
Guest User

pace-project

a guest
Apr 26th, 2015
221
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 2.03 KB | None | 0 0
  1. This looks great. I will read the paper soon. We want to use a Poisson distribution for the arrival process. The next step is to build a simple simulation, where the time clicks each day. So, for each day, we figure out how many people are in the hospital, how many in the ICU, and how many were turned away from the ICU. I would like to see this simulation as a graphic, and with numbers cvs formate. The graphic can be simple, just a simple table, where you can see which rooms have patients, and which do not. Each day, the clock clicks, and you get a new population in the hospital, which is based on how many patients check out, and how many come.
  2.  
  3. -------------------------------------------------------------------------------------------------------------------------------
  4.  
  5. Thank you for the reply and for the suggestions.
  6.  
  7. I tried to clean up the data to reflect what you’ve suggested.
  8. For example, I simulate patients' admissions into a hospital for a period of time (e.g. two months March and April 2015).
  9. The timing is days; so multiple patients can be admitted on the same day regardless of the time. But each with a different length of stay.
  10. Here is a link to the simulated dataset with the plots as of now.
  11. Also, I found a paper that uses Poisson distribution for a queuing system in healthcare. It looks very related to what we are supposed to do.
  12. So for now, I am exploring the poisson probability so I can use it to visualize and explore the dataset.
  13. Please let me know what you think.
  14.  
  15. -------------------------------------------------------------------------------------------------------------------------------
  16.  
  17. Looks like a good start. Here are a few suggestions:
  18.  
  19. For the data lets start with using days as the timing, so for each day, we generate a list of patients, along with how long they stay in the hospital. Lets assume that everybody is admitted, but that 10% go into the IC rooms. We dont need age, or responce time.
  20.  
  21. The next thing I would like to see if a graphical display of the queuing system.
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement