Presentation at the 2013 modeFRONTIER user meeting. Image courtesy of ESTECO.

modeFRONTIER’s biennial users’ meeting will be taking place later this week in Novi, Michigan, near Detroit. The event will draw more than 100 ESTECO customers looking to network, exchange information and learn computer-based optimization technologies. Companies attending will include AES, Bombardier, Ford Motor Company, MMI Engineering, Toyota and Whirlpool.

While at the user meeting, these attendees will:

See new features ESTECO added to modeFRONTIER and SOMO
See how modeFRONTIER can control CAE design of experiments (DOE) and optimizations
Discuss the modeFRONTIER and SOMO with ESTECO’s development team
Attend training
Learn about lifecycle management and multidisciplinary design optimization (MDO)

An interesting presentation at this year’s event will cover the development and application of enterprise MDO (EMDO) at Ford. As technical leader of business strategy and engineering optimization at Ford, Yan Fu’s work with modeFRONTIER and SOMO has done a lot to help her and her team rethink design strategies to anticipate manufacturing and market requirements in the early design cycle.

Fu’s work, outlined below, looks to find the trade-offs important to the automotive industry and has led to many innovations with respect to environmental initiatives, performance, safety and cost.

Ford Consolidates Their Lines, Allowing for Regional Tweaks

One model optimized for North American and European safety standards. Image courtesy of ESTECO.

Environmental initiatives, performance, safety and cost are typically competing objectives, which teams in the automotive industry need to balance with each design.

Though tools such as modeFRONTIER can help find the balance by performing optimizations and DOE, optimization itself still constitutes a significant amount of work.

To mitigate the work, Fu explains that many manufacturers have increased the number of cars that stem from a single engineering frame. Using this derivative approach, Ford has been able to reduce their product line of engineering frames. The company had 27 lines in 2007 and their target is nine by 2016.

“These ‘derivatives’ have numerous common product elements not visible to the consumer (e.g., common chassis, body structures, core components) in order to make differentiation of consumer-facing features valuable,” ESTECO recently wrote in a white paper.

“Developing an increasing number of derivatives per shared platform increases the complexity of managing the product development processes,” ESTECO continued, “given the explosion in the number of design models generated using different simulation tools and containing large numbers of design variables, responses, objectives and constraints.”

The complexity of simulation is growing and requires more people from various disciplines to have access. Image courtesy of ESTECO.

Though the development of this single engineering frame is made more complex, and the time to optimize this complex design increases, the strategy is that the overall workload decreases because other engineering frames are abandoned.

However, the more complex the design, the larger the team needs to be to get the job done, and this is where EMDO plays a vital role.

What’s the Framework of Ford’s Hub Design Strategy?

Multidisciplinary design optimization can become a multidisciplinary global project. PLM is key to ensuring the project is on track. Image courtesy of ESTECO.

At Ford, a vehicle line is assigned to one engineering center, which acts as an informational hub. For efficiency, this center makes sure that parts, designs, manufacturing information and suppliers are common for the line.

Regional teams, whose job is to tweak the product to the local market, support the hubs.

At R&D department level, this global collaboration was made possible with SOMO, modeFRONTIER, private cloud systems, shared model libraries, and PLM to allow teams to work on the products 24/7 from around the globe.

The global team looks into all the design objectives, from safety and durability to aerodynamics and noise, vibration and harshness (NVH), all at the same time. Due to multiple trade-offs and design space exploration software in various disciplines, modeFRONTIER becomes vital.

But modeFRONTIER needs to control a collection of software tools to assess these objectives. Many of these tools do simulation. 1D simulations typically come from model-based design software, such as Simulink, or computational software like MATLAB. More complex simulations of subcomponents, such as multiphysics simulations of parts, typically use CAE tools such as Nastran and LS-DYNA.

“Multidisciplinary design optimization (MDO) . . . has significantly streamlined the design cycle,” wrote ESTECO. “In fact, by enabling cross-attribute synchronization, MDO helps teams gain a better, earlier understanding of design limitations, signalling problems early on in the design process and allowing for anticipated and cheaper interventions. This means that the days of a simulation expert working alone on his/her own discipline are something that belongs to the past.”

For more than a decade, ESTECO and Ford have partnered to make process integration and design optimization support possible. This, in part, led to SOMO, which expands the desktop capabilities of modeFRONTIER to the web. SOMO distributes modeFRONTIER applications and standardized templates allowing for a more EMDO capabilities. This democratization of simulation and optimization allows for employees to centralize their efforts, regardless of their function and level of expertise. SOMO also includes a queue manager to control the computations on the high performance computer (HPC).

