Search the Community

Hello, I am interested in having routing information read into a model from an Excel spreadsheet. The basic problem is that I have several entity types that must be processed at one of many servers. There are capability restrictions (i.e. EntityA can only be processed at Server1, EntityB can be processed at Servers1, 2, 3, etc). I have a “capability matrix” as input data that is a table in Excel showing where each entity type can be processed. In the Simio model, I am using a transfer node to route each entity to an available (and capable) Server by using the “Select from list” entity destination. My customer is interested in analyzing the impacts of changing capabilities of the servers. Currently, all other inputs to the model are fed in using bound Excel files. However, the routing logic requires lists, which I have to manually create and/or edit. I would like to have these bound to a data table, or somehow reflect the data in the master input Excel file. Thoughts? Thanks, Adam

Yeah, running models from bound Excel files is not seamless. However, even that doesn't do what I'm after (unless I'm missing something). I would need a way to get data from the table (bound to an excel/csv file) into a list. Alternatively, I need a way to get the transfer node (or routing group) to select locations from a table. Would you mind summaraizing your workflow for using the API for input data? Do you build the model in the GUI and create properties for all your input data? What is your typical input data like? Just curious if I could use a similar approach... Thanks! Adam

I have not looked too deeply at this specific problem for a while (there may be some obvious gotchas I have forgotten), however... have you looked at the Write step, and the bind excel sheet to table? You can bind excel sheets to tables, which will solve the feeding problem, and you can write out to csv files and use macros in the main spreadsheet to suck in that data. I have gone off excel sheet DRIVEN models though. The capability within Simio was just not there in making it easier for users. It is painful having to explain to end-users of models that they need to fix the absolute path references to the sheets... and that they must manually import the data into the tables if it caches incorrectly... etc. I only use excel for examining the results now, and for the rest I use experiments via the API and a written custom application.

Simple approach: Take advantage of the fact that each model has an OnRunInitialized process that is called at the start of each run. --Create a table that contains all your initialization data. This might optionally be bound to an external file and automatically read if it changes frequently. --Define the OnRunInitialized process to search that table, create the entity(s), initialize the entity, and transfer it to the correct location. Look at the Simio Example (not SimBit) named RPsixample for an example of this approach applied to a simple scheduling model. More comprehensive approach: Take advantage of the fact that each object has an On Initialized add-in process that is called at the start of each run. --Create a table that contains all your initialization data identified by object instance. This might optionally be bound to an external file and automatically read if it changes frequently. --Define the On Initialized process to search that table for a matching object instance, create the entity(s), initialize the entity, and transfer it to the correct location. This can also set object characteristics as well, such as status, capacity, failure data, learning curve, … While this could be used by simply specifying the add-on process in each object instance, it is even more powerful if you create a custom object that includes a custom object-specific initialization process.

I have a source (exp(1)) --> Server (exp(1/1.2)) --> sink. This way, the Server gets a load of lambda=1 and serves at a rate of mu=1.2, giving a util of rho=1/1.2 = 0.833.. Endtime is set to 1 million seconds, so nbr. of jobs to the server is 1 million. Simio computes a rho of .833, so the sim is correctly set up, I believe. But, to run this takes me 115 seconds in Simio (I clicked on Fast-Forward, Trace is turned off). Are there ways to make Simio run this any faster? (I have searched the Intro PDF and the Reference PDF for things such as "fast-forward", "fastest" ... but found nothing new) I have an average Windows-7 64 bit Dual Core E8400 @3ghz, 4GB Ram Thanks. --Ketil

Today I had to do a quick and dirty model in Arena. I just thought I would take this opportunity to do a quick post on my findings. By no means is this a comprehensive analysis of the two tools, but I just felt like sharing my opinion of the two. Being the predecessor to Simio, it is natural to compare the two tools, and even after using Arena for the short amount of time I did... the differences were astounding. I will admit that Simio was my first immersion into the world of using corporate modelling tools. It was a tough journey to get to the stage where I felt comfortable using Simio, but looking back... I wonder how I got by using the archaic tools I did. The learning curve to Simio was extremely steep (in my humble opinion that is), and I still don't believe I am anywhere near utilizing 100% of Simio's capabilities. That being said... Simio is a relatively new tool and is still under active development. After launching Arena for the first time, the first thing that struck me was how bland/simple it was. Putting that to one side, I was impressed with how intuitive it was to use. This may be a side effect of using Simio, but barring my inability to find the Connect button for linking the objects together, I simply dragged objects into the process view and linked them up. The objects all had names that made sense (workstation... server... what were you guys thinking? >_>), and ignoring the fact that most of the symbols looked like they were drawn by pre-schoolers... the process of getting a simple model running was child's play. Within 5minutes I had my first working model. (I can tell you that it took a lot longer than that for my first Simio experience) One of the features I would love to see the most in Simio... AVI export. It was right there, and it was easy. Come on Simio! After placing multiples of an object, there was a table at the bottom of the screen with every other object of the same type, allowing you to quickly locate and edit properties of said objects... Simio... adding something like this would immeasurably improve productivity (mine atleast) and increase my levels of sanity after working with large models. Another point to note was that Arena felt very responsive, and while this normally would not be an achievement for a piece of software... Arena did not crash or give cryptic errors. (Simio crashed multiple times during my first 30minute experience with it... and I didn't even know how to do anything more advanced with it at that stage). The last point to make in this section of the post, is how human readable the reports were. My eyes were not assaulted with a million numbers. You could argue this is a bad thing or a good thing, but at the end of the day, some of these reports matched what I typically look for in Simio, and have to waste time trying to get. They were also laid out clearly and concisely. I am sure Arena has a fairly similar learning curve to Simio once you get to the more advanced stuff, but I have got to say that Arena definitly has the edge of many years of polish and tweaking. I am not here though to plug Arena. Let's get to the meat of this post now. The things I saw in my short time using Arena in comparison to Simio, that Simio did better. Even though I feel Simio has a lot more work to do on getting their product to the kind of a level Arena sits at (polish and usability) in terms of: aesthetics, power, time-savings (gained from re-usability of objects in Simio, instead of hard coded processes in Arena), feature base, and on-going development... Simio wins hands down. The benefits of object based models in comparison to process based models are imeasureable, and while some of the more basic features you would expect (or take for granted) in a modelling tool are lacking or missing in Simio, a lot of thought and has gone into the development of this tool. You can really see how Simio is the next generation of tool. I guess the point I am really trying to convey is, Simio rocks. It may be buggy as hell, it may not have every bell and whistle to satisfy my unreasonable demands, and it may not be as smooth or intuitive as Arena, but it is actively developed and a lot of love and care went into its development. I dare say that given time and a bigger/more active community, Simio will truly come into its own. The Arena community may be larger, but it does not appear as if Arena itself is evolving to meet the ever changing demands of the market. I guess it won't be long until the remaining Arena modellers move on and embrace newer technologies like Simio. Sorry for the long post. TLDR: Keep up the good work guys!