Stories from an AI company — Empower Operations

This article will provide a brief history of the evolution of design optimization technologies. The content covers mathematical programming, metaheuristic optimization, surrogate-based optimization, and today’s AI-driven optimization techniques. It will introduce the ideas and methods behind the so-called “intelligent optimization” that are driven by AI and machine learning technologies, its current applications, and potential applications in the field of design and manufacturing. The author foresees a third wave of revolution in design technologies, mostly enabled by intelligent optimization. This revolution will liberate designers from tedious and time-consuming trial-and-errors and empower them with the AI learning tools in arriving at the best design. From this perspective, Intelligent CAD (ICAD) is happening and will be the dominant technology in the near future.

Please click full article to access the full article on Pages 9-10.

Volition 1.0: OASIS Optimization API

OASIS now fully supports the volition optimization API at version 1.0. This API allows users with enabled simulation software to integrate with OASIS more easily. The Volition API is for users and software vendors looking to integrate world-leading optimization capabilities into their software platforms. To get started with the volition API, see github.com/EmpowerOperations/Volition

Babel 15: Multi-line & Temporary Variables

OASIS Babel math expressions now support multiple lines and the use of temporary variables.

With this new capability users can more compactly express redundant work, making integration easier.

LGO 2021: Better Discrete Variable Support

LGO has been updated to run more quickly with discrete variable setups. LGO is now capable of running larger discrete variable setups with less redundant work resulting in modest to substantial performance improvements in both workloads with exclusively discrete variables and workloads with mixed discrete and continuous variables.

New Optimization Settings

The Optimization settings screen has been updated to reduce the chance of error and more simply express the options available to users for their optimizations.

Reproducible Optimization Runs

OASIS now takes an explicit “Random Seed” that can be used to create reproducible optimization runs. This is helpful for use-cases where the optimization must be auditable and re-executed on multiple machines or the same machine multiple times.

Engineering design plays an important role in achieving fast transition from a product idea to commercialization or improving an existing product. Design process differs among companies where their accumulated expertise and exposure to modern software tools dictate their success.

Design optimization is a concept that more and more engineers start to realize its power. The increase of computing power and new advanced optimization algorithms make optimization practical and useful for iterative search of better designs, based on state-of-the-art simulation tools (FEA, CFD, multi-physics simulation, etc.).

Simulation is only used for design verification in most of the cases. Huge potential of incorporating simulation in looking for better design is lost in the conventional sequential process. Optimization-Driven Design is happening and is becoming the standard. The savings in time and costs, as well as the potential for product/process quality improvement, are substantial. With “one-click” optimization, non-experts can feel at ease with optimization; the latest AI-driven optimization algorithms offer much more than you know. Whoever rides the trend will be the leader of tomorrow.

Optimizing early and often has the following key benefits:

  • It creates a lasting ripple effect on the designs and on the financial impact throughout the design and product life cycle.
  • The number of designs explored increases drastically while empowering engineers to understand their models better to improve results.
  • The data gained from multiple iterations opens a whole new vista of insight.

In this article, we’ll guide you through these areas of opportunity while contrasting the traditional vs. modern optimization mindset.

In attendance at the recent CAASE (Conference on Advancing Analysis and Simulation, cohosted by NAFEMS and DE, June 5-7) was Dr. Gary Wang, CEO of Empower Operations. An ASME Fellow, Wang is also Professor of Mechatronic Systems Engineering at the Simon Fraser University, Vancouver, Canada.