需求预测综合模型及其与库存决策的集成研究

准确的预测有助于企业做出有效的决策,包括生产计划、定价和促销决策等,以减少库存、提高客户满意度和企业竞争力． 以快速消费品为研究对象,分析其需求影响因素,建立了时间序列分析与多元回归整合的需求预测综合模型; 将此预测模型引入到库存决策中,构建了基于库存成本最小的需求预测与库存决策集成模型,并借助变邻域搜索算法获得模型的参数值; 最后,选用实际数据,验证了所构建的需求预测综合模型、物流需求预测与库存决策集成模型及其求解方法的有效性．     

Modelling the tradeoff between postponement capacity and forecast accuracy

The ability to offer rapid delivery of a wide variety of customised products requires companies to maintain high levels of product inventories to quickly respond to customer demands. One alternative for reducing final product inventories while providing the required customer service level is delayed product differentiation, known as postponement. This strategy, however, can result in significant costs of increasing capacity at the postponement stage. Another alternative is to improve forecast accuracy, resulting in costs associated with more sophisticated forecasting methodologies. In this study we model the costs associated with each alternative and the resulting reductions in inventory levels, while maintaining a constant service level. We illustrate the interaction between these variables using a numerical example motivated by our work with a local manufacturer of non-durable household goods. Our findings show that large cost differences can exist between the two strategies, and that these costs play a significant role in determining the best strategy. Also, the value of the product (through holding cost) sets a limit on the benefit that can be realised by either strategy. These results have important managerial implications that should be considered when making the postponement decision.     

基于需求预测的两级动态配送路径优化研究

针对传统响应式配送难以应对海量动态客户需求的困境,在此提出基于需求预测的两级动态配送路径优化方法。利用历史数据从需求预测、需求聚类和需求配额三个维度,处理电商物流配送过程中的动态需求;建立基于需求预测的单阶段和多阶段两级车辆配送路径优化模型,并针对问题特性利用分支定界算法与CW-禁忌搜索算法生成两个阶段的最优配送路径。最后结合实验案例对模型与算法的有效性进行验证,试验结果表明本文设计方法具有较好的动态场景适应度和客户响应能力。     

基于GARCH M模型的人民币汇率预测

在经济全球化的形势下,人民币的走势是关系到中国外贸企业生存、国际地位以及国家金融环境的重要因素,因此对人民币/美元汇率进行预测是十分有必要的。通过对GARCH-M模型在预测人民币美元汇率的可行性,时间序列存在异方差性和自相关性的论证,建立相应的GARCH(1,1)-M模型,并运用模型对美元/人民币汇率进行预测。表明在现实中可以运用GARCH-M模型进行汇率趋势预测,但是由于检验的数据较少,所以不能达到精确的预期目的。     

罪刑法定原则视域下兜底性条款的限界设定——以非法经营罪为视角

罪刑法定原则要求法律明确性和禁止溯及既往,以此满足国民预测可能性要求,从而为人们行动自由提供前提性条件。法律的稳定性使得刑法一时难以应对社会情势的变迁,为了严密刑事法网,于是刑法条文中出现了兜底性条款的规定。为了满足司法明确性的要求,司法机关对兜底性条款作出补充解释,使得兜底性条款外延不断扩张．如司法机关对非法经营罪的解释是一明显例证。由于司法解释具有溯及既往的效力,使得国民难以预测自己行为的合法性,从而限制人们行动的自由。为此,应秉持罪刑法定原则,对刑法中兜底性条款作出合理的限定。     

对辽宁省高等教育发展规模的分析与预测

以经济发展水平与城市化作为两个控制变量,利用BP人工神经网络对辽宁省未来高等教育规模发展进行了分析与预测。     

基于多种模型组合的我国2015年人口总数预测

首先建立了线性回归模型、曲线拟合模型、指数函数模型、时间序列模型、灰色系统模型5个单一模型,就2001-2004年的人口数分别对这5个模型的预测结果进行了比较;然后考虑将这5种模型用5种组合方法组成新的组合模型来预测人口总数;最后对我国2005-2015年的人口总数进行了预测,预测结果为我国2015年人口总数为14.30亿。     

镇江市解放南路与南门天桥连接方案研究

在对镇江市南门天桥周边道路及交叉口交通量的调查分析及预测的基础上 ,对将解放南路直接连接上天桥路的可行性进行了分析。提出了首先应对该地区路网进行完善 ,在结合路网完善的基础上 ,提出了解放南路与天桥路的连接方案     

水电站FcDSS中洪水过程预测模型研究

本文深入研究了水电站FcDSS中入库洪水过程预测模型、区间洪水过程预测模型的理论和方法 .研究成果已在水电站FcDSS中应用 ,取得满意效果 .     

The predictive distributions of thinning‐based count processes

This paper shows that the term structure of conditional (i.e. predictive) distributions allows for closed form expression in a large family of (possibly higher order or infinite order) thinning‐based count processes such as INAR(p), INARCH(p), NBAR(p), and INGARCH(1,1). Such predictive distributions are currently often deemed intractable by the literature and existing approximation methods are usually time consuming and induce approximation errors. In this paper, we propose a Taylor's expansion algorithm for these predictive distributions, which is both exact and fast. Through extensive simulation exercises, we demonstrate its advantages with respect to existing methods in terms of the computational gain and/or precision.