![]() |
|
|||||||
| Simulation Analysis simulation software (Arena, AutoMod, Enterprise Dynamics, ExtendSim, PlantSim, ProModel, SIMUL8, WITNESS), input and output analysis, experimental design, optimization, simulation model verification and validation |
![]() |
|
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
|
|||
|
Hi everyone,
due to a large number of user-specified output data my Crystal reports literally explode since all replications of my simulation are listed there separatly. Is it possible to let Crystal present only the summary statistics at the end of the report? I couldn't find any settings for this. Thanks a lot in advance. Paul |
|
|||
|
I always get following error when displaying the category overview of the Crystal report:
Not supported. DAO Error Code: 0xc6f DAO.Database Not enough space on temporary disk. I've tried it several times on two differnet computers: 2.5 GHz, 2.0 GB RAM and 2.8 GHz, 8.0(!) GB RAM Especially at the latter one, there is plenty of space on the hard drive (>200 GB). Anyone an idea, why the reports don't work? Or does anyone have an idea, how to get the summary data otherwise? The SIMAN Summary Report (.out file) cannot be used since the number of letters per output statistic is limited there, which is why I'm not able to distinct between various statistics. Thanks a lot in advance. Paul |
|
|||
|
Just in case nobody has an answer to my problem description above, I probably would need to run my simulation with considerably less replications than orginially planned. That would be somewhere between 100 and 500, depending on when Crystal being able to generate the summary report.
In that case I'd like to compare my performance measueres on consistency by running the same model several times. Since Arena applies pseudo-random numbers, the random number generator (RNG) would always generate the same random number streams... Is there anything like the SEEDS element for the current RNG, so that I would not need to add the stream number in the distribution's parameters but could use them for all distributions? That would be much work and isn't possible everywhere, as in the case of arrival schedules or "by chance"-decide modules! Thanks again. Paul |
![]() |
| Thread Tools | Search this Thread |
| Display Modes | |
|
|