模糊时间约束下多周期混合运输配送网络优化

发布时间：2018-12-17 07:15

【摘要】：随着市场竞争的加剧,供应链处于一个复杂多变的动态环境中,对多周期动态下的供应链配送系统的优化,成为供应链配送网络优化的发展趋势。混合运输(供应商与客户的匹配运输与非匹配运输相结合)作为一种全新的配送模式,有效地解决了供应链配送中的空车运输问题,降低了供应链配送的成本。但在已有的研究中,都是将混合运输模式应用在供应链单周期运营的情况下,忽略了在不同时间里系统的动态连续性,这可能会导致最优结果在实际应用中出现偏差。因此,研究动态多周期环境下的混合运输配送网络问题更加具有现实意义。另外,在现实的配送环境中,存在着大量不可预知的因素,这些不确定因素对配送车辆的到达时间造成干扰和影响,导致配送网络中存在模糊的客户预约时间和模糊的车辆配送时间。但现有研究中,多数只考虑了客户的模糊时间窗,极少考虑车辆配送时间的模糊性。因此,将模糊时间理论与配送网络优化问题相结合,更符合现实,具有一定的研究价值。本文对多周期下的混合运输配送网络问题进行研究,并将该问题与模糊时间理论相结合,从多个角度出发,综合考虑了供应链配送系统的动态性、配送时间和客户时间窗的模糊性、配送车型的多样性,提出了带有模糊时间约束的多周期多车型混合运输配送网络优化问题。同时,将这一问题应用到多工厂、多客户点、多供应商的供应链三级配送网络中,建立了两个0-1整数优化对比模型：一是带有模糊时间约束的单周期多车型混合运输配送网络优化模型,二是带有模糊时间约束的多周期多车型配送网络优化模型。基于以上思路建立的模型,更加符合物流配送活动的实际情况。对于模型的求解,本文采用效率和准确性都较高的微粒群智能优化算法进行求解。通过算例验证了微粒群算法的有效性和两个模型的实用性。同时,在对比实验中,分别对算法的重要参数和车辆单位运输费用进行设置,验证了参数对算法性能的影响和模型的稳定性,稳定的模型对企业做出长期决策具有良好的指导作用。算例的结果表明多周期下的供应链配送模型要比单周期下的供应链配送模型更加优越,可以有效的对企业资源进行合理配置和规划,使得到的优化成本更低。
[Abstract]:With the aggravation of the market competition, the supply chain is in a complex and changeable dynamic environment. The optimization of the supply chain distribution system under the multi-cycle dynamic condition has become the development trend of the supply chain distribution network optimization. As a new distribution mode, hybrid transportation (the combination of supplier and customer matching transport and mismatched transport) effectively solves the problem of empty car transportation in supply chain distribution and reduces the cost of supply chain distribution. However, in the existing research, the hybrid transport mode is applied in the case of supply chain single cycle operation, and the dynamic continuity of the system is ignored in different time, which may lead to the deviation of the optimal result in the practical application. Therefore, it is of practical significance to study the mixed transportation distribution network in dynamic multi-cycle environment. In addition, in the actual distribution environment, there are a large number of unpredictable factors, these uncertain factors on the delivery vehicle arrival time caused interference and impact, There are fuzzy customer reservation time and fuzzy vehicle delivery time in distribution network. However, most of the existing studies only consider the fuzzy time window of the customer, and rarely consider the fuzziness of the vehicle delivery time. Therefore, the combination of fuzzy time theory and distribution network optimization is more in line with the reality and has certain research value. In this paper, the problem of mixed transportation and distribution network under multi-period is studied, and combining this problem with fuzzy time theory, the dynamic of supply chain distribution system is considered synthetically from many angles. Due to the fuzziness of distribution time and customer time window and the diversity of distribution models, the optimization problem of multi-cycle and multi-vehicle hybrid transportation distribution network with fuzzy time constraints is proposed. At the same time, this problem is applied to the three-level distribution network of multi-factory, multi-customer and multi-supplier supply chain. Two 0-1 integer optimization models are established: one is the single cycle multi-vehicle hybrid transportation network optimization model with fuzzy time constraints, the other is the multi-cycle multi-vehicle distribution network optimization model with fuzzy time constraints. Based on the above ideas, the model is more in line with the actual situation of logistics distribution activities. For the solution of the model, the particle swarm optimization algorithm with high efficiency and accuracy is used to solve the problem. The effectiveness of PSO and the practicability of the two models are verified by an example. At the same time, in the contrast experiment, the important parameters of the algorithm and the unit transportation cost of the vehicle are set respectively, which verifies the influence of the parameters on the performance of the algorithm and the stability of the model. The stable model has a good guiding effect on long-term decision-making. The results show that the multi-cycle supply chain distribution model is more superior than the single-cycle supply chain distribution model, which can effectively allocate and plan the enterprise resources, and make the optimization cost lower.
【学位授予单位】：福州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F274;TP18

【相似文献】