# Experimental Design Space Explorer

Design space explorers allow designers to rapidly explore the vast combinatoric spaces that represent valid solutions to a design problem. This allows investigation of a wider variety of concepts before settling on a final design choice; moreover, it encourages more daring design directions by removing the risk of developing an invalid design. We are constructing such an explorer to aid in the design of experiments.

Although essentially an optimization problem that finds a balance between the cost of the experiment and the questions it can be used to answer, the design process cannot be automated in general. The design decisions require specialized knowledge about the problem space, while the software only represents the experimental design space. This makes experimental design an ideal area for cooperative problem solving where human design is facilitated - not replaced - by computational power. The usual design process for an experiment involves leafing through a book on experimental design and picking the exemplar that best matches the problem at hand. This is problematic for two reasons: First, choosing an appropriate model is difficult because one must find a mapping between examples provided by the book and the current problem. Second, as the designs provided are often not an exact fit, experimenters can be led to reach invalid conclusions by misapplying a design. Instead of presenting a set of boilerplate examples, this project allows exploration of the space of all possible experimental designs. This allows the designer to rapidly converge on an ideally tailored design balancing cost and power.