this is the second video in the introduction to Simo module in the first video we built our initial uh Sim model of our uh two-station flow line and you can see here let me just go ahead and and click run here and you can see that the source object creates entities uh the two servers process those entities and then the sync tabulates results uh and destroys those entities we also looked at the results so if I stop and go to the results tab we see that pivot grid and then we demonstrated how we can filter this pivot grid to highlight the particular metrix that we want in this second video we'll pick up where we left off and we will learn how to specify object properties so that we can tell the model or Implement in the model the arrival process for the source and the two service processes uh for the server and once we have the object property specified we'll introduce the notion of simio experiment here we can see the initial information about the arrival process and the two service processes so we're going to assume that customers arrive at the rate of 25 per hour service at server one takes 2 minutes on average service at server 2 takes 1.71 minutes on average so at this point it makes sense to do a quick reality check here and if we think about the service rate at server one if there's 60 minutes per hour and each uh entity or each customer takes 2 minutes the nominal service rate is 30 per hour similarly the nominal service rate for service 2 is 35 per hour and since the arrival rate to the system 25 is less than the minimum service rate of the two servers we know that the uh system can handle the arrivals and so in other words we know that our system will be nominally stable this is very useful information because it gives me some idea of what I expect to see once I add the property values and so in this case what I would expect to see is that the servers can keep up with the arrival rate adding some detail to our arrival and service processes we're going to assume that the inal times and the service times are exponentially distributed so for our customer arrival process we assume that the in Ral times are exponentially distributed with a mean of 2.4 minutes for Server one exponentially distributed with a mean of 2 minutes and finally server two exponentially distributed with a mean of 1.71 minutes so we need a way to sample a random variant from an exponential distribution with a provided mean either in this case 2.4 minutes 2 minutes or 1.71 minutes in simio this expression is random. exponential with the mean value X and so as we'll see building Simo models is all about seio Expressions understanding how to interpret semio expressions and under understanding how to construct SEO Expressions but before we look at these expressions in a little bit more detail let's use the Expressions that we know and apply to the model that we've been working on back in our model here and the basic idea is that we're going to set the property values for the object instances to the Expressions that we just saw so for example when I go to the source object Source One and over here on the right is my property window the property that I want to set is enter arrival time so you can see that the default value is random. exponential point2 0.25 I need to set this to 2.4 and so you can see when I do that there's a unit um property here and so we've just set the in Ral time property for the source object instance to the expression that we saw in the previous slide note that the source object and all of the objects in the standard Library have lots of different property values that you can set for each individual instance and it's setting these property values that fine-tune the behavior that's ultimately defined in the object definition so I place a source object uh in the facility view to create the instance and I know what that does is it creates entities but setting the properties is what lets you specify exactly how that happens and so in our case we're saying we want exponentially distributed in Ral times with a mean of 2.4 minutes so in this video module we are just going to zero in and focus on those properties that we use but in the modules that deal with the standard Library objects we will do a much deeper dive and look at many of the properties for all of the standard Library objects for our server object the processing time property is what we want to set and you can see the default value here it's a triangular distribution and so instead of just typing the value in for this one I'm going to use the Expression Builder so I just highlighted the property value I'm going to click on the little down arrow here and this brings up our Expression Builder which helps you build expressions and understand expressions and what happens is when I start typing it jumps to the location where it thinks you want to go based on what you've typed so I typed RA in and it immediately jumped to random I can now just hit the period and once I hit the period it then opens up the submenu so in this case I have the submenu of the random uh keyword and so now I know that I want the exponential and I can just jump directly to the exponential and now you can see that it tells me it's expecting a parameter the mean value so for this one it's 2 minutes so we have random. exponential 2 for Server one and then for server two I'll just type this one in random exponential 1.71 also in in minutes so either type the value in if you know it or type the expression in if you know it or use the Expression Builder we'll come back to the Expression Builder it turns out that the it's it's quite a useful tool for as I said learning expressions and building Expressions now that we have the object property value set let's run our modified model so I'm going to go back to the Run ribbon and just click run and let me increase the speed Factor a little bit and we can see the behavior is not any different than what we saw in the previous model just we just changed those in Ral time and and the service times for the servers and so I can just fast forward to the end and you can see if if you remember that we're just using the default run length of 24 hours if I want to see the results I can stop the model and then go back to our results in this case I'm showing the the filter results I can do that or I can always uh reset um to the default view and also note here that if I look at my observations I can do a little bit of model verification I would expect there to be around 600 entities that flow through our system because we have an arrival rate of 25 per hour and we ran for 24 hours so that would be 600 and because our system is stable we would expect everything that enters the system to exit the system and so this 600 uh gives me um at least some indication that our model has been implemented correctly when we started this modeling exercise I mentioned that creating Sim models in interactive mode consisted of dragging objects from the standard Library into the facility view connecting those objects and then setting the property values and that's exactly what we did so remember that we added the source the two servers