2017 MCM Problem A: Managing The Zambezi River
The Kariba Dam on the Zambezi River is one of the larger dams in Africa. Its construction was controversial, and a 2015 report by the Institute of Risk Management of South Africa included a warning that the dam is in dire need of maintenance. A number of options are available to the Zambezi River Authority (ZRA) that might address the situation. Three options in particular are of interest to ZRA:
(Option 1) Repairing the existing Kariba Dam,
(Option 2) Rebuilding the existing Kariba Dam, or
(Option 3) Removing the Kariba Dam and replacing it with a series of ten to twenty smaller dams along the Zambezi River.
There are two main requirements for this problem:
Requirement 1 ZRA management requires a brief assessment of the three options listed, with sufficient detail to provide an overview of potential costs and benefits associated with each option. This requirement should not exceed two pages in length, and must be provided in addition to your main report.
Requirement 2 Provide a detailed analysis of Option (3) - removing the Kariba Dam and replacing it with a series of ten to twenty smaller dams along the Zambezi river. This new system of dams should have the same overall water management capabilities as the existing Kariba Dam while providing the same or greater levels of protection and water management options for Lake Kariba that are in place with the existing dam. Your analysis must support a recommendation as to the number and placement of the new dams along the Zambezi River.
In your report for Requirement 2, you should include a strategy for modulating the water flow through your new multiple dam system that provides a reasonable balance between safety and costs. In addition to addressing known or predicted normal water cycles, your strategy should provide guidance to the ZRA managers that explains and justifies the actions that should be taken to properly handle emergency water flow situations (i.e. flooding and/or prolonged low water conditions). Your strategy should provide specific guidance for extreme water flows ranging from maximum expected discharges to minimum expected discharges. Finally, your recommended strategy should include information addressing any restrictions regarding the locations and lengths of time that different areas of the Zambezi River should be exposed to the most detrimental effects of the extreme conditions.
Translation:
赞比西河上的卡里巴水坝是非洲较大的水坝之一。它的建设是有争议的,南非风险管理研究所的2015年报告包括一个警告-大坝是急需维护。赞比西河管理局(ZRA)可提供若干选择,以解决这一问题。 ZRA特别感兴趣的有三个选项:
这个问题有两个主要要求:
要求2对选项(3)进行详细分析 - 删除Kariba水坝,并用赞比西河沿岸一系列十至二十个较小的水坝替代。这个新的水坝系统应该与现有的Kariba水坝具有相同的整体水管理能力,同时为现有的水坝提供与卡里巴湖相同或更高水平的保护和水管理选择。您的分析必须支持关于沿赞比西河新坝的数量和位置的建议。
在您的要求2报告中,您应该包括一个策略,用于调节通过您的新多坝系统的水流,从而在安全和成本之间提供合理的平衡。除了解决已知或预测的正常水循环,您的战略应为ZRA经理提供指导,解释和证明应当采取的行动,以正确处理应急水流情况(即洪水和/或长期低水位状况)。您的策略应为从最大预期排放到最小预期排放的极端水流提供具体指导。最后,您的建议战略应包括解决对赞比西河不同地区暴露于极端条件最有害影响的位置和时间长度的任何限制的信息。
####参考论文 [非洲水资源及利用现状_李淑芹.pdf](https://share.weiyun.com/25bfbf1b2b2759fc3c3f4c4652c55ad0) [赞比西河流域水库实施环境适应性管理.pdf](https://share.weiyun.com/7725a34960ad9e7cf1f0e9a5f66dd45b)
####个人思路 仅供参考 a warning that the dam is in dire need of maintenance 考虑到题目说明是需要maintenance,保养,那么成本consideration 是一定的。 其实题目已经给了我们一些参考信息,给出了三种可行的方案。我们要做的,就是评估,预测成本和收益的关系,那么答案自然而出。
- 构建Cost-benefit Prediction model 成本效益预测模型
- 考虑 k-NN ,或者Random forest
#### 关于 water management _providing the same or greater levels of protection and water management options _ [RIVER WATER QUALITY MANAGEMENT USING MATHEMATICal modelling.pdf](https://share.weiyun.com/8636fd1de11beae42eed6003c25da6a7) ![Model](http://img.blog.csdn.net/20170121032038346?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvemhhbmd0ZW5neXVhbjIz/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast) ![](http://img.blog.csdn.net/20170121032056346?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvemhhbmd0ZW5neXVhbjIz/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast) ![](http://img.blog.csdn.net/20170121032115228?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvemhhbmd0ZW5neXVhbjIz/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
##Problem B: Merge After Toll
2017 MCM : Linked from github.com
微博:分享技术,分享生活,欢迎交流
Github:Baidu.Inc Mobile Security
CSDN :技术分享
Multi-lane divided limited-access toll highways use “ramp tolls” and “barrier tolls” to collect tolls from motorists. A ramp toll is a collection mechanism at an entrance or exit ramp to the highway and these do not concern us here. A barrier toll is a row of tollbooths placed across the highway, perpendicular to the direction of traffic flow. There are usually (always) more tollbooths than there are incoming lanes of traffic (see former 2005 MCM Problem B). So when exiting the tollbooths in a barrier toll, vehicles must “fan in” from the larger number of tollbooth egress lanes to the smaller number of regular travel lanes. A toll plaza is the area of the highway needed to facilitate the barrier toll, consisting of the fan-out area before the barrier toll, the toll barrier itself, and the fan-in area after the toll barrier. For example, a three-lane highway (one direction) may use 8 tollbooths in a barrier toll. After paying toll, the vehicles continue on their journey on a highway having the same number of lanes as had entered the toll plaza (three, in this example).
