Statlab - a virtual environment for design of experiments
K. Rijpkema (Eindhoven University of Technology), M. Boon and A. Di Bucchianico (Eindhoven University of Technology)
A good working knowledge of DOE (Design of Experiments) is essential for both industrial statisticians and engineers. It is therefore essential that courses in statistics pay sufficient attention to this topic. However, a distinctive feature of DOE is that it is pro-active, unlike many other statistical techniques. Hence, this requires a teaching approach that forces students to actively think about several aspects of setting up an experimental design, without steering the student too much. In order to create such a teaching environment to be used in statistics courses at various departments of Eindhoven University of Technology, a web based tool called Statlab has been developed. This tool adapts itself to the student, who is being led through one of several possible case studies. We describe the teaching philosophy behind this tool, as well as its technical implementation. Students generally consider using Statlab as a stimulating teaching environment. We consider our tool to be a useful addition to existing teaching tools like the well-known helicopter experiment or the experiments described in e.g. [1] and [2].
We are unaware of teaching software comparable to our software, although we do know about several small Web applets that perform subtasks. Statlab is written in Java and it can be freely used through the web site http://www.win.tue.nl/statlab/.
The current version of Statlab consists of two parts. During the first part, the assignment is to set up a screening experiment. During this screening experiment the student needs to create a two-level factorial design in order to determine the significant factors in a (generally chemical) process. The student has to take decisions concerning design size, blocks, high and low factor values, aliasing structure, centre points, replication and randomisation. When the student has found the significant effects, the second part of Statlab starts: the optimisation phase using response surface methods (RSM). In this part the student tries to find the optimal values of the significant effects using the method of steepest ascent.
Some of Statlab's features are: it is multilingual, the program flow is determined by the choices of the student, automatic grading system by sending a commented email of the student's actions to the teacher, automatic generation of available designs using an implementation of Franklin and Bailey's algorithm [3], graphical visualisation of fitted response surfaces.
References
[1] Anderson-Cook C.M. (1998). Designing a First Experiment: A Project for Design of Experiment
Courses, American Statistician 52, 4, 333-342.
[2] Kopas D.A. and McAllister P.R. (1992). Process Improvement Exercises for the Chemical Industry. American Statistician 46, 1, 34-41.
[3] Franklin M.F. and Bailey R.A. (1977). Selection of Defining Contrasts and Confounded Effects in Two-level Experiments. Applied Statistics 26, 1, 321-326.