Can I merge FlexSim Model together?

[pert_p]

[ FlexSim 22.1.2 ]

I have 4 Models and I want to combine into FlexSim file for reading on the same global Table and look the changing on 4 difference models. Can I do like that?

[Jeanette_Fullmer · Accepted solution]

Hello @Pert P,

If you have a Global Table that all 4 models will use info from what you could do is create a CSV that loads into the 4 models as a Global Table. When you want to make a change you update the CSV and have it reload when there are changes.

FlexSim cannot merge models together. Some things can be copy and pasted but it is not guaranteed. However if one model relies of the input of another you can create and export data from one model that is then used in another like what is mentioned in this post.

[pert_p]

Thank You for informtion

On global table, I have coded for generate random number on global table. That mean when run replication of each times It will generate a new data into table then the process flow. will read the data then complete process and start a next replicate and begin at global table. I'm not sure that you said "When you want to make a change you update the CSV and have it reload when there are changes." I've to add CVS every times of each replications, right ? or It will change automatically?

By the way, I want to read the same table and each replication will generate as wrote on coding data table. I need to compare result of 4 models

[Jeanette_Fullmer]

Hello @Pert P,

When importing a table from excel into a global table you can select the following so it updates.

In terms of random number generation a stream will generate the same sequence of random number for that stream. The only worry you run into is whether all the objects in your models use the same streams as the other model. This is where you can run into some statistical problems when comparing 4 different models vs comparing one model with 4 scenarios. In the single model each object will have the same number stream that is used for each object in between scenarios. In the 4 different models you do not necessarily know what number streams are used when you compare object between the models. With that being said you cannot compare replication 2 of model 1 with replication 2 of another model. You may be able to run a statistically significant number of runs of each model and compare the aggregation of statistics of each model with each other. I am not a data scientist though so I am not sure I have considered all the problems you need to consider when comparing your statistics.

[pert_p]

Thank You so much. I found that if I coding for all model is the same and I us e repeat Random Streams, It is one way to get the same data for all model too.

[pert_p]

I'm worry about on my global table there is a lots of algorithm such as probability , random etc. Even though I will use the same Random stream number. The Algorithm maybe made the data on global table of each models different. but I need it the same data but I don't want load every run replication to all models