Consider a toll highway having L lanes of travel in each direction and a barrier toll containing B tollbooths (B > L) in each direction. Determine the shape, size, and merging pattern of the area following the toll barrier in which vehicles fan in from B tollbooth egress lanes down to L lanes of traffic. Important considerations to incorporate in your model include accident prevention, throughput (number of vehicles per hour passing the point where the end of the plaza joins the L outgoing traffic lanes), and cost (land and road construction are expensive). In particular, this problem does not ask for merely a performance analysis of any particular toll plaza design that may already be implemented. The point is to determine if there are better solutions (shape, size, and merging pattern) than any in common use.
Determine the performance of your solution in light and heavy traffic. How does your solution change as more autonomous (self-driving) vehicles are added to the traffic mix? How is your solution affected by the proportions of conventional (human-staffed) tollbooths, exact-change (automated) tollbooths, and electronic toll collection booths (such as electronic toll collection via a transponder in the vehicle)?
#### Link 2005 MCM Problem B (Not Problem-B 2017)
Heavily-traveled toll roads such as the Garden State Parkway , Interstate 95, and so forth, are multi-lane divided highways that are interrupted at intervals by toll plazas. Because collecting tolls is usually unpopular, it is desirable to minimize motorist annoyance by limiting the amount of traffic disruption caused by the toll plazas. Commonly, a much larger number of tollbooths is provided than the number of travel lanes entering the toll plaza. Upon entering the toll plaza, the flow of vehicles fans out to the larger number of tollbooths, and when leaving the toll plaza, the flow of vehicles is required to squeeze back down to a number of travel lanes equal to the number of travel lanes before the toll plaza. Consequently, when traffic is heavy, congestion increases upon departure from the toll plaza. When traffic is very heavy, congestion also builds at the entry to the toll plaza because of the time required for each vehicle to pay the toll.
Make a model to help you determine the optimal number of tollbooths to deploy in a barrier-toll plaza. Explicitly consider the scenario where there is exactly one tollbooth per incoming travel lane. Under what conditions is this more or less effective than the current practice? Note that the definition of "optimal" is up to you to determine.
收取过路费(多车道划分限制访问的收费公路)使用“坡道收费”和“障碍收费”来收取驾驶员费用。斜坡收费是在高速公路的入口或出口匝道处的收费,并且这些不关心我们在这里。障碍收费是一排跨过高速公路的收费站,垂直于交通流的方向。通常会有(总是)比交通车道更多的收费站(见前2005年MCM-B)。因此,当驶出收费站时,车辆必须从较大数量的收费站出口车道“散开进入”到较少数量的常规行驶车道。收费广场是高速公路需要用于便于障碍收费的区域,包括在障碍收费之前的未散开插口区域,收费路径本身以及收费路径之后的分离子路段区域。例如,三车道高速公路(一个方向)可以在障碍通行费中使用8个收费站。在支付了费用之后,车辆在具有与进入收费广场相同数量的车道(在该示例中为三个)的高速公路上继续行驶。
考虑在每个方向上具有L个行驶车道的收费高速公路和在每个方向上包含B个收费站(B> L)的障碍通行费。确定跟随收费障碍的区域的形状,尺寸和合并模式,其中车辆从B过街出口车道下行到L个车道。在您的模型中纳入的重要注意事项包括事故预防,吞吐量(每小时通过广场末端加入L外出车道的车辆数量)和成本(土地和道路建设昂贵)。特别地,该问题不仅仅要求可能已经实现的任何特定收费广场设计的性能分析。重点是确定是否有比任何常用的更好的解决方案(形状,大小和合并模式)。
确定您的解决方案在轻和重的流量的性能。随着更多无人驾驶(辅助驾驶)车辆添加到交通组合中,您的解决方案如何改变?您的解决方案如何影响常规(人员配备)收费站,精确更换(自动)收费站和电子收费站(例如通过车辆中的应答器收集电子费用)的比例?
