Many of the great scientists were also philosophers, and a number of great philosophers were scientists. I am of the opinion that it is where boundaries are blurred that the best scientific advancements are made, be it between sciences, or very different subjects. It is a philosophical approach that will allow for the inclusion of unexpected behaviours and observations, as this will encourage the consideration of a different approach to the problem.
Descartes is often quoted most significantly as a philosopher, yet many scientific interactions owe a debt to his system of co-ordinates, and much of mathematics and the physical sciences would not exist without that underpinning.
In 2006 there was discussion over the realisation that Van Gough (the artist) had depicted in his painting, Starry Night, something which mathematics is still struggling with, the concept of turbulence. Whilst Van Gough is not known as either a scientist of a philosopher like both of those occupations artists depict the world as they see it in a medium for others to interact with.
Both philosophers and scientists search for further meaning, and the abstraction that philosophy supplies allows for a different kind of understanding compared to that of a physical, measurable science.
That abstraction has been a foundation of science, as with cosmology. Kant, a philosopher, describes the balance between various forces which we now use as a basis of cosmology which Laplace, a mathematician, moved on.
In a similar fashion, observations that did not conform to the current model of the solar system lead to the hypothesis of extra planets. This was true in one case (Neptune and the effect it had on Uranus), but it was a different theory which explained the other (General Relativity rather than Vulcan existing). There was no way at the time of each hypothesis to be certain which of the two were correct, and it is that which I perceive to be the heart of philosophy, considering a thesis and examining it.
The Anthropic Principle is a philosophical concept that is used by many scientists, even if they don't know the term. The aim is to strike that balance between what is observed and what is understood. A good explanation will make something seem more usual, or more acceptable to the person being explained to. This principle can be applied not only to cosmology and the related fields, but to the other end of the size spectrum to subatomic interactions. The principle could also be applied to artists, as they can depict graphically what a scientist might depict mathematically.
In my experience we teach abstractions which are out dated as scientific fact, and then tell people that this theory has been superseded by this one, and here is the experimental proof for it. For example, an atom was once taken to be the smallest possible example of an element which was the simplest possible thing. This then developed into the existence of components for an atom, and their charges. Many places still teach atomic structure with pictures of electrons orbiting a nucleus as if they were planets within a solar system. Yet, the behaviour is more chaotic than that. The introduction of Leptons/Mesons is often left to a high level of learning, and yet there are consistent ways of using muons ('discovered' in 1936) to perform other experiments and learn more still about the behaviour of the universe we live in. Muons were discovered through experiments relating to cosmic rays, an observation was made as to the behaviour of a particle which did not conform to those that were known about. As such a new thesis was required, and this is the philosophical contribution. If Anderson and Neddermyer (who are attributed with finding the muon) had been dogmatic in their approach, and merely observed, and decided that their readings were incorrect, or their equipment faulty, then whole areas of research may have been impossible to understand. Instead, they considered options and considered whether or not there were other reasons for the observed behaviour, and applied a philosophical approach to the problem at hand.
In conclusion, the biggest contribution philosophy can give to the physical sciences in my opinion is that it challenges the narrow view of only the evidence, and encourages the theoretical view of possibilities that have yet to be proved.