“The MDO expert doesn’t have to re-run the single-discipline workflows, nor understand all specific details related to them,” Fu explained in the report. “He/she combines the different domains using the ESTECO collaborative design optimization environment: without SOMO, he/she would have had to knock on everybody’s door, get their models and understand their requirements, review the whole workflow together with all attribute engineers . . . it would have taken a lot of time.”

Ford Puts EMDO into Action

Multidisciplinary workflows are combined into one assessment of the design space. Image courtesy of ESTECO.

Naturally, with a large projects, it is hard to tell where to start. The answer to that is to determine the target performance requirements. Common goals are to minimize costs and weight while maintaining or improving the durability, safety, drivability and NVH.

“For each attribute, engineers build the optimization workflow, create a specific DOE, train relevant response models and then publish the results to the central library, making them available to the MDO expert,” said Fu.

“He/she is now able to compose the comprehensive MDO workflow based both on his/her multidisciplinary expertise and the attribute experts’ knowledge made available in the central repository,” she continued.

The MDO expert then creates a DOE, or optimization algorithm, and modeFRONTIER uses simulation and response surface model (RSM) approximations to assess the design space.

Various simulations needed to assess a car’s performance and the computational cost. Image courtesy of ESTECO.

RSM and other approximations are used to reduce the computational demand of the overall study by limiting the time more rigorous studies, such as FEA and CFD, are required when searching for global maxima and minima.

These simplifications become more and more important the more complex the model becomes.

For instance, one of Ford’s case studies contained 49 variables and 12 constraints. Instead of running hundreds of CFD simulations, the RSM estimation used fewer models to make for a faster optimization. To ensure the results are accurate, the RSM are validated to ensure the results are similar to those that would be discovered by the finite element simulation.

For this Ford case study, RSM helped to assess about 20,000 variations within five minutes. An analytics tool then discovers the better performing design candidates. Over the next day or so, the best performing candidates are then validated using the original finite element simulations.

SOMO ensures that all the backup designs are versioned, archived and available to be reviewed if needed.

These charts, created by SOMO, help engineers to assess which design candidates are optimal. Image courtesy of ESTECO.

“When it’s time to present our results to top management, we not only have the comparison between baseline and optimized designs, but now we can also rely on an effective web-based tool that enables a quick decision-making process across attributes,” said Fu.

“SOMO’s insightful charts help highlight how much weight we are saving, what are the trade-offs between considered disciplines, how sensitive the designs are, and much more,” she continued.

“In the past, it would have taken us a month or more to collect all the models, formulate the MDO problem and build the workflow,” concluded Fu. “Now, with SOMO, tightly integrated with Ford private cloud computing and IT security systems, attribute engineers can do their own work and publish the latest models so the MDO team can build the top level MDO process right away. This is something new that has revolutionized our way of working.”

To learn more of Ford and Yan Fu’s work with SOMO, follow this link.

Other Interesting speakers at the User Meeting Include:

MARIO J. FELICE Manager of Global Powertrain NVH & Systems CAE Dept., Ford Motor Company
- KEYNOTE MULTI-PHYSICS SIMULATION AS APPLIED TO THE NVH REFINEMENT OF FORD MOTOR COMPANY’S POWERTRAINS
ROBERT LOTZ PhD, Sr Project Engineer - aerodynamics BorgWarner Turbo Systems
- SPECIAL TOPIC: DESIGN PROCESS TURBOCHARGER COMPRESSOR IMPELLERS USING MODEFRONTIER
ANDREA VACCA Associate Professor, Purdue University
- MODELING AND OPTIMIZATION OF POSITIVE DISPLACEMENT PUMPS FOR HIGH PRESSURE FLUID POWER APPLICATIONS
PINAR ACAR | Graduate Research Assistant, University of Michigan
- OPTIMIZATION OF MULTI-SCALE MICROSTRUCTURE DESIGN TO IMPROVE MACRO-LEVEL ENGINEERING PERFORMANCE IN AEROSPACE APPLICATIONS
JINGWEN HU University of Michigan, Associate Research Scientist
- RESTRAINT SYSTEM OPTIMIZATION FOR BELTED AND UNBELTED OCCUPANTS IN FRONTAL CRASHES

modeFRONTIER Events

North American modeFRONTIER Users’ Meeting, Novi, MI, Nov 4–5, 2015
International modeFRONTIER Users’ Meeting, Trieste, Italy, May 17–18, 2016

ESTECO has sponsored this post. They have no editorial input to this post - all opinions are mine. Shawn Wasserman