###个人思路 仅供参考:
对于此题,首先需要从侧面了解美国高速公路具体收费模式,不能完全使用中国收费站模式套用。
对于纵横交错,非常发达的美国高速公路网来说,收费的便捷性与效率应显得尤为重要,而且,据我所知,90%的美国高速公路不收费。因为美国人交的税里h和汽油费里就有一部分是专门用于高速公路的新建与维护。
####目的
- 确定跟随收费障碍的区域的形状,尺寸和合并模式(事故预防,吞吐量和成本)
- 构建更好的形状,大小和合并模式模型
- 无人车或者辅助驾驶(谷歌无人车L4,特斯拉L3)对模型的影响
- 构建的模型对『常规(人员配备)收费站,精确更换(自动)收费站和电子收费站(例如ETC)的比例?』带来怎样的影响?
####思路1
- 考虑车流量情况[metrocosm.com-data] (http://metrocosm.com/get-the-data/#interstatetraffic) 构建机器学习模型,通过数据的训练,学习到最优的 features' weight. (其实,题目中已告诉我们features)
- Time-Cost: 根据模型车流量的输入,最优化(损失时间Time-Cost,包括车辆通过后回合的时间cost),得出整体最优的形状,尺寸和合并模式模型参数。
- Price-Cost : 同时,模型还有一个Price-Cost,成本,同时达到成本的低于某一阈值或收敛值
- 对于合并模式 feature,过多的并道会增加 Cost-time,过少甚至一条收费通道,在车流量过大的情况下,导致通过车速过慢,也会增加Cost-time. 对于 combine-feature ,可以根据模型训练量化出 feature 的 weight.
####思路2
- 进一步看问题,可以转化为网络流量问题
- 参看Frank-Wolfe算法,通过目标函数近似线性化,迭代,可行下降方向进行一维搜索。求解最优点。
####思路3
- 排队论,或者有人叫做"服务系统设计",整体思路和#1类似
- 粗略介绍下:一个排队系统的好坏要以顾客与服务机构两方面的利益为标准。就顾客来说总希望等待时间或逗留时间越短越好,从而希望服务台个数尽可能多些但是,就服务机构来说,增加服务台数,就意味着增加投资,增加多了会造成浪费,增加少了要引起顾客的抱怨甚至失去顾客。
- 根据到达间隔时间与服务时间的分布,泊松分布;负指数分布;爱尔兰分布; 计算最优服务台数量! 参考下图
####思路4 交通流的特点在于动态、随机、反馈、自适应、多变为主导,以非线性复杂性。可以使用元胞自动机模拟交通流.
####**模型对人员配备,自动收费站和电子收费站(例如ETC)的影响** 1、分散式 该方式通常在互通立交的各个匝道设置收费系统,或者多个匝道设置一套收费系统。分散式的优势在于避开了车流的交叉通行,缩短车辆运行时间,同时缓解了收费站附近的交通阻塞,提高了收费站设计和选址的可行性;缺陷在于收费系统设施和人员不集中,建设费用较高,收费可控性较差,因此分散式在实际中较少使用。 2、集中式 集中式收费系统把每个立交的匝道进出口设置在同一位置,然后在该处设置双向收费系统,连接了收费路段和普通道路。与分散式相比,该方式的优势在于对收费系统的 设备和人员进行集中控制和管理,提高了系统工作效率;缺陷依然明显,由于匝道出入口必须设置在同一位置,对立交的规划设计产生了干扰,而且出入口车流集中,道路车流量较大时,极易产生交通拥堵,影响了道路的通行能力。
####收费效率: (1)人工收费 人工收费是比较传统的收费方式,收费的整个流程全部由收费人员人工完成。收费人员在收费站内对缴费车辆的不同类型进行收费,手续完成即可抬杆放行。人工收费最显著的特点是收费系统的建设投资较少,收费流程简便易行,缺点是可能会出现逃费和漏费现象,另外系统的工作效率也大打折扣。 (2)半自动收费 半自动收费是在传统人工收费的基础上引进计算机辅助设施,二者一同完成车辆缴费流程。通常收费人员负责收费工作,计算机用于车辆识别、费用结算、收据的输出打印等工作。由于有了计算机的介入,收费系统可以对各个收费通道进行集中控制,也能对相关路段进行联网计费。因此,半自动收费可以在一定程度上消除车辆逃费漏费现象的发生,也可以对收费车辆进行信息反馈,缺点是收费人员的存在也会导致系统的收费 效率下降。 (3)自动收费 自动收费方式也就是ETC电子不停车收费系统,该系统采用了计算机和网络通信技术,车辆的整个缴费过程不需要减速停车即可完成,因为ETC系统能够自动实现车辆的识别、费用汇总结算和监控过程,这种方式特别适合封闭式和开放式收费系统使用。
####关于自动驾驶 _How does your solution change as more autonomous (self-driving) _ 关于自动驾驶,我还是蛮感兴趣的,我厂L4,L3部门正好是做Auto-Car 上半年我也了解了些技术点。 #####简单介绍自动驾驶 我指 L4, 可以做到感知周边的环境(道路线,车辆,并线,行人等等) 通过已有的训练模型得出最准确的行使方案决策,通过最优化算法,计算并线,转弯等行车最佳方案,控制系统执行。
added to the traffic mix? 出题人还是想知道关于自动驾驶和人驾驶 mix 的情况,在我看来,需要分析 mix 后的影响 features 是哪些,能影响的范围程度需要分析. 我认为: 1、影响直线行驶的效率通行效率,准确的说是提升通行效率,根据研究自动驾驶在流比较密的时候,会保留一定的车距,减少突然刹车次数。比正常人驾驶习惯更好(关于)。如图 关于这一点,注意,参赛同学需要多找新文献,引用【】 说明下。 2、还有,应该考虑带来的安全性 feature 的变化,根据现有的研究,自动驾驶能够提升整体的驾驶安全性(主要根据,驾驶员驾驶员生理及心理品质,闪光融合临界值测定,检测驾驶员注意程度及疲劳等主要指标)
