基于深度学习LSTM对交通流状态的预测

当今道路交通状态对城市管理和人们出行愈加重要,影响着人类生活的方方面面.以深圳交通为研究对象,由基础车辆数据和道路坐标构建了路网系统,从车辆速度和密度两个方面导出了交通流状态评价指数TSI.利用深度学习长短期记忆神经网络(LSTM)对车辆速度和密度两个指标进行预测,并通过对比极限学习机(ELM),时间序列(ARMA)和BP神经网络,进行仿真实验,结果表明相对于传统预测模型,所采用的LSTM网络具有更优的预测精确度和对远期预测的稳定性.最后利用预测结果计算出更能直观反映出道路交通拥堵情况的TSI指数,为人们提供了准确的交通状态预测.     

多行程车辆路径问题中变邻域搜索算法的应用

多行程车辆路径问题是标准车辆路径问题的一个变体,每个车辆在运行期间可以使用不止一次.对于这种NP-HARD问题,提出了一个改进变邻域搜索算法并设计了四个邻域结构用于求解和制定多行程路径问题的调度规划.算法测试了一组标准实例问题,获得的解决方法与文献中提出的三种不同数据集进行比较计算证明,算法提供了较高质量的求解结果.最后采用三个标准函数进行数值计算,与PSO和GA算法进行比较证明,提出的VNS算法虽然运行花费时间较长,但是达到全局收敛性的比率和全局收敛性都远超其他两种算法.     

多目标定位路线问题模型及禁忌搜索算法研究

在一个完整的物流系统中,总费用和工作量平衡是影响决策的两个基本标准,拓展对员工重要的公平原则-工作量平衡,并权衡工作量与总费用之间的平衡.在此基础上,考虑了带路线分配决策的多目标定位-路线问题,建立了数学模型,采用禁忌搜索算法对模型的求解,为检验车辆多次使用的效果,针对算法设计了联立和序贯两种不同的车辆路线分配形式.算法分析结果表明:区域特征在区分算法两种形式的性能时具有重要的地位.     

基于模式搜索的轮胎压力监测方法研究

提出了一种含有两个参数的轮胎纵向力模型,并通过该模型利用基于模式搜索的最小二乘法对轮胎的纵向刚度和滚动半径进行了动态估计.该算法应用于车辆的实际试验数据,结果证明轮胎的压力与轮胎纵向刚度近似成反比,这为利用轮胎纵向刚度监测轮胎压力变化,提供了可靠的理论依据.     

改进蚁群算法优化周期性车辆路径问题

周期性车辆路径问题(PVRP)是标准车辆路径问题(VRP)的扩展,PVRP将配送期由单一配送期延伸到T(T＞1)期,因此,PVRP需要优化每个配送期的顾客组合和配送路径。由于PVRP是一个内嵌VRP的问题,其比标准VRP问题更加复杂,难于求解。本文采用蚁群算法对PVRP进行求解,并提出采用两种改进措施——多维信息素的运用和基于扫描法的局部优化方法来提高算法的性能。最后,通过9个经典PVRP算例对该算法进行了数据实验,结果表明本文提出的改进蚁群算法求解PVRP问题是可行有效的,同时也表明两种改进措施可以显著提高算法的性能。     

多行程车辆路径问题和配送中心定位问题的研究

为了同时解决多行程车辆路径问题和配送中心的定位问题,首先开发了一个以最小化总成本为目标的数学模型,其中总成本包括运输成本和车辆启动成本.然后设计了一个启发式算法解决这个问题,包括三个阶段:第一阶段是找到初始定位并进行路线安排,第二阶段采用模拟退火(SA)的逻辑和交换算法来获得更好的路线,最后阶段是改善由模拟退火算法中当前温度控制的位置.通过标准样例进行的实验结果表明,该算法可以更好地获得一个配送中心定位和有效的相关路线安排.最后,数值实验指出:1)选择不同类型行程的配送方式取决于每辆车的启动成本和单位距离的运输成本;2)使用大容量车辆可以更好地减少运输距离.3)增加服务时间可以有效地减少所需车辆的数量,这三个结果对于多行程车辆路径问题和配送中心的定位问题的管理决策都具有一定的实用价值.     

基于主成分-灰色关联度的城市暴雨灾害应对能力评价:以江苏省13个地级城市为例

对城市暴雨灾害应对能力进行评价是城市暴雨灾害管理的重要手段.以江苏省13个地级城市为研究对象,基于2010-2013年的面板数据,综合运用主成分与灰色关联度相结合的分析方法,从日常防御、灾前预警、灾中应急和灾后恢复4个方面评价江苏省城市暴雨灾害应对能力.研究发现,江苏省城市暴雨灾害应对能力整体水平逐年提升,但是扬州和泰州小幅波动并略有下降;全省城市暴雨灾害应对能力具有明显的空间分布特征,呈现出由南向北能力递减的演变趋势.研究表明,该方法切实有效地解决了城市暴雨灾害应对能力的评估问题.     

