Integration of Genetic Algorithms and Simulation for Stockyard Layout Planning

This paper describes the development of a process simulation model and integration of the Genetic Algorithms (GA) with the model as optimisation techniques using a case study of stockyard layout planning in precast concrete products industry. Genetic algorithms have been used for two purposes: firstly, to identify clusters of concrete products by analysing sales historical data such that frequently ordered products are grouped together; and secondly, to identify the allocation of the identified clusters to the storage locations. The simulation model was developed in ARENA (simulation software) which evaluates stockyard layout scenarios with different spatial layouts and different allocation of products to storage locations. The evaluation parameters include throughput time for loading and dispatch of the products to service customer orders, queuing and waiting times of lorries and cost of loading and dispatch of the products in the stockyard. The genetic algorithms are used to identify the best allocation of products to storage locations by evaluating the objective function values such as throughput time. The finding from the experiments conducted using a case study are presented and discussed.