####模型评价 Determine the performance of your solution in light and heavy traffic. 机器学习的好办了,收敛速度,re-call 召回率,准确率等。
Important considerations to incorporate in your model include accident prevention 需在模型中考虑这个 feature,邻近开放应急收费通道,邻近路段交通流疏导、车速控制等方式可以考虑。
###参考示意图
###barrier tolls
###ramp tolls
####相关论文
A Research on the Intra-Regional Accessibility and Eco- nomic Development in the Cardi City Region Toll Plaza Merging Traffic Control for Throughput Maximization 基于节能减排的高速公路瓶颈处拥挤车流控制技术研究.pdf 区域交通导向发展_TOD_模式研究_陈磊磊.pdf 数据:新泽西每日汽车流量.pdf
高速公路收费站模型优化设计的研究 排队论在公路收费站服务台设计及管理中的应用 基于排队论的高速公路服务区停车位合理规模的计算_王殊 排队论在高速公路收费系统中的应用 基于MultiGen的高速公路交通流仿真技术研究 基于缓堵的主线收费站设计关键参数研究 基于元胞自动机的高等级公路复式收费站能耗仿真与应用 基于M_G_K排队模型的北京地区高速公路收费站通行能力研究 北京地区高速公路收费站通行能力研究
参考如下基础信息
在19世纪中后期,西部的许多州,比如加利福尼亚、内华达等,修建了许多私人的收费道路。进入二十世纪后,由于汽车的普及以及州际、州内高速公路的兴起,绝大部分收费道路均由州政府接手管理,在州的范围内,收费有了统一的标准。
根据运输部高速公路政策资讯办公室(Office of Highway Policy Information)的统计,如果将公路、桥梁、隧道、摆渡收费点全部统计在内的话,截至2011年1月1日,全国38个州有收费站,12个州没有任何收费站。全国的收费里程为8633公里。
根据联邦高速公路管理署的不完全统计,美国现有州际高速公路近7.6万公里,州内、郡县高速公路19.5万公里。以此来看,在27.1万公里的高速公路上,8633公里的路段需要收费,所占比例为3.18%,可见还是比较少的。难怪许多到美国旅游的外国人,特别到是西部、中部、南部旅游的人,会错以为美国的高速公路是完全免费的。
我们住在首都华盛顿附近的人,都知道最近环城495高速公路弗吉尼亚境内段又新增了一个快速道路,双向四车道,便于通勤的人上下班使用。不过,使用这个快车道是需要付费的(车上如果有三个人的话则免费),费率是动态的,根据在这个快车道上行驶车辆的多少随时调整:车辆多,费用就往上调整;车辆少,费用就下降。这个浮动收费方式与绝大部分采用固定费率的高速公路不同,是个新生事物,尚未得到大多数驾驶者的认可,目前使用该快速道路的车辆还比较少。
为了鼓励更多的人使用这条耗资14亿美元、花了四年多的功夫才建成的快车道,上个周末道路管理部门免费让民众使用,希望吸引更多的人。这个快车道的特点是不但没有人工收费站,连收费点也没有,全部使用高空电子收费器,从经过的汽车电子付费卡上自动收取费用,这样就避免了汽车接近收费点时车速减慢或大排长龙的不便。
美国高速交费方式
美国高速公路,收费有两种方式:一种是电子收费卡(E-PASS),经过收费站时自动感应,费用直接从卡上扣除;一种是现金支付,有投币式或专人收费。
[Barrier toll system](https://en.wikipedia.org/wiki/Barrier_toll_system)
[Ramp toll](https://en.wikipedia.org/wiki/Ramp_meter)
2017 MCM Problem C
MCM 2017 Link from github.com
微博:分享技术,分享生活,欢迎交流 Github:Baidu.Inc Mobile Security CSDN :技术分享
Traffic capacity is limited in many regions of the United States due to the number of lanes of roads. For example, in the Greater Seattle area drivers experience long delays during peak traffic hours because the volume of traffic exceeds the designed capacity of the road networks. This is particularly pronounced on Interstates 5, 90, and 405, as well as State Route 520, the roads of particular interest for this problem.
