Conditional Initial Routing Made Easy

Conditional initial routing

welcome to this module of the simulation video lab series in this module we're going to talk about different entity routing methods and how to implement them in simio so in terms of the objectives particularly we're going to talk about probabilistic routing conditional routing and we will use a selection weights and the links to do these two types of routing we will also talk about compound expressions that we can use in delinquents or the selection weights on our links to support both probabilistic and conditional routing we will also talk about dynamic routing so we can make decisions as the model is running and also we will talk about routing with sequences so we are going to talk about these different types of entity routing in the context of an example that is based on the operation of the airport's security checkpoint so as most of you are familiar with this system we the airport security checkpoint has basically two main steps the first one is a document check where TSA checks your dock and your ticket and your ID and then it will go through the advanced imaging machine which which they call AIT that stands for advanced imaging technology which is basically an x-ray machine so also note that there is another option that is called TSA PreCheck which is a program that you can participate in if you register for it and then if you're accepted there will allow you to leave your shoes and jackets on and also you don't have to take your laptop of your backpack so as you can see we have in this model we assume that passengers arrive at a rate of 720 per hour into this security checkpoint system and 90% of the passengers are irregular people that go through the regular screening equipment and also note that we have this possibility of 10% of failing the screening process which most of the times happen because people forget to take everything out of their pockets or they just don't stand the right way when when the machine is scanning so they have to repeat the process and go through the Machine again and finally we have 10% of the passengers that are really are eligible for pre check so let's go to simyo and start building this model and what I need is my source object and my three server objects and finally my sync object I'm going to connect these up with paths I'll have a switch to multi placement mode and I also have this reentrant flow back into the screening server so now I have my servers and and I have my paths set up let me put my model entity or default entity in and the screen and the facility view and I'm going to rename my object so I have passengers so the first one is document check doc check and then we have AIT and we have pre check so now I'm just going to hit run just to make sure that everything is connected and my entities flow in the system so the next step I'm going to do is implement my probabilistic routing logic so as you can see I have ninety percent of my passengers that go from the first stage to the AIT server to the advanced imaging technology server so I'm going to have my selection wait for the path that goes from doc check to AIT set to point nine and the alternative will be point one and I also have a possibility of ten percent of failing the screening process here so on this path I'm going to use a selection weight of 0.1 and the alternative is 0.9 so this is something that we have seen in the previous labs and the most of you are already familiar with using selection waits for probabilistic routing let me just make this note that it really doesn't matter what value you put in the selection weight as long as the proportions are right your your routing logic will we'll do what you wanted to do so right now I have 0.9 and point 1 on these two paths objects I can also have 90 and 10 or 9 and 1 so again as long as the proportions are right simio will route the entities according to 2 to the proportions that you put in so let me go ahead and put in my inter arrival time so I know my arrival rate is 720 per hour so I'm just going to let simio calculate my inter arrival time so I'm going to put in 1 over 720 and I'm going to set the units to hours and also I have different and units of capacity and my server so I have three components of capacity or 3 TSA officers that check the IDs and tickets and I have three AIT equipment so I'm going to change the capacity of my AIT to three also and I will leave the capacity of the pre-check to one also note that in this lab of focus is more on the implementation of different types of routing and we're not going to really do any analysis with with the models that we developed and as you can see we will leave most of the parameters of the model to their default values just to save some time but of course when you're doing this model as part of a project you you will need to verify your model as you add more features to it so now if you would like to verify this a model that we currently have what you need to do is develop a static model a queueing model of this and then adjust the parameters of your simulation model so that it corresponds to your QA assumptions and then you will need to run an experiment to run your model for a long time and warm-up period and replicate your model in order to be able to compare the results of your simulation model with your queueing model to gain confidence that your probabilistic routing is implemented correctly but we're not going to do these steps for the models that we develop in this lab so the first model that implements only probabilistic routing was easy and we have seen an examples of it before in the second model we are going to add a new component to our model which is a pat-down station