AC 2011-2547: AN EXPERT SYSTEM TO TEACH ENGINEERING STUDENTS SUTAINABILITY AND CREATIVITY DESIGN METHOD SELECTION

The objective of this paper is to present the development of an expert system guide test-bed for the selection of design methods to solve design problems. The authors observed that for senior design courses, instructors are faced with the dilemma of which methods to teach for a particular design problem. On one extreme an instructor can teach a plethora of methods with the urgency of transferring the largest amount of knowledge to the students. This may result counterproductive since the students will be overloaded with information. Another extreme will be to teach them only one or few methods; although this avoids an overload, the authors have observed that most instructors do this, teaching just a couple of preferred methods. The premise of this research work is that there is a lack of guidance when it comes to design method selection. This situation is also reflected in the professional world where it is well known that designers will stick to just a couple of methods in each category for the most of their professional life. This is understandable because (1) the designer is familiar with the method, and (2) there is not a clear guide to help decide which other methods to use. A test-bed for an expert system to guide in the selection of design methods has been developed at The University of Texas at El Paso (UTEP) from which specific guides are built upon. The ultimate goal is to help designers (students and professionals) understand what are their needs based on the design problem (initial sate, desired state, and other constraints) and provide them with method options that are appropriate while also explaining why these were suggested to help build their experience. The expert system will suggest the closest methods matching the criterion, and also recommend if the criterion needs to be relaxed or constrained to expand or reduce the number of suggested methods. The characterization of problems and methods are two much needed contributions of this research work as well as the mapping process which reflects the experience of skilled designers. This research work has two immediate benefits: (1) Design methods can be taught to students using problems that are appropriate to the particular method (using the wrong problem-method pair produces frustration) and (2) industry can be more efficient if they know what the best approach is to solve certain problems. An additional benefit of this research is the possibility of identifying areas in the problem space that are not currently addressed to existing ideation methods, this disconnection represents an area of opportunity to generate targeted design methods.