内蒙古体彩十一选五下载:Simulation Data Analytics
Whenever software like this is run, it computes with great accuracy how the designed model behaves according to some specific test situation. The output, in the background, is usually a large amount of data produced by the high-quality simulation engine. However, it is difficult for engineers to take advantage of these rich datasets full of raw data (typically, hundreds of thousands of time series). They usually end up checking some features they have learned are important from their personal experience and skip any deep review of the raw datasets.
Yet, when decisions impacting the design or production of any complex system are taken on the basis of such simulations, it is of the utmost importance to ensure that no hidden defect could possibly invalidate the model or the results of the simulation. The analysis of the logs of the simulation software, combined with a fine grained investigation of the simulation engine raw data, allow, along with INENDI, to quickly find those anomalies and prevent engineers from making less than optimal decisions.
In addition, data-driven (meaning no a?priori hypothesis) approaches,?such as those provided by the Machine Learning algorithms in MINESET, can offer?alternatives to the traditional way of?looking at the data. MINESET,?ESI's?easy-to-use, integrated visual analytics tool can provide simulation engineers with the following benefits:
- quick validation of the initial hypothesis, using a non-scripting-required method for charting and plotting all variables?
- discovery of patterns inherent in?the data, using unsupervised methods, such as Clustering and Associations
- guided analytics tour of?the factors that lead to instabilities and extreme results
- quantified method of exploring?the effects of individual input factors