Jump to content

Welcome to the Simio Forum

Welcome to the Simio Forum! Experience the full experience of the forum by becoming a Simio Insider!

Click_Here_button.png

New Software Release

Our latest release, Simio 10.174, will be live in July 2018!

Save the Date!

Simio Sync is May 14-15, 2018! More details coming next month.

hugo

Members
  • Content count

    42
  • Joined

  • Last visited

Community Reputation

0 Neutral

About hugo

Converted

  1. Hi Glen, I have been using the DBWrite step in my current model with MySQL. It works like a charm, but it slows the model down considerably. (Can take almost 50% to 70% more time) Is there any reason why it is slowing the model down so much? Any work around suggestions? Thank you and regards, H
  2. hugo

    Updating values in Data Tables

    Hi, Can any of these methods be used to write time series data during the simulation run and then at the en of the simulation write it into a database? Thanks, H
  3. hugo

    Categories for States

    Hi, I must agree. Being able to categorise states and Elements would be great! Thanks
  4. Thanks for all your input! Amy it is as you said, it is the last part of the tail or the negative aspect of the distribution that give you errors within your model. I used the re-sampling method in my case. I however constructed my own library of object for the project I am currently working on, so I have complete flexibility in how I wanted to model the failure's MTBF and MTTR distributions. In addition I had groups of objects which were dependant on one another's failures. Thus when an object fails there are a certain set of other objects which needs to fail too, and vice versa in the repair of objects. The only problem I did however run into was that a Timer element in Simio initial time before firing an event is before I can even do my own iteration sample of a distribution. Had to build my own timer with a delay and a fire step, but works very well. Dave, in my case the best alternative would be to use the actual data instead and sample randomly from that, rather than defining a continues function. Like Amy said it would get very long to define. But non the less, all these alternative needs to be kept in the back of all our minds for the day we can use them. Once again thanks for all your inputs. H
  5. Hi Dave, thanks for your input on the matter. I agree with you that truncating a distribution is not perfect. Perhaps the example I stated above was a bit harsh truncating the normal distribution, but it was only to show the impact of truncating incorrectly. I usually only truncate from the 95th, 98th percentile. Thank you for revering to the new version of Simio and its input distribution capability, will take a look at it, currently busy with a project and I tend to focus on the project more than on the newest updates. Regards, H
  6. Hi GreerAmy, I have tried using just the expression to achieve this. The problem is you have to re-sample, otherwise you are creating a new distribution which will be skewed at your cut-off points. The first image is the distribution defined as Random.Normal(50,15) with no constraints. The second image is an example of when the Min and Max approach is taken. It unfortunately gives a very skew distribution. Truncated to be below 70. The third image is re-sampled to ensure that is is within the bound set. Truncated to be below 70. My suggestion is to have a function in Simio to allow re-sampling within the expression. For example: Random.Truncation(Distribution ,Lower Constraint, Upper Constraint) I fully agree that one needs to be careful, thus the reason for the images of the distributions. But for example, an exponential distribution does not end, it goes until infinity, with a very small likely hood, but it still does. In my case my MTTR will never be over 80 hours. But sometimes the exponential distribution representing the MTTR samples 1.345x10^5 hours. That just is not possible. Thanks for your help though.
  7. Hi all, I have a problem in truncating some distributions within Simio. I usually use Palisades @Risk to fit distributions to data, but distributions such as Exponential etc. sometimes needs to be truncated. Now I know this is not possible at this moment in Simio, but can it be done in some other way to let Simio re-sample within certain bounds. Now the answer to that is to build a process to re-sample until the distribution is in the right range of the distribution, but how do you do that if your distribution is used in a Time Element? I may be just missing something simple. Can truncation perhaps be implemented in a future sprint? Thank you.
  8. Hi all, Made a mistake. The right code is: String.Substring(DateTime.ToString(TimeNow), 7, 4) + "-" + String.Substring(DateTime.ToString(TimeNow), 1, 2) + "-" + String.Substring(DateTime.ToString(TimeNow), 4, 2) + " " + String.Substring(String.FromDateTime(TimeNow), 12, Antecedently swapped the Date and Months around. H
  9. Hi all, Thanks to those for showing the building blocks of Date Time conversions. Here is an additional expression. If you need to import a Time and Date into MySQL the following expression will convert the Simio Date Time format to the MySQL format. String.Substring(DateTime.ToString(TimeNow), 7, 4) + "-" + String.Substring(DateTime.ToString(TimeNow), 4, 2) + "-" + String.Substring(DateTime.ToString(TimeNow), 1, 2) + " " + String.Substring(String.FromDateTime(TimeNow), 12, {See corrected posting below} Hope it helps. Regards, Hugo
  10. Hi Dave, Thank you for your reply. That makes sense. I will try your suggestion and get back to you.
  11. Good day all, I don't know if my question has been discussed in a previous post or not, did try to search for something similar. When placing a sub-model into a model it is possible to update the sub-model's internal process logic etc. However any animations updated in the Sub-model's external view is not updated to the sub-model previously placed in another model. Is there a way that the external view can be updated? This would greatly reduce the time to each time replace sub-model into the model just to update the visuals/ external view. I am probably just missing something small. Any help will be appreciated. Thank you and regards, Hugo
  12. Hi, Thank you for reading this. I have 2 tables, Table1 is the primary table of Table2 with Table 2 containing the Foreign Key relating to Table1. Now within a process I would like to get a couple of node reference out of a few rows within Table2, but narrow my field down by using the already in place relationship with Table1. Thus, SetRow in the process for Table1 and then select the first (relationship row) of Table 2. Then from there on Row 2 of Table2, but still via the relationship etc. The rows relating to Table1 does not necessarily follow on one another. How does one accomplish this? Thank you Regards Hugo de Vos
  13. Hi, Thanks for your comment. I experienced almost the same as you described. I can get the vehicle through the sub-model, but when it excited one cannot change the ModelEntity or the Vehicles destinations. So the the vehicle just circles and circles... I think Simio should be able to make the connection through a sub-model just as it does when all the paths are in a single level. A vehicle is just a more complicated Entity, so I believe it is possible. I'm going to play around with the model a bit and upload an updated one.. Thanks again for your response. H
  14. Good day all, I need a vehicle to pick-up an entity in the "Model" and then transport the entity through the submodel. I have attached a example of what I need doing (not working example). This is just to test the concept, there will be other logic in the submodel, my first mission is to get the transporter and entity in the submodel. Now I know this is possible with other submodels such as the server, were a vehicle can be processed by the server (which is essentially the same as what I am trying to achieve). Please assist if you can. Regards. H Vehicle into submodel.spfx Vehicle into submodel.spfx
  15. hugo

    Reference a vector state in Assign step

    Sorry, Found the problem. One needs to first add the model's name before the state. For Example: Table1.ColumnName.ObjectName.VectorState[VectorRow] Thanks H
×