Resource usage in numbers

[tor son]

I have a comprehensive DES in simmer. I want to get information about the usage time of resources. In particular the active time and idle time of each resource. 

This plot creates a good overview.

plot(get_mon_resources(envs[[6]]), metric = "usage", "doctor", items = "server", steps = TRUE)

But I want the numbers of it, so that I can see: activity time = x minutes and idle time= y minutes. 

As I repeat the simulation often, I would like to get the mean time for activity and idle time. It's possible to achieve this in a cumbersome way by doing several filtering and analysing steps of get_mon_resources(). Is there an easier way to calculate the acitivity time and idle time of resources? Some resources even have a higher amount than one.

[Iñaki Ucar]

On Mon, 24 Jan 2022 at 14:18, tor son <tors...@gmail.com> wrote:

I have a comprehensive DES in simmer. I want to get information about the usage time of resources. In particular the active time and idle time of each resource. 

This plot creates a good overview.

plot(get_mon_resources(envs[[6]]), metric = "usage", "doctor", items = "server", steps = TRUE)

But I want the numbers of it, so that I can see: activity time = x minutes and idle time= y minutes. 

Here's the code to that function: https://github.com/r-simmer/simmer.plot/blob/86bef86c7dc5f77320011e27d9eaddac053969fb/R/plot.resources.R#L51.

 

As I repeat the simulation often, I would like to get the mean time for activity and idle time. It's possible to achieve this in a cumbersome way by doing several filtering and analysing steps of get_mon_resources(). Is there an easier way to calculate the acitivity time and idle time of resources? Some resources even have a higher amount than one.

As you can see, the code above deals with multiple replications out of the box. It should be easy to adapt to your particular use case with an additional summarization step.

We strive to provide all the necessary raw data via get_mon_* so that any analysis is possible, but there you are "on your own". In simmer.plot, we provide some quick visualization tools, but they are meant to be just an example of what could be done. Different simulation scenarios require very different analyses, so it would be impossible for us to try to cover them all. That said, if you identify some non-trivial post-processing step that is general enough to be applied in a wide range of use cases, we could consider adding it as a helper somewhere (e.g., in simmer.plot).

--

Iñaki Úcar