Simulation: A Revolution in Design and Manufacturing: Part 1

The abstract concept of simulation has a long history. Artisans use models of their work in order to share a vision or explore new ideas, architects construct intricate scale models of buildings to better conceive the way light and space might work together. In engineering, simulation has assumed new importance, as it provides a stepping-stone between two-dimensional drawings and a three-dimensional product. More importantly, it allows engineering teams to develop multiple scenarios in a virtual world where the cost is much lower.

Digital transformation has helped define simulation over its lifetime. The advent of the computer meant that modeling could move into a virtual domain. This domain could support more functionality than a real-world model, allowing it to be manipulated, changed, and improved with ease. More significantly, it allowed other factors to be introduced into the virtual domain, enabling insights into the model’s reaction to additional variables. This is when the science of simulation took a leap forward in becoming a truly enabling technology.

Today, simulation software is used in the most demanding environments. For example, NASA has its own Mission Simulation Facility (MSF), a framework that supports the development of autonomous technology for planetary exploration and includes virtual robots. The aerospace industry uses virtual reality to simulate human-machine interfaces to improve optimization, they also apply simulation software when using additive manufacturing to achieve ‘print right, first time’ success.

As these applications develop, the need for better simulation solutions also increases. This means better support for more detail, faster computation, and real-time responses. Simulation is no longer just used to provide an estimation of what a final product may look like, it is now used as a fundamental part of the design and manufacturing process.

In the first of two blogs on this topic, we will look at the importance of simulation software and how it is being used at every level, from improving the way we design components to the way they are manufactured in an automated production environment and even how they are used in the real world.

Simulation at this level relies heavily on mathematics computation, predominantly using techniques like finite element analysis (FEA), the finite volume method (FVM), and the finite-difference time-domain (FDTD) method. All of these have the word ‘finite’ in common, which essentially means breaking large problems down into smaller and more manageable tasks without losing information in the process. “Finite” refers to the level of reduction that can be achieved in the simulation process. As computational resources improve, this limit is being continuously extended.

Despite the continued increase in available resources, enabled by high-performance computing (HPC) and hyperscale architectures, there will always be a compromise between the accuracy that can be achieved with simulation and the time it takes to arrive at those results. New technologies, such as Artificial Intelligence (AI) and Machine Learning (ML), are being employed to redefine this balance. AI in simulation introduces the concept of using expert systems to accelerate the decision-making and analysis part of a simulation, but it will also be used to explore alternative scenarios when the simulation flags an issue. The speed and accuracy with which AI-accelerated simulation operates are also expected to improve over time, as the system learns to do things better.

The tools now used in simulation, based on the underlying methods and technologies, can address a nearly unlimited variety of functions. At a very basic level, fundamental forces such as electromagnetism, influence the way properties from the very small (electrons) to the relatively big (objects) behave when exposed to other forces, such as energy in the form of heat or vibration. Because these elements can be modeled, simulation is the ideal environment in which to run these scenarios.

For a design and manufacturing company like Molex, where every productivity gain is important to the overall business, the use of simulation has become an imperative. The technology is used in the design of the product, it informs how the product will be used in an automated process, and it also simulates how the final product will be manufactured on a production line. At each stage, simulation addresses different but crucial elements, some of which may only become apparent during the simulation of the lifecycle.  

For example, the molding of a plastic component can be simulated and the output analyzed from a mechanical point of view. This would typically involve analyzing the potential for defects and the impact it would have on overall yield. Modifying the simulated component design allows for it to be optimized long before it is released for tooling. Further, modeling how the plastic component is then used in the next production stage also spotlights incremental opportunities to improve efficiency.

Molex uses flow simulation in the tooling design stage to identify issues such as poor fill patterns, shrinkage, warpage, and the thickness of walls. This simulation helps identify the presence of an issue as well as predict the actual location of the fault, allowing the mold and the process parameters to be optimized at the simulation stage.

In practice, the shape of the mold could have a substandard fill pattern, resulting in too many parts coming off the production line with missing features, or high levels of warpage. Addressing these potential deficiencies at the simulation stage may require the mold design to be changed by adding, removing, or redistributing some of the mold material. Simulation at each phase enables the creation of a more predictive and comprehensive picture, and that diminishes both time, waste, and cost due to rework.

Mold steel failure prediction is critical in molding optimization and can result in higher yield and subsequently lower unit price. It also notably influences the lifetime of the tool. If an unfavorable condition is present, it will put stress on the tool in unexpected ways. This could drastically reduce the tool’s lifetime.

Another key element where simulation can deliver positive results is in the overall cycle time of the molding process. This is the accumulated time it takes to produce a unit from the mold, including fill, pack, cooling, and mold open time. Simulation allows each of these parameters to be adjusted and the effects of those changes to be measured, with predictions on how a shorter fill time or shorter cooling time will affect the final product. By carrying out many thousands of simulated cycles the result can be significant, savings of several seconds in the overall cycle time can result in many more units per hour.

Manufacturing end products requires components made from multiple materials. Metal stamping is an important element of product design. Stamping involves forming a piece of metal into a predefined shape through force applied evenly across the surface. Stamping is another area where simulation is having a positive impact. Stamping simulation software predicts how the stamped product is formed.

Molex is using these tools to improve the way it designs stamped products, by first simulating the product to assess its design feasibility, and then through incremental simulation steps to design the die and/or punch and the size or shape of the blank. This work has proven useful in diminishing iterations required during design.

Incremental simulation allows tiny changes to the tool to be assessed virtually, without the time and expense of creating a new tool. Because all of the data is known, the effect of any small change can be analyzed, this would be much more difficult in the ‘real world’, where the tool would first need to be manufactured and then the resulting product measured carefully before any conclusions could be drawn from the design changes. Of course, simulation also supports easy roll-back to previous designs if the effects of those small changes prove detrimental.

Simulation software is continuously improving and new technologies like AI and ML will only accelerate this to ensure that modelling and simulation tools can provide a reliable and cost-effective solution to optimize design and manufacturing.  Molex is utilizing a variety of simulation tools in every stage of the design cycle, from flow simulation to production line time frames, all of which enable us to ensure maximum productivity and reduce time, wastage and cost – all of which ultimately benefit our partners.

In part two, we will be looking in more detail at how simulation can optimize the assembly process and our vision of the future of simulation, including the use of digital twins and AI.