Algoritmos genéticos

Solo disponible en BuenasTareas
  • Páginas : 29 (7226 palabras )
  • Descarga(s) : 0
  • Publicado : 15 de noviembre de 2010
Leer documento completo
Vista previa del texto

Journal of the Chinese Institute of Industrial Engineers, Vol. 20 No. 3, pp. 282-294 (2003)

Abhyuday A. Desai and Yung-Nien Yang* Department of Industrial Engineering Texas Tech University, Lubbock, TX 79409, USA Hamid R. Parsaei Department of Industrial Engineering University ofHouston, Houston

This paper investigates three scheduling algorithms and three inventory control policies to evaluate their effect on the system performance measures using computer simulation. The model chosen is a flow-shop type, computer assembly plant. Analysis is conducted on two scheduling measures (summation of process time and tardiness) and two inventory control measures (orderingcost and holding cost). The scheduling algorithms used are Shortest Processing Time, Genetic Algorithm and Simulated Annealing. The aim of the applied algorithms was to minimize the sum of flow times. Since it is known that SPT minimizes sum of flow times, it provides a benchmark for comparing the efficiency of genetic algorithm and simulated annealing search methods. EOQ, MRP and JIT are used asinventory control methods. An experimental design for the comparison is presented to evaluate which combination of control methods produces the best results under the designed scenario. Keywords: genetic algorithms, simulated annealing, economic ordering quantity, material requirement planning, just-in-time

At present, various inventory control algorithms are applied in industry.Choice of an inventory control method affects the system performance measures such as the ordering cost, holding cost and shortage cost. Manufacturers choose methods depending upon their manufacturing model. For example, make-by-order assembly plants choose the Just-in-Time technique for controlling plant inventory. Another aspect of manufacturing is the choice of the scheduling method.Scheduling problems are known to be NP-hard. As a result, the correct choice of the scheduling method is a major factor that decides the performance of the plant. Factors such as the utilization of the plant facility, make span and tardiness are depending on the scheduling method employed. Various approaches have been used to solve scheduling problems. Heuristics have been an important tool used since theproblem is NP-hard. These approaches aim to find the near-optimal solution in real time. Based upon local research, some modern heuristic techniques for

combinatorial problems have been successfully used for scheduling problems, in recent years, including Simulated Annealing and Genetic Algorithms. Inventory control and scheduling are the most important aspects in the current productionenvironment, especially in this networked era. Improvement of communication cannot prevent mistakes from inventory or scheduling but shorten the chance to make corrections. Therefore, the study of interaction between these two factors is a vital issue for traditional and virtual manufacturing environments. Also, the ready information in virtual manufacturing environments makes the integratedconsideration possible and reasonable. This paper studies this interaction between the inventory control method and the scheduling technique employed.

2.1 Inventory Control Methods
The traditional methods of inventory control use EOQ models. However, the basic EOQ model

Corresponding author:

Abhyuday A. Desai et al.: Effect of Activity Schedulingsand Inventory Control was based on the assumption that demand is constant, no shortage is considered and the lead-time is zero or constant. These assumptions are not faced in real life applications. The EOQ model does not take into consideration the demand pattern of the end product before determining the inventory levels of parts and materials. This is another major shortcoming of this model....
tracking img