Modifying colourings between time-steps to tackle changes in dynamic random graphs

Abstract

Many real world operational research problems can be formulated as graph colouring problems. Algorithms for this problem usually operate under the assumption that the size and constraints of a problem are fixed, allowing us to model the problem using a static graph. For many problems however, this is not the case and it would be more appropriate to model such problems using dynamic graphs. In this paper we will explore whether feasible colourings for one graph at time-step t can be modified into a colouring for a similar graph at time-step t+1 in some beneficial manner.