Understanding how tires age is one goal of a research partnership between Sandia and Goodyear Tire & Rubber Co. Computational tools reduce the need to build and test prototypes -- and get new products to market sooner.

Sometimes duplicating a manufacturing process can be like trying to follow a friend's chocolate chip cookie recipe: It just never seems to turn out exactly the same. Oddly, some manufacturing processes can only be performed successfully by certain people. Even when procedures are well documented, the results cannot be duplicated exactly. While a disappointing batch of cookies is probably of no great economic consequence, the inability to duplicate a manufacturing process can have major financial implications.

Another puzzling manufacturing issue is processes that are based on a substantial amount of "lore" -- methods that work for unknown reasons, or processes that must be constantly "tweaked" in different ways. Both of these situations result in manufacturing processes that are delicately balanced between success and failure and allow little or no latitude for modification.

Manufacturing Processes That Work...Every Time
Sandia National Laboratories is addressing these and other serious manufacturing problems with computational simulation of manufacturing processes. Until the development of high-performance computing, it was not practical to perform the huge number of calculations necessary to analyze a process or replicate the many tiny details. Now, high-performance computing makes it possible to more completely define a manufacturing process. This means the process can be duplicated and will work more predictably, and that success depends less on human factors. In addition, it is possible to predict how modifications will ripple through the entire process before all the equipment must be purchased and installed. The ability to quickly simulate an entire process makes it possible to eliminate many of the initial failures that occur when a new manufacturing process is brought on-line.

Sandia, in partnership with a variety of firms in the microelectronics industry, including Intel and Motorola, has simulated the performance of microelectronics processing equipment to develop a fundamental understanding of what controls performance and, based on the information gleaned, has modified the equipment to improve results.

Sandia's modeling and simulation capabilities are being applied to improving microelectronics processing equipment.

In Less Time
Sometimes, just figuring out the most efficient and effective way to complete a manufacturing process is the most time-consuming task. For instance, in brazing (joining two pieces of metal by melting a piece of metal between them), one of the most time-consuming aspects is simply trying to figure out which furnace settings consistently produce the best bonds. Fast computer workstations can perform a simulation, but it takes more than 100 hours to perform the necessary calculations. With high-performance computing, the same calculations take less than an hour.

New Products Without Prototypes
High-performance computing also offers the advantage of creating new products without the expensive, time-consuming physical prototype process. Born from Sandia's requirement to create small numbers of highly specialized weapon components, this ability can be particularly valuable to small-lot manufacturers.

To be truly useful, simulation models must be accurate. Simulation of small details is often beyond the expertise of manufacturing designers and is very time-consuming. Sandia has worked with various partners, including the Goodyear Tire and Rubber Co., to develop simulations that are easy to create and use effectively.