Cooperative project scheduling with controllable processing times: a game theory framework

Cyril Briand (1, 2) , Jean-Charles Billaut (3, 1)

1 LAAS-ROC - Équipe Recherche Opérationnelle, Optimisation Combinatoire et Contraintes
2 UT3 - Université Toulouse III - Paul Sabatier
3 LIFAT - Laboratoire d'Informatique Fondamentale et Appliquée de Tours

This paper considers a project-scheduling environment assuming that the activities of the project network are distributed among a set of actors (or agents). Activity durations are modeled as time intervals and are assumed con-trollable, meaning that every actor is allowed to shorten the duration of some activities by adding extra-money. For performing the project, actors have to collaborate with each other intending to satisfy the desired project duration , defined by the project customer as a time interval. In this work, every actor's payoff corresponds to a fixed percentage of the total customer's payment, which itself depends on the ability of the actors to achieve the project in time, provided daily penalty costs are applied in case of tardiness. This problem can be modeled as a game, where players (actors) have to select a strategy (a duration vector for their activities) intending to maximize their profit. In this paper, the focus is put on the modeling of this project game, and on the connections between various decision problems, arising either in decision or game theory. We also study the particular case where each activity is assigned to one specific agent.

Domaines: Recherche opérationnelle [math.OC]

Cyril Briand, Jean-Charles Billaut. Cooperative project scheduling with controllable processing times: a game theory framework. 16th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), Sep 2011, Toulouse, France. 7p., ⟨10.1109/ETFA.2011.6059081⟩. ⟨hal-01233563⟩