The Design of Experiments (DOE) Method And How It's Used in Optimus Software

The design of experiments (DOE) method is a statistical methodology that projects the best possible set of experiments needed in a design space to derive the most relevant and accurate information or outcome for a particular project. The efficiency that DOE brings to a project or goal therefore is second to none, saving organizations valuable time and resources.

It was first developed in the 1920s by a British statistician named Ronald Fisher. Some 70 years later, Fisher's method has become a powerful software tool for engineers and researchers. DOE can assist in the design of any information-gathering exercises where potential variation is present and needs to be well understood.

In design of experiments, the experimenter is often interested in the effect of a process or system (which make up the ‘treatment’) on an object or group of objects (otherwise know as the ‘experimental units’). Both the treatments and the experimental units can represent a wide variety of people, live forms, products or things. Design of experiments is thus a discipline that has very broad application across all the natural and social sciences and engineering.

Today’s competitive design environment mandates that engineers and design teams consistently search for ways to save time, money and valuable resources. One of the current best methods to do so is the utilization of design of experiments (DOE) tools. Specifically design of experiment software. Optimus software is one well respected DOE software that uses DOE to capture the design space for each individual project.

Would leave out this entire paragraph – it is more related to RSM than DOE

Well chosen experimental designs maximize the amount of valuable information that can be obtained for a given amount of effort. By using DOE software, engineers, scientists and researchers in virtually any industry can optimize responses and discover winning combinations. This enables them to cut significant time off of production or development schedules and a substantial annual savings (often in six figures or more).

Design of Experiments can be combined with Response Surface Modeling (RSM) to predict responses for designs that were not explicitly part of the initial DOE – and this can be done with limited computational effort. RSM thus allows further post-processing of DOE results. Optimus software can also generate the best RSM automatically from design of experiments, drawing from a large set of RSM algorithms and optimizing the RSM using a cross-validation approach.

To maximize the power of a DOE tool it should have an easy and intuitive interface and a broad selection of designs for optimizing and screening processes or product formulations. It is also important that it randomizes the order of experiments run since this will ensure that interference factors are spread randomly across all control factors.

The use of design of experiments methodology is no longer an option in today’s engineering landscape. Therefore tools like Optimus software that use this methodology to help teams develop the best product are even more valuable and vital.