Memoirs from a Small-Scale Course on Industrial Math

Overview In this article I describe my experiences from a graduate course on industrial math that I taught at Duke in the spring semester 1995. When I volunteered to teach this course, several months before the beginning of the semester, it seemed like a very manageable task; but as the time to start teaching drew closer, I grew apprehensive that I would be unable to develop a successful course in this rather unfamiliar area. My worries were all the more severe because (i) I was operating solo, without any colleagues at Duke working on industrial problems; (ii) I hadn’t had significant contact with industrial problems for several years; and (iii) my skills on the computer are rather limited. In the final analysis, I believe the course was very successful, as I will explain below. In writing this article, I hope that others who may be considering such a course may profit from what I learned. Moreover, I want to add a personal note: I thoroughly enjoyed the course, and I plan to teach courses in this format regularly in the future. Because of my own background, I restricted the focus of the course to problems with a PDE/analysis slant. With some digging I found that industrial math has many challenging problems of this type which, because of both their intellectual content and their obvious relevance to a larger social context, provide an excellent vehicle for learning techniques of applied math. Moreover, understanding the current job market, students appreciated the broader training such a course offers, and they also liked the opportunity for active participation included in the structure of the course. For classes meeting twice per week (lasting 75 minutes), Duke’s semester includes 29 meetings. These were divided among the following four activities: