Some optimization problems are too complex to be solved exactly, using specific software tools. For this reason, in many cases it is more convenient to define and design a heuristic procedure even though the final solution is generally sub-optimal. One of the most promising approaches in the traditional computing environment is the Iterated Local Search method. It is based on an exploration of the neighbor of the current solution and its performance is estimated to be very high for a large number of problems. The main drawback of the approach could be the required computational time, in particular when the neighbor to be explored becomes too large. We propose a general distributed framework, based on Iterated Local Search, and we show a concrete application in logistics, related to the optimal assignment of products to storage locations in a warehouse.
A general distributed framework based on iterated local search
PISACANE, ORNELLA;
2009-01-01
Abstract
Some optimization problems are too complex to be solved exactly, using specific software tools. For this reason, in many cases it is more convenient to define and design a heuristic procedure even though the final solution is generally sub-optimal. One of the most promising approaches in the traditional computing environment is the Iterated Local Search method. It is based on an exploration of the neighbor of the current solution and its performance is estimated to be very high for a large number of problems. The main drawback of the approach could be the required computational time, in particular when the neighbor to be explored becomes too large. We propose a general distributed framework, based on Iterated Local Search, and we show a concrete application in logistics, related to the optimal assignment of products to storage locations in a warehouse.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
