Improving Particle Swarm Optimization to Solve the Cutting Steel Bars Problem

• Main Authors: Shu-min Ou, 歐淑民
• Other Authors: Hsien-tang Wu
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/33393006291111071009

碩士 === 朝陽科技大學 === 營建工程系碩士班 === 99 === The steel bars cutting plan is a combinational optimization problem, which discuss how to cut the specific quantities and length of steel bars and to make the rest of the raw steel bars at least. In recent years, the price of the raw material fluctuated very much. If we can reduce the quantities of the rest of the raw steel bars, it is more helpful to control the cost in construction. The objective of the study is to solve steel bars cutting optimization with improving particle swarm algorithm. The design of the particle coding is used to determine the steel bars cutting order by the real number vector of the 2 times dimension of requiring steel bars quantities. The next fit packing method is used to carry out the packing operation, calculate objective function value, determine the steel bars cutting plan, and memorize excellent order of the cutting steel bars. The excellent cutting order pertubation mechanism is to avoid particles falling into the local optimal solution and to increase the opportunity of the particles in search of better solution. The results show that the study in cutting steel bars mode can get most suitable solution more quickly and better.