A multi-stage stochastic integer programming approach for a multi-echelon lot-sizing problem with returns and lost sales

Franco Quezada, Céline Gicquel, Safia Kedad-Sidhoum, Dong Quan Vu

Abstract: Highlights•We study a multi-echelon stochastic lot-sizing problem within remanufacturing environment.•It is modelled as a multi-stage stochastic integer program and solved by a Branch & Cut algorithm.•We propose a new family of tree valid inequalities generated by a mixing procedure.•A heuristic algorithm is studied to solve the separation problem.•The time needed to obtain guaranteed optimal solutions is significantly reduced and the value of the stochastic solution is assessed providing significantly improvement.