How do I display statistics for partitions within a zone?

Question

question

Cameron Pluim asked Jul 12, '16 Adrian Haws edited Jul 14, '16

How do I display statistics for partitions within a zone?

Assume I have Queues A,B,C,D and TEs 1,2,3,4. I can have product arrive at any queue and require any specific TE to transport it.

I'm trying to have a zone collect the following statistics: Content Vs Time (Per Queue, and Per Queue and TE), Min/Max/Average Staytime (Per Queue, and Per Queue and TE).

So lets say Queue A has 4 products, 1 needs TE 1, 2 need TE 3, and 1 needs TE 4. I would need:

Content Vs Time (Queue A)
- Content Vs Time (Queue A TE 1)
- Content Vs Time (Queue A TE 3)
- Content Vs Time (Queue A TE 4)
Min/Max/Average Staytime (Queue A)
- Min/Max/Average Staytime (Queue A TE 1)
- Min/Max/Average Staytime (Queue A TE 3)
- Min/Max/Average Staytime (Queue A TE 4)

Currently in my model I have a fixed resource process flow that is just gathering the statistics for these queues, where they enter a global zone which separates them into subsets based on queue, and partitions based on queue&TE. This is the best way I found to do it so I only have to create subsets for the number of Queues I have in the model, and not for each QueueTE interaction, but I've found that the dashboard statistics don't have a clear way of showing the partition statistics like they do the subset statistics.

Is there a good way to display these Zone Partition statistics, or is there a better way to collect these statistics?

Software Version:

FlexSim 16.1.0

statistics zone partitions

Answer 1 · 2016-07-13T22:03:48Z

Adrian Haws answered Jul 13, '16 Cameron Pluim commented Jul 14, '16

After looking at your model I was able to find a way to generate these charts without using zone partitions or exporting raw data. For now it's not quite as convenient as using the partitions, but you can simply create more subsets specifying each combination.

I've attached a revision of your sample model below. You'll see that there are charts for the content of each Queue, as well as the content of each type in QueueA. I also demonstrated the Minimum, Maximum, and Average Staytime for specific types in QueueA.

Here's an example of defining each label's value in "Token Selection Criteria" for each new subset:

Queue = "QueueA" AND TE = 4

Then in the "Series" tab for the chart properties you can only select the subset(s) that you need in order to display the data.

Here's how the charts look that I've set up so far:

If you still want to export the data after your model run let us know, but this is how you could avoid having to do that with your specific model.

zone-stats-4.png (15.4 KiB)

zone-stats-3.png (42.9 KiB)

1273-zonepartitionstats.fsm (39.8 KiB)

· 1

Cameron Pluim commented · Jul 14, 2016 at 12:54 PM

@Adrian Haws

Unfortunately I was planning on using partitions because I'm looking at 30+ queues and probably around 15 TEs, so if I didn't have to specify each combination that would be ideal. Now granted each TE won't service each queue, but I was hoping to avoid the prework of needing to identify each combination that is needed.

For now it sounds like I'll need to use subsets for each combination, unless there is a way to get the data from the partitions.

0 ·

Answer 2 · 2016-07-13T16:15:56Z

Adrian Haws answered Jul 13, '16 Adrian Haws edited Jul 14, '16

Hi Cameron,

Although you can display list partition statistics I'm not aware of an easy way to do so with zone partitions.

It would help to see how you have your model set up, but I created an example model that I hope works similarly to yours but on a smaller level (although I just used a general process flow). I created a label to represent the four different types of products and made a zone subset for each of these label values. The built-in zone statistics in the dashboard make it really easy to display things like Content vs Time and Staytime. For example, to display the average staytime for each type of product (I called it "itemType" in the example) you can drag a "Zone Calculation" chart into the dashboard and select the Zone as an object. Then in the Properties' "Statistics" tab you add "Staytime" within "Zone Subset Statistics," select it, and change the Value in the window from "Current" to "Average." This might be easier to see in the model I've attached.

If this doesn't answer your question feel free to let us know and we'll do our best to help you.

zone-stats.png (170.8 KiB)

partition-stats-3.fsm (30.3 KiB)

· 3

Regan Blackett ♦ commented · Jul 13, 2016 at 04:24 PM

I brought this to @jordan.johnson's attention a couple of weeks ago, it surprised him that that zone partition chart options were missing so it must have been a mistake in the last release and it should be something we see in a future release.

1 ·

Cameron Pluim commented · Jul 13, 2016 at 08:23 PM

@Adrian Haws

Thanks for your model, and your explanation. I would have attached a model earlier, but I had to recreate it in a sample model in order to share it.

I love the stats that your model shares, but I want to be able to get stat "Zone (Current Content)" or "Zone (Avg Staytime)" for each individual queue.

According to @regan.blackett's comment, the zone partition chart options are missing, and I'm ok with them being missing for now, but I would like to know where I can get them for a data dump at the end, to create my own charts if need be.

I see this data here:

But I would like to be able to dump each individual entry to create the time chart.

0 ·

zonepartitionstats.fsm (31.2 KiB)

zonepartitiontree.jpg (248.8 KiB)

Phil BoBo ♦♦ Cameron Pluim commented · Jul 14, 2016 at 03:19 PM

@Cameron Pluim Those are tracked variable nodes. You can get their value with get(). They aren't tracking history so they don't have all the data in them that you need to create a time chart. If they were tracking history, there would be a sub-node with the historical data on it.

You would need to get their values at different points in time in order to collect a history if you wanted to get the data from there.

You may be able to programmatically add all the combinations you need as subsets if it would take too long to do it via the GUI.

1 ·

question