the sync and the default entity we drag those from libraries onto the facility view connected those and then set the property values in this case we set the in arrival time property for the source and the processing time properties for the two servers and that's all we had to do to create this initial model before we talk about seio experiments I want to spend just a couple of more minutes talking about semio Expressions here are the three property values that we set in our previous model we set the in arrival time for the source and the processing time for the two servers the expression that we use random. exponential then the mean of the exponential distribution uses the random keyword and the corresponding distribution Simo offers large number of distributions you can see all of them in this help reference that I'm giving and what I've shown here are a couple of commonly used or a few commonly used uh distributions the uniform distribution triangular distribution exponential distribution which is what we saw normal and then log normal so again I would encourage you to have a look at this help reference and you can see all of the distribution that that are built into Simo and accessible through the random keyword the last expression related topic that I want to cover in this video is using the expression editor or the Expression Builder we saw this briefly when we were adding property values to our model and basically you can use the Expression Builder anytime simio is waiting for you to enter an expression and you see that when you click the mouse in the expression Val in the property value and you see the little down arrow when I click the down arrow it brings up the expression editor and you can see the notation here uh if you see the double greater than sign to the right of an item that means there is a subl list so we saw this when we used random so we typed random saw the double arrows and hit the period and it pulled up the the submenu If an item is bold like you see default entity and entity and false and Global here that means it returns a value so it's valid to just hit enter and that will return a value for expression and finally if you hover the mouse over an item it will bring up a tool tip that will give you a brief description of what that item is and so we will use the Expression Builder over and over throughout all of our video modules uh to help us build and interpret seio Expressions SEO experiments is the last topic for this video before we do that though we have to talk a little bit about the Run ribbon so we've seen the Run ribbon before where we use to control the interactive run we've seen the Run button and the fast forward button we took a look at the starting and ending time and the speed factor and so what we want to do for our model is we want to uh set the ending time instead of being an ending time we want it to be a run length and we want to set it to 20 240 hours we want to run in Fast Forward mode and then we want to uh go to the um to the pivot grid so let's jump into our model and go back to the Run ribbon and then set the ending type again it's set to a specific ending time this was the default setting instead of that I want to have a fixed run length of 240 hours and then our task was to run in Fast Forward mode and then look at the pivot grid if you can look down here at the semio clock at the simulation clock I'm sorry you can see that it we are in fact at 240 hours so now if I stop the model and go look at our Run results one thing to note here is our observations so we see it's 597 so we would expect this to be around 6,000 because we ran for 240 hours or 10 10 times what we ran previously so again a little bit of evidence that our model is working as we designed it so I'm going to change the view and go back to my uh queuing metric View and I'm going to add that observation uh value so I'm going to go back to my statistic filter the little funnel shape and then scroll down and pick observations click okay and now we see we brought that observation the number of observations into our named view to save that I go back to ad View and just Cho choose the name that I already have it's going to say you want to overwrite it and in this case I do so the reason I did that is because I want to go back to my facility View and reset my model and then run again in Fast Forward mode so if we do that and then we go back and look at our results we can see 597 and if you go back and look at all the individual statistics you'll find that we found the exact same thing and if we run again we will see the exact same thing again so I go back and reset I run in Fast Forward mode I go look at my results and lo and behold we see the exact same thing again so this is a bit unusual if you're thinking about using random variates so we have you know random inter arrival times and random service times but we're always ending up with the same thing this is something that we refer to as predictable Randomness and so the predictable Randomness while it seems a little bit unusual is actually by Design this is something that we want the reason this happens is because simote starts the random number generator it's the same place every time when you run an interactive mode unless you tell it to do otherwise we will explore this notion of predictable Randomness Sim's random number streams and how you can control those streams in more detail in a future video but for now what we want to do is we want to talk about simio experiments now we know that because we're using random variates in our model for the interal times and the service times that we have to replicate our model in order to have statistically valid results in order to be able to interpret our results correctly that is so that default way to do that in simeio is by using anmo experiment so I'm going to go back to the project home ribbon and click on new experiment and you can see over in the navigation window our experiment that we just created and our interest here is in the required number of replications we're just going to use all of the default values as with other things related to seio there are lots of different property settings and we will explore these in detail when we get to the a future video but for now we're just going to run in its default mode of 10 replications and so it runs you can see it's running our 10 replications and once it completes we can then look at the results of those 10 independent replications so I go to the pivot grid and we see the results are very similar to what we saw when we ran in interactive mode except in addition to our single observation we now have the minimum the maximum and the half width and instead of that single observation value we have the numerical average of the 10 replications also note that we can uh use our filter so there's our queing metric filter uh to see the um the descriptive statistics from our 10 replications now we have a completed model that runs 10 independent replications of that model so that's the end of this video in the next video we'll pick up here and show how we can use the experiment for interpretation of the results