有顾客需求变动的车辆调度干扰管理研究

为解决来自顾客需求变动所引发的物流配送干扰问题,提出基于干扰管理思想构建扰动恢复策略与方案.应用虚拟单车场实现了车辆调度扰动恢复问题转化,提出了车辆调度扰动恢复策略以及扰动度量方法,以作为车辆调度干扰管理建模的基础;对有顾客需求变动的车辆调度干扰管理问题,分析了顾客需求变动造成的扰动并进行了辨识;建立了相应的干扰管理模型,并提出了归一化处理办法对VRPTW问题进行有效兼容;结合干扰管理模型的特点,改进了基于顾客的编码表示方法;根据干扰管理思想,设计了遗传算法对干扰管理模型进行了求解.最后通过实验验证了模型与算法的有效性.     

基于能量增强的运动车辆车标的“四阶段法”定位技术研究

基于视频图像的运动车辆识别系统主要是由汽车牌照识别和汽车类型识别两大核心技术构成,它在智能交通领域中有着广泛的应用,同时也是计算机视觉、图像处理和模式识别等交叉学科研究的热门课题,运动车辆车标定位是车辆类型识别的核心技术之一,提出了基于能量增强的运动车辆车标定位的"四阶段法"的概念,介绍了运动车辆车标定位的方法和关键技术,对不同天气状况下不同类型的运动车辆进行了车标定位实验,实验结果表明该方法具有较强的科学依据和实用价值.     

半车车辆-道路耦合动力分析模型的研究与应用

在不平整路面上行驶的车辆会产生道路与车辆的耦合振动.为简化分析,先将车辆和道路视为两个子系统,然后再耦合为一个整体系统.车辆采用1/2车辆4自由度模型,道路采用弹性多层体系有限元模型,通过轮胎与路面的位移相容条件,建立车辆-道路整体耦合动力分析模型,推导出了系统的动力平衡方程组,并探讨了方程组的解耦方法,提出了反映车辆运行品质和道路设计参数的理论计算指标,为车-路耦合动力分析的深入研究,以及系统研究道路结构设计参数与车辆运行品质的相互关系提供理论基础和求解依据.此外,通过解耦和编程分析,还对沥青路面和水泥混凝土路面两种典型路面结构的运行品质指标进行了分析和评判,以期为不同路面结构设计和分析提供参考.     

应急物资联合配送决策研究

救援物资的高效投放是应急物流的关键活动之一。考虑到灾后救援车辆短缺,分配时受灾群众产生非理性心理,以及灾后道路状况对军、地救灾车辆速度产生影响等因素,本文从军民融合的角度出发,根据军地运输车辆特点,设计科学合理的车辆分配方案与运输路线。论文研究构建了基于车辆分配公平性和应急配送及时性的“分配-运输”集成决策模型,设计改进后的NSGA-Ⅱ算法进行求解,最后通过算例验证本算法的有效性和合理性,并通过调节军民运输车辆数,找到应急救援过程中的最佳军、地车辆比例,从而为灾后各灾备中心的车辆调度决策提供参考。     

基于粒子群算法的汽车租赁短期车辆配置问题研究

合理的车辆配置与调度是租车公司运营管理考虑的主要问题之一,也是提高租车公司的租车率和收益的有效途径。针对目前我国租车公司普遍缺乏历史数据、预定提前期短、租期短、门店间距离短等主要运营特点,本文将租车公司运营中频繁而复杂的短期车辆配置问题作为研究对象,提出单日的车辆配置方法,构建随机期望模型,并采用合理的方法分解模型,选择粒子群算法对子模型进行求解,并用数值算例验证了该方法的可行性与效果。该方法能够帮助租车公司管理者做出正确的决策,在提升顾客满意度的同时,提高租车率和租车公司的收益。     

Probabilistic programming models for traffic incident management operations planning

This paper proposes mathematical programming models with probabilistic constraints in order to address incident response and resource allocation problems for the planning of traffic incident management operations. For the incident response planning, we use the concept of quality of service during a potential incident to give the decision-maker the flexibility to determine the optimal policy in response to various possible situations. An integer programming model with probabilistic constraints is also proposed to address the incident response problem with stochastic resource requirements at the sites of incidents. For the resource allocation planning, we introduce a mathematical model to determine the number of service vehicles allocated to each depot to meet the resource requirements of the incidents by taking into account the stochastic nature of the resource requirement and incident occurrence probabilities. A detailed case study for the incident resource allocation problem is included to demonstrate the use of proposed model in a real-world context. The paper concludes with a summary of results and recommendations for future research.     

Planning vehicle transhipment in a seaport automobile terminal using a multi-agent system

A multi-agent system (MAS) for supporting the planning of transhipments of imported finished vehicles via a seaport is presented. The focus is on storage allocation, i.e. the allocation of parking areas for the temporary storage of vehicles, and on deployment scheduling, i.e. the allocation of drivers to the vehicles that have to be moved in the terminal area. These planning tasks, which in practice are usually carried out by different operators, are assigned to two different agent types. A further agent, the coordinator agent, is responsible for combining the local sub-plans into an overall plan in such a way that the demand for drivers in the planning period is minimised and balanced. The MAS is tested using randomly generated problem instances with different distributions of the manufacturer shares in the vehicle streams. The tests verify a certain robustness of the MAS with regard to changes in the problem data, in particular to the number of permanently employed drivers and the cost surcharge for hired drivers. In addition, the results highlight that the minimum overall (relative) costs of the drivers depends on the number of permanently employed drivers and on the level of the cost surcharge for hired drivers.     

