CAJ | 학술논문

针对车辆在负载和空载状态下不同的成本核算模型，提出了两阶段算法求解最小配送成本：第一阶段用改进的扫描算法求得满足问题约束条件的若干组非同质化的初始解；第二阶段采用这些解作为多样化的初始种群，并用改进的单亲遗传算法进行全局、大范围搜索，最终求得满意解．实例计算表明，算法能在较短的时间内求得理想解，满足了物流配送企业深挖内部潜力、有效控制成本的现实需要．
침대차량재부재화공재상태하불동적성본핵산모형，제출료량계단산법구해최소배송성본：제일계단용개진적소묘산법구득만족문제약속조건적약간조비동질화적초시해；제이계단채용저사해작위다양화적초시충군，병용개진적단친유전산법진행전국、대범위수색，최종구득만의해．실례계산표명，산법능재교단적시간내구득이상해，만족료물류배송기업심알내부잠력、유효공제성본적현실수요．
Aiming at the different costing model for vehicle at load or no-load state, a two-stage algorithm for solving the minimum distribution costs was proposed. At the first stage, an enhanced sweeping algorithm was used to obtain various groups of non-homogeneity initial solution which met the constraints. At the sec- ond phase, these solutions were used as diverse initial population, an improved single parent genetic algo- rithm was used to perform overall and large range searching and finally the satisfactory result was obtained. Practical calculation shows that the algorithm can obtain a rnore satisfactory solution in a short period of time, which meets the actual needs for logistics companies to root out the internal potential to control the cost effectively.