Self-driving, cooperating cars have been proposed as a solution to increase capacity of highways without increasing number of lanes or roads. The behavior of these cars interacting with the existing traffic flow and each other is not well understood at this point.
The Governor of the state of Washington has asked for analysis of the effects of allowing self-driving, cooperating cars on the roads listed above in Thurston, Pierce, King, and Snohomish counties. (See the provided map and Excel spreadsheet). In particular, how do the effects change as the percentage of self-driving cars increases from 10% to 50% to 90%? Do equilibria exist? Is there a tipping point where performance changes markedly? Under what conditions, if any, should lanes be dedicated to these cars? Does your analysis of your model suggest any other policy changes?
Your answer should include a model of the effects on traffic flow of the number of lanes, peak and/or average traffic volume, and percentage of vehicles using self-driving, cooperating systems. Your model should address cooperation between self-driving cars as well as the interaction between self- driving and non-self-driving vehicles. Your model should then be applied to the data for the roads of interest, provided in the attached Excel spreadsheet.
Some useful background information:
- On average, 8% of the daily traffic volume occurs during peak travel hours.
- The nominal speed limit for all these roads is 60 miles per hour.
- Mileposts are numbered from south to north, and west to east.
- Lane widths are the standard 12 feet.
- Highway 90 is classified as a state route until it intersects Interstate 5.
- In case of any conflict between the data provided in this problem and any other source, use the data provided in this problem.
Definitions: milepost: A marker on the road that measures distance in miles from either the start of the route or a state boundary. average daily traffic: The average number of cars per day driving on the road. interstate: A limited access highway, part of a national system. state route: A state highway that may or may not be limited access. route ID: The number of the highway. increasing direction: Northbound for N-S roads, Eastbound for E-W roads. decreasing direction: Southbound for N-S roads, Westbound for E-W roads.
问题C:“合作和导航”
2017年MCM
美国许多地区由于道路的数量限制,交通容量有限。例如,在大西雅图地区,由于交通量超过道路网络的设计容量,司机在交通高峰时段经历长时间的延误。这在5号,90号和405号州际公路以及520号国道在这个问题上显得尤为明显。
自动驾驶车己经作为增加公路能力而不用拓宽车道的方案提出,但是在这一点上,人们并未很好的理解自动驾驶车与现有交通工具的关系。
华盛顿州州长要求就自动驾驶车对于Thurston, Pierce和Snohomish等地交通的影响。(详情请参见提供的地图和Excel表格)。重点是当这种自动驾驶车的数量占比从10%增加到50%后,效果会有什么变化?到90%呢?其中是否存在均衡点?是否有明显的变化临界点?假设条件满足,是否应该为这些车设立专用车道?你的模型分析是否提出了改善政策的建议?
