I here want to point out briefly how Marx's critique of political economy in Capital is a work of science, albeit, not an exact science.
Firstly, it's a work of science in that it reveals the essence of phenomena (their causal foundations) that lies hidden behind their appearances (the immediately observable features of things). It thus satisfies the principal condition of a scientific realist conception of the central aim of science: to reveal the hidden causal structures of the phenomena of the world which in turn produce their observable features.
However, it's not an exact science in the way that Newton's theory of our solar system and universe is, with its mathematical laws of motion - which in turn can be used to make precise quantitative predictions about various celestial phenomena like the orbits of planets and comets.
This is because Marx's critique of political economy deals with the social phenomenon of a historically specific mode of production called capitalism and not a natural one like our solar system and universe. In short, it deals with a phenomenon comprised of socio-historical contingencies (the rate of unemployment goes up and down depending on certain factors being in place) rather than one which consists of real natural necessities (the elliptical orbits of planets around the Sun).
In consequence, it does not deal with a phenomenon which can be described in exact mathematically quantitative 'laws of nature' which in turn state that something will always happen in a precise way and in a precise time, etc., without any exceptions, like the elliptical orbit of any planet about the Sun. At best, it can only derive 'laws of motion' which basically capture and describe the developmental tendencies of the phenomenon of capitalism, like the tendency to economic crises.
We can thus conclude that Marx's critique of political economy in Capital is an inexact science, which is still far superior to the vulgar economics of our times - modern orthodox neoclassical economics - which, as Marx might say, fails to penetrate the appearances of things (and hence fails to be a science according to the central criterion of scientific realism).