and the way the pat-down station works is that if an entity or a passenger fails the screening the x-ray twice the passenger is still checked or screened using the x-ray machine for the third time but regardless of what the outcome of that screening is the the passenger will also need to go through pat-down so again if if a passenger fails the screening twice two times the passenger is still still needs to go through the through the x-ray machine for the third time but regardless of whether the passenger passes or fails the test or the screening the passenger will will need to go through the pat-down as well so let's go to simio and see how we can implement this logic the first thing I need is my server object which is my pat-down station let me sink let me move my sink object a little bit so I'm going to connect AIT to my new server and server to say and also I'm going to rename my sink to pan down and change its capacity to two and as you can see I have this conditional routing that sends entities to the pat down station if the entity has failed the screening two times so when you think about it what we really need is a mechanism to count the number of times that each entity goes through AIT basically what we need is a model entities state a state or an attribute that each passenger carries with itself so I'm going to go to model entity definitions States and I'm going to define an a new integer state and I'm going to call it num screened so now when an entity is created it has this attribute associated with it and that initially has a value of zero stored in it so now I go to the model and the next thing that I need to implement is the logic to increment the value of the NAM screen for each entity when the entity goes through the AIT so I'm going to do this by incrementing the value when the entity enters the AIT station and I'm going to use a state assignment process also recall that and we can we can also use add-on processes to do state assignments so I'm going to change or increment the value of my model entity num screened I'm going to increment it by 1 whenever the entity enters the AIA AIT server so when the entity enters the AIT from dark check it becomes wine so remember the default value was 0 and then if it fails for the first time it becomes 2 and it the entity fails the screening for the second time the value becomes 3 so what I need to tell simio to do is send the entity if the value of model entity dot naam screened is equal to 3 so I'm going to use my selection way to implement this condition and the expression I need is model entity Nam screened equal to 3 and I also need to make modifications to the selection weights for my other two paths so what I'm going to do is multiply 0.9 by model entity dot num screened less than 3 and I'll explain shortly why we do this and I'm going to do the same thing for my other path that goes back to AIT so it's just paste here so now let's see what happens when an entity is about to be transferred out of the output node of AIT so if model entity num screened value is less than less than 3 that means the entity is processed and AIT either 1 or 2 times then the a this expression right here becomes 1 so my link weight or selection weight essentially becomes 0.1 so the entity will be rounded back to the AIT with a probability of 0.1 similarly if the entity is processed less than three times on AIT then this expression will become 1 and my selection weight will become 0.1 so the entity will be mounted to the sink with the probability of 0.9 otherwise if the entity has already failed the the the AIT is screening process twice and it's been processed on AIT three times then the entity will be then this expression will become true and the the entity will will be routed to pat-down so in order to verify that this logic actually works I'm going to increase the size of my passenger model entity instance a little bit and now what I would specifically like to do here is change the color of my or the symbol of my passenger based on the number of times that the end model entities are processed on AIT so what I'm going to need is two additional symbols so for my second symbol I'm going to use light the red color I'm going to apply it to my passenger model entity instance and then I'm going to select my third symbol and I'm going to use a dark red two and represent the passengers that need to go to pat-down so now as you can see I have my symbols defined so those that do not fail the AIT have the default green color those that fail once have this this color and then those that fail twice has this red color the last thing I need to do is to implement the logic to change the model entity that picture state so I'm going to do this on the path and I would like to use a state assignment and the state that I would like to change is model entity that picture right here and I would like to increment its value by one so now when the entity fails for the first time model entity that picture for that entity becomes one so I will be using the second item or color in my symbol list and when it fails for the second time this model ends in that picture state becomes two and I will be using the third symbol in my symbol list so now let's run the model and see what happens so as you can see if they fail for the first time they they're represented with a light green a light red color and those that fail more than once hopefully we'll see one of them soon will be changed to a dark red color and will be routed to the pat-down station so let me increase my run speed a little bit to see if we see an entity that fails twice and as you can see at some points will see some entities that have a dark red color that go through the pat-down station