A hybrid clonal selection algorithm for the location routing problem with stochastic demandsA hybrid clonal selection algorithm for the location routing problem with stochastic demands Peer-Reviewed Journal Publication Δημοσίευση σε Περιοδικό με Κριτές 2018-10-292016enIn this paper, a new formulation of the Location Routing Problem with Stochastic Demands is presented. The problem is treated as a two phase problem where in the first phase it is determined which depots will be opened and which customers will be assigned to them while in the second phase, for each of the open depots a Vehicle Routing Problem with Stochastic Demands is solved. For the solution of the problem a Hybrid Clonal Selection Algorithm is applied, where, in the two basic phases of the Clonal Selection Algorithm, a Variable Neighborhood Search algorithm and an Iterated Local Search algorithm respectively have been utilized. As there are no benchmark instances in the literature for this form of the problem, a number of new test instances have been created based on instances of the Capacitated Location Routing Problem. The algorithm is compared with both other variants of the Clonal Selection Algorithm and other evolutionary algorithms.http://creativecommons.org/licenses/by/4.0/Annals of Mathematics and Artificial Intelligence761-2121-142 Marinakis Ioannis Μαρινακης Ιωαννης Marinaki Magdalini Μαρινακη Μαγδαληνη Mygdalas Athanasios Μυγδαλας Αθανασιος Springer Netherlands Clonal selection algorithm Iterated local search Location routing problem with stochastic demands Variable neighborhood search