A new approach for scheduling independent tasks with multiple modes

Heuristic algorithms for scheduling tasks with multiple modes and minimizing the schedule length involve in general two distinct phases, task mode assignment and then task scheduling. We propose a novel approach where these two features are managed in an integrated mechanism with mode assignment embedded in scheduling. The problem is first reformulated as a special single-mode task scheduling problem, and then is modeled as a graph interval T-coloring. Finally, a tabu-like metaheuristic is proposed for this latter graph coloring problem, and the performance of our approach is compared to known multi-mode scheduling heuristics.