Showing 2 results for Genetic Algorithm.
R. Eskrootchi, M. H. Sebt, F. Jazebi,
Volume 12, Issue 3 (9-2014)
Abstract
In different projects the speed of different machinery can be estimated using manufacturer's handbooks and a number of modification factors to consider the environmental effects, type of the project and status of site management. Since the statuses of different factors of the domestic projects are totally different from those of the international projects, there is a wide discrepancy between the determined speed by handbooks and the actual values in the domestic projects. This paper is aimed to develop a fuzzy system to estimate soil excavation rates at earthmoving jobsites. The proposed fuzzy system is based on IF-THEN rules a genetic algorithm improves the overall accuracy. The obtained results clearly revealed the capability and applicability of the proposed system to properly estimate soil excavation speed. The average error of fuzzy system, handbook method and nearest neighbor interpolation are 10 , 92 and 32 percent, respectively.
A. Sheikholeslami, Gh. Ilati, M. Kobari,
Volume 12, Issue 3 (9-2014)
Abstract
We consider the problem of continuous dynamic berth allocation to containerships in a tidal seaport. In some container ports, low water depth in coastal area causes many restrictions on providing vessel's services. Therefore, berth allocation planning for relatively large vessels with high draft is subject to tidal conditions when the vessels are in the access channel as from anchorage area to the quay. Tidal conditions sometimes have a significant effect on possibility of entrance and departure of these ships to or from ports. Shahid Rajaee Port Complex, Iran's largest container seaport and the case study of this research, located at northern coast of Persian Gulf and has low water depth in its area. Historical data of seaside operations in this port is applied to the proposed model. This model also takes into account the variations of water depth in different berths. Simultaneous programming for two or more container terminals and exertion of priority and precedency coefficients based on vessel size and voyage type altogether are other attributes of this model. Here, genetic algorithm in combination with pattern search algorithm was used for solving the problem. Computational experiments have indicated that the proposed heuristic is relatively effective just for small size instances.
