As remote robotic space satellite servicing technologies develop, each servicer satellite will need to account for a number of servicing scenarios and consider a variety of alternate design solutions to best meet the most servicing scenario requirements. This thesis presents a graph transformation method for systematically down-selecting the number of design options available, and highlighting trade-offs in sets of design solutions which best meet satellite servicing task requirements while also reducing total mass, maximum power needed and servicing time. The proposed method successfully identifies for further consideration several best design solutions from a set of approximately 10,000 potential solutions in the first test case examined, and from a set of approximately 2*1026 in the second test case examined.