你需要构建一个模型,包括对车道流量,峰值、(和/或)平均交通量,以及自动驾驶车所占的比例 的协作系统。你的模型分析应该含有对自动驾驶车和非自动驾驶车之间的关系分析。你的模型分 析应该满足附件中Excel提供的指定道路数据。
你的MCM文件应该包括:1页摘要表,1-2页的州长回信,以及你的解决方案(不超过20页),整份文件最多23页,注:附录和参考文献不计入23页的限制当中。
一些有用的背景信息: 平均而言,每日交通量的8%发生在高峰旅行时间。 所有这些道路的名义速度限制为每小时60英里。 里程数从南到北,从西到东。 车道宽度是标准的12英尺。 高速公路90被分类为状态路线,直到它与州际5相交。 如果此问题中提供的数据与任何其他来源之间存在冲突,请使用 这个问题提供的数据。 定义: milepost:在路上测量距离,从路线的起点或a 状态边界。 平均每日交通量:在道路上行驶的平均每天的汽车数量。 州际公路:作为国家系统的一部分的有限进出高速公路。 国家路线:可能受限或不受限制的国家公路。 路由ID:高速公路的编号。 增加方向:N-S道北行,E-W道东行。 下降方向:N-S道南行,E-W道西行。
###个人思路 仅供参考 ####目标
- 模型: 车道流量,峰值,平均交通量,以及自动驾驶车所占的比例对协作系统的影响(利用道路数据)
- 自动驾驶车和非自动驾驶车之间的关系分析
####思路: 初步看来,此问题是网络流,交通流方面的问题,会涉及网络最大流算法知识。
宏观方法:不关心单个车辆的特性,利用流体力学的方法研究 道路上所有车辆的集体平均行为--流体力学连续模型 微观方法:从单个车辆的动力学行为入手,通过考察单个车辆之间的相互作用,推导出整个系统的统计性质--车辆跟驰模型、元胞自动机模型 介观方法:将交通流中的车辆看成具有相互作用的粒子,然后利用分子动理论对交通进行来研究--气体分子动理论模型
####交通流元胞自动机模型 采用离散的时间和空间变量,用一系列的演化规则来描述车辆间的微观相互作用,进而推出系统的动态演化规律。 #####假设:
- 道路被均分为若干元胞,每个元胞的大小为一个车长 (7米)或更小同一时刻,每个元胞或者为空,或者仅被一辆车占据;
- 车辆的位置和速度都是离散的整数值,速度更新过程也被离散为以1 秒为单位的跳跃式更新;
- 在绝大多数元胞自动机模型中,道路上车辆的速度更新是并行的,且单道上不允许超车;
这方面我们可以参考下:
元胞机模型—184号模型 和**(NS) 模型***
Nagel-Schreckenberg(NS) 模型
数学定义
在这一模型中,时间、空间以及速度都被 整数离散化 。道路被划分为离散的格子(即元胞),每个元胞或是为空,或是被一辆车占据,每辆车的速度可以取0,1,2...,vmax,vmax为最大速度。在 t->t+1 的过程中,模型按照如下规则进行演化:
加速:vn-> min( vn+1, vmax)
减速:vn-> min( vn-1, dn)
随机慢化:以概率p,vn-> max(vn-1, 0)
运动: xn-> xn+ vn
其中 dn= xn+1- xn- lveh,代表序列号为n的车辆与在其前方序号为n+1的车辆之间的距离 xn代表序列号为n的车辆的位置,vn为其速度,lveh为其车辆长度
**Velocity-effect(VE) 速度效应** [Ref]Li, Q. S. Wu and R. Jiang. Cellular automaton model considering the velocity effect of a car on the successive car. Phys. Rev. E 64, 066128 (2001)
以外的绝大多数元胞自动机模型中,有一个共同特征,在从 t->t+1 的时间步中,车辆速度更新规则只考虑了t时刻两车的距离,而没有记入前车运动的影响,即都把前车作为静止的粒子处理。由此造成模拟速度小于实际车辆速度。
规则将减速步改为:
vn-> min(vn+1, dn+ v‘ n+1)
v‘ n+1= min(vmax-1, vn+1, max(0, dn+1-1))
其中v‘ n+1是n+1车在 t->t+1 时间步里的虚拟速度。它由NS模型演化规则所能得到的最小可能速度。一方面考虑了前车的速度效应,另一方面又确保在模型的更新过程中不会发生撞车。
速度差依赖随机化
基于对无人驾驶和 L3跟车系统的研究,可以有效提高跟车速度差的随机化程度,
具有速度差的连续两辆车会影响到后者,Auto-car 修正了这一速度差,可以增加模型的容量,获得更多的模拟模型最大流量。
所以我想从无人车带来的随机速度差的补偿方向去研究
题目给的数据我感觉颗粒度不够,能做网络流的训练。。 我是从NS方向中的自动驾驶带来的连续跟车速度补偿角度出发的,这方面有很多国外的论文研究,建议多看看!
####无人驾驶对现有交通流的影响
还有一篇关于无人驾驶对于现有城市交通的影响的论文还没看完,之后再更新吧,这题如果按元胞自动机微观相关作用思路去解,还是很靠谱的,不过需要你有较强的实践能力,数据分析能力。Good Luck! 加油!
####关于无人车
1、**无人车专用道: ** 之前我有了解到我厂无人车团队的一些想法,在条件允许的情况下,建立一条无人车专用道。 我认为,关于 Electronic toll collection 电子收费系统,可以根据无人车的比例,建立适量的无人车专用道。算是一种思路吧,可以用模型、数据去验证.
2、**无人车车队: ** 还有这点也是一次我厂无人车团队 report 时提及的 ,基于自动驾驶的车队很有实践意义与商业价值(你可能不了解,无人车的车队跟车系统,在测试上能大大提高汽油的使用率,节油节省费用非常明显)
对于这点,我查阅资料,有这样的新闻,同样也验证了可能性。
Recent statements by Ford, Uber, BMW and others clearly show that fleets of self-driving cars will emerge early and have the potential to capture a significant share of individual motorized mobility.
