In my experience, having done primarily design work, but also some behavioral science research, g/ood/ design is just as analytic, empirical, and rigorous as "scientific" methods of problem-solving.
Just to muddy the waters, there's also design based research (coming out of the learning sciences) which is relatively new but is formed specifically for scientific inquiry. The key distinction between design based research and other forms of R&D lies in its concern with theory.
There's actually a bit of an issue with framing traditional problem solving as a "S"cientific activity. When we talk about the sciences (hard or soft) we're usually talking about developing and testing theories (which is of course a design activity in and of itself although most scientists don't talk about it in those terms). Meanwhile, most people engaged in pragmatic forms of problem solving have very little interest in theory (except perhaps their own espoused theories as discussed by Argyris and Schon) in the sense that science is usually concerned with it. The result is of course gaps between theory and practice which folks have sought to bridge through a variety of methods, sometimes with greater success than others.