An integrated FANP–MOLP for supplier evaluation and order allocation

In the face of acute global competition, supplier management is rapidly emerging as a crucial issue to any companies striving for business success and sustainable development. To optimize competitive advantages, a company should incorporate “suppliers” as an essential part of its core competencies. Supplier evaluation, the first step in supplier management, is a complex multi-criteria decision-making (MCDM) problem, and its complexity is further aggravated if the highly important interdependence among the selection criteria is taken into consideration. The objective of this paper is to suggest a comprehensive decision method for identifying top suppliers by considering the effects of interdependence among the selection criteria, as well as to achieve optimal allocation of orders among the selected suppliers.     

Simulation Study of an Automated Guided-vehicle System in a Yugoslav Hospital

A simulation study of an automated guided-vehicle system connected with an automated vertical transportation system in a large hospital is described. The aim of the study was to determine the allocation and size of storage space for vehicles, the minimum number of vehicles required, the accomplishment of a transportation-network timetable and the queueing situation in the system. This study is part of a simulation study of internal transportation in the new, 1100-bed University Hospital in Zagreb, Yugoslavia, which is under construction.     

A new mixed integer linear model for a rich vehicle routing problem with docking constraints

In this paper we address a rich vehicle routing problem that arises in real-life applications. Among other aspects we consider time windows, simultaneous delivery and pick-up at customer locations and multiple use of vehicles. To guarantee a coordinated material flow at the depot, we include the timed allocation of vehicles to loading bays at which the loading and unloading activities can occur. The resulting vehicle routing problem is formulated as a two-index vehicle-flow model which integrates the routing under real-life conditions and the assignment of vehicles to loading bays at the depot. We use CPLEX 11.0 to solve medium-sized instances that are derived from the extended Solomon test set. The selective implementation of preprocessing techniques and cutting planes improves the solver performance significantly.     

The use of fuzzy sets for resource allocation in an advance request vehicle brokerage system—a case study

Community transport vehicle brokerage operators match the needs of groups without minibuses with the spare capacity of groups which own minibuses. They take bookings in advance, and allocate vehicles without knowing what the total demand will be for travel at a certain time. An approach has been developed which ranks vehicles according to a sequence number which indicates how ‘difficult to book’ they are, and presents the most difficult to book to the broker first, with the intention of maintaining the maximum flexibility in the system. The concept difficult to book is not easy to define exactly. Fuzzy set theory provides a mechanism for dealing with imprecise concepts, and through fuzzy logic for reasoning about such concepts. Fuzzy inferencing systems consist of linguistic variables, fuzzy sets, rules, and methods to combine the sets and produce a final result. Fuzzy logic has been used to provide a means of automating the allocation of sequence numbers to vehicles. Three different approaches to developing the rules for this system have been tested, by simulating bookings, and the one which produced the best results was based on a knowledge acquisition process involving booking staff. This approach resulted in the allocation of the highest number of trips and in leaving the least hours of travel unallocated.     

基于收益管理和仿真的第三方仓储资源分配

针对不确定市场需求条件下第三方仓储资源的能力规划与分配问题,构建随机数学规划模型,理论分析证明了最优资源分配量的存在性,并指出最优资源分配量是单位资源成本的递减函数、单位资源收益和单位损失成本的递增函数。鉴于解析求解的复杂性,基于收益管理思想,结合离散事件仿真技术和响应曲面法,提出一种新的分析求解框架:收益管理用于细分顾客、构建资源分配策略,仿真模型刻画系统随机特性并评估系统绩效指标,响应曲面法则优化分配策略并探寻绩效改进方向。案例研究和仿真实验结果显示,根据顾客类别分配仓储能力的策略优于传统的先到先服务策略,收益管理、响应曲面法与仿真的综合集成,能够提高系统收益,从而使本文所提方法体系得到了有效验证。     

Multi-objective metaheuristics for a location-routing problem with multiple use of vehicles on real data and simulated data

We address an integrated logistic system where decisions on location of depot, vehicle routing and assignment of routes to vehicles are considered simultaneously. Total cost and workload balance are common criteria influencing decision-making. Literature on location-routing problems addressed the location and vehicle routing decisions with a common assumption of assigning one route to one vehicle. However, the cost of acquiring vehicles (and crew) is often more significant than the routing cost. This notion of assigning several routes to a vehicle during the routing procedure is explored in our integrated model. We apply metaheuristics of tabu search and simulated annealing on real data and simulated data, to compare their performances under two versions: simultaneous or sequential routes assignment to vehicles. A new statistical procedure is proposed to compare two algorithms on the strength of their multi-objective solutions. Results show that the simultaneous versions have advantage over the sequential versions in problems where routes are capacity-constrained, but not in the time dimension. The simultaneous versions are also more effective in generating non-dominated solutions than the sequential versions.