基于这点,车队将寻求'最大化其所有车辆的吞吐量',而不是单个车的通行速率。『还有一点,如果无人驾驶车队的数量大量增长,更具团队strategies ,以及精巧的算法提高整体车队的粘度,或者叫密度~ 』
####关于无人车比例
how do the effects change as the percentage of self-driving cars increases from 10% to 50% to 90%? Do equilibria exist?
题目提及的10-percent,50-percent,90-percent 不知道是无意还是有意透露, 不过基于上面2条的分析,对于无人车对的比例增加,可以做如下分析:
- 首先,得明确一些现实已经存在的假设~『无人车车队可以根据传感器的数据分析,精确控制车队的运行策略,并上传给管理端』 『车队管理人员,或者城市交通管理者可以通过传感器获得的实时交通信息,以及无人车队的具体位置』
- 模型研究方向:
3. 无人车车队车辆超过10%-50%的交通流量规模时,保留一些车道用于无人驾驶车辆,因为它们比单个车辆具有更高效的移动性(有研究无人车车辆专用车道上的吞吐量可以是人驾驶车道的吞吐量的两倍) 4. 50%-90% 那说明大部分交通流量都是无人车辆,完全可以考虑控制红绿灯,交通信号灯的,来提供一半以上车辆的运行效率! 5. 90%以上意味着基本全城流量都是无人车,刚我们做了『能够预测在任何给定时间点的行程的实际持续时间,并且将旨在最小化引起严重拥塞的行程』的假设,那么,全局上看,可以实时政策干涉整体交通流量,例如,驱动无人车队改变路线策略,对空闲道路施行奖励或者对繁重压力的路段施行拥堵费的征收。 6. 既然我们已经打开思路,都立法,人为干预了,那么,更大胆的预测,无人车达到90%的比例,交通事故的出现概率将降低到极低的值,在北京的同学应该很明显的感觉体会到,由于交通事故造成的交通拥堵非常严重,而且时常很长!那么,你也可对这个feature 进行研究,增加在模型中。
Does your analysis of your model suggest any other policy changes? 题目也提及政策的变化
针对这点:我非常感兴趣,希望能和大家交流,有好的想法都可以提处理来!
参考文献:
多辆车优化速度权重的合作驾驶模型的数值模拟_刘玲玲.pdf
考虑相对流量影响的交通流合作驾驶格子模型_庹满先.pdf
Self-driving cars will change cities.pdf
Cooperative intersection collision avoidance in a constrained communication environment .pdf
####2017 ICM Problem D Problem D: Optimizing the Passenger Throughput at an Airport Security Checkpoint
Following the terrorist attacks in the US on September 11, 2001, airport security has been significantly enhanced throughout the world. Airports have security checkpoints, where passengers and their baggage are screened for explosives and other dangerous items. The goals of these security measures are to prevent passengers from hijacking or destroying aircraft and to keep all passengers safe during their travel. However, airlines have a vested interest in maintaining a positive flying experience for passengers by minimizing the time they spend waiting in line at a security checkpoint and waiting for their flight. Therefore, there is a tension between desires to maximize security while minimizing inconvenience to passengers.
During 2016, the U.S. Transportation Security Agency (TSA) came under sharp criticism for extremely long lines, in particular at Chicago’s O’Hare international airport. Following this public attention, the TSA invested in several modifications to their checkpoint equipment and procedures and increased staffing in the more highly congested airports. While these modifications were somewhat successful in reducing waiting times, it is unclear how much cost the TSA incurred to implement the new measures and increase staffing. In addition to the issues at O’Hare, there have also been incidents of unexplained and unpredicted long lines at other airports, including airports that normally have short wait times. This high variance in checkpoint lines can be extremely costly to passengers as they decide between arriving unnecessarily early or potentially missing their scheduled flight. Numerous news articles, including [1,2,3,4,5], describe some of the issues associated with airport security checkpoints.
Your Internal Control Management (ICM) team has been contracted by the TSA to review airport security checkpoints and staffing to identify potential bottlenecks that disrupt passenger throughput. They are especially interested in creative solutions that both increase checkpoint throughput and reduce variance in wait time, all while maintaining the same standards of safety and security. The current process for a US airport security checkpoint is displayed in Figure 1.
Zone A: o Passengers randomly arrive at the checkpoint and wait in a queue until a security officer can inspect their identification and boarding documents. Zone B: o The passengers then move to a subsequent queue for an open screening line; depending on the anticipated activity level at the airport, more or less lines may be open. o Oncethepassengersreachthefrontofthisqueue,theyprepareallof their belongings for X-ray screening. Passengers must remove shoes, belts, jackets, metal objects, electronics, and containers with liquids, placing them in a bin to be X-rayed separately; laptops and some medical equipment also need to be removed from their bags and placed in a separate bin. o Alloftheirbelongings,includingthebinscontainingtheaforementioned items, are moved by conveyor belt through an X-ray machine, where some items are flagged for additional search or screening by a security officer (Zone D). o Meanwhile the passengers process through either a millimeter wave scanner or metal detector. o Passengers that fail this step receive a pat-down inspection by a security officer (Zone D). Zone C: o The passengers then proceed to the conveyor belt on the other side of the X-ray scanner to collect their belongings and depart the checkpoint area. Figure 1: Illustration of the TSA Security Screening Process.
