This Website is not fully compatible with Internet Explorer.
For a more complete and secure browsing experience please consider using Microsoft Edge, Firefox, or Chrome

The Basics of Successful Simulation Data Management, Part 2: Practical Automation of Simulation Activities


Abstract


One of the major opportunities for a simulation data management (SDM) system is that the activities for routine simulation applications can be automated, which can significantly reduce effort and improve quality. Indeed, this can lend itself to a template driven workflow approach that, once developed, can enable simulation analysis to be accessible and used by less experienced users ? so called democratization. For successful automation, the SDM data structure must be flexible enough to evolve with the requirements of the analysis, and the metadata must be easily retrieved in a useful form. This presentation demonstrates a simple SDM tool in the context of an automated CFD dispersion study. Specifically, it covers the following aspects: 1) Democratization through a template driven workflow approach, as part of an integrated SDM system. Democratization does not mean a free for all. With a template driven approach a workflow can be robustly designed and rigorously tested by expert users, such that less experienced users can subsequently undertake analysis using the template. This has the benefit that the work is unsupervised at the point of use, but since all simulation data is recorded within the SDM system, every simulation can be easily checked and scrutinized to the same degree as if the experienced user had undertaken the simulation themselves. 2) Deviations from the workflow. Regardless of how well a template may have been designed, it is always possible (or even probable) that in some instances there will need to be some deviation from the intended workflow. By having a flexible data structure that can capture such deviations, intermediate users are empowered to modify the workflow and, since this deviation is captured in the SDM system, there is a trigger for this to be brought to the attention of the experienced user/owner of the workflow for discussion and future improvement of the workflow. 3) Automated reporting. Manual reporting of simulation predictions is often a laborious, repetitive task involving the copying and pasting of simulation plots into a report to be delivered to the end client. Such repetitive tasks are well executed with an automated reporting tool as part of an SDM system. The SDM tool is programmed within Microsoft Office, and is therefore easily accessible to many simulation users.

Document Details

ReferenceNWC21-554-c
AuthorHowell. S
LanguageEnglish
TypePresentation Recording
Date 28th October 2021
OrganisationAbercus
RegionGlobal

Download


Back to Previous Page