Approximately 45% of passengers enroll in a program called Pre-Check for trusted travelers. These passengers pay $85 to receive a background check and enjoy a separate screening process for five years. There is often one Pre-Check lane open for every three regular lanes, despite the fact that more passengers use the Pre-Check process. Pre-Check passengers and their bags go through the same screening process with a few modifications designed to expedite screening. Pre-Check passengers must still remove metal and electronic items for scanning as well as any liquids, but are not required to remove shoes, belts, or light jackets; they also do not need to remove their computers from their bags.
Data has been collected about how passengers proceed through each step of the security screening process. Click here to view the Excel data.
Your specific tasks are: a. Develop one or more model(s) that allow(s) you to explore the flow of passengers through a security check point and identify bottlenecks. Clearly identify where problem areas exist in the current process. b. Develop two or more potential modifications to the current process to improve passenger throughput and reduce variance in wait time. Model these changes to demonstrate how your modifications impact the process. c. It is well known that different parts of the world have their own cultural norms that shape the local rules of social interaction. Consider how these cultural norms might impact your model. For example, Americans are known for deeply respecting and prioritizing the personal space of others, and there is a social stigma against “cutting” in front of others. Meanwhile, the Swiss are known for their emphasis on collective efficiency, and the Chinese are known for prioritizing individual efficiency. Consider how cultural differences may impact the way in which passenger’s process through checkpoints as a sensitivity analysis. The cultural differences you apply to your sensitivity analysis can be based on real cultural differences, or you can simulate different traveler styles that are not associated with any particular culture (e.g., a slower traveler). How can the security system accommodate these differences in a manner that expedites passenger throughput and reduces variance? d. Propose policy and procedural recommendations for the security managers based on your model. These policies may be globally applicable, or may be tailored for specific cultures and/or traveler types.
In addition to developing and implementing your model(s) to address this problem, your team should validate your model(s), assess strengths and weaknesses, and propose ideas for improvement (future work).
Your ICM submission should consist of a 1 page Summary Sheet and your solution cannot exceed 20 pages for a maximum of 21 pages. Note: The appendix and references do not count toward the 20 page limit.
References: [1] http://www.wsj.com/articles/why-tsa-security-lines-arent-as-bad-as-youd-feared-1469032116 [2] http://www.chicagotribune.com/news/ct-tsa-airport-security-lines-met-20160823-story.html [3] http://www.cnn.com/2016/06/09/travel/tsa-security-line-wait-times-how-long/ [4] http://wgntv.com/2016/07/13/extremely-long-lines-reported-at-chicago-midway-airports-tsa-checkpoint/ [5] http://www.cnbc.com/2016/04/14/long-lines-and-missed-flights-fuel-criticism-of-tsa-screening.html
问题D:在机场安全检查站优化乘客吞吐量
继2001年9月11日美国发生恐怖袭击事件后,全世界的机场安全状况得到显着改善。机场有安全检查站,在那里,乘客及其行李被检查爆炸物和其他危险物品。这些安全措施的目的是防止乘客劫持或摧毁飞机,并在旅行期间保持所有乘客的安全。然而,航空公司有既得利益,通过最小化他们在安全检查站排队等候并等待他们的航班的时间,为乘客保持积极的飞行体验。因此,在希望之间存在最大化安全性同时最小化对乘客的不便的张力。
在2016年,美国运输安全局(TSA)受到了对极长线路,特别是在芝加哥的奥黑尔国际机场的尖锐批评。在此公众关注之后,TSA投资对其检查点设备和程序进行了若干修改,并增加了在高度拥堵的机场中的人员配置。虽然这些修改在减少等待时间方面有一定的成功,但TSA在实施新措施和增加人员配置方面花费了多少成本尚不清楚。除了在O'Hare的问题,还有在其他机场,包括通常有短的等待时间的机场不明原因和不可预测的长线的事件。检查站线路的这种高差异对于乘客来说可能是极其耗时的,因为他们决定要尽早到达,因为可能延迟错过他们的预定航班之间。许多新闻文章,包括[1,2,3,4,5],描述了与机场安全检查站相关的一些问题。
您的内部控制管理(ICM)团队已经与TSA签订合同,审查机场安全检查站和人员配置,以确定可能干扰乘客吞吐量的瓶颈。他们特别感兴趣的创意解决方案,既增加检查点吞吐量,减少等待时间的方差,同时保持相同的安全和安全标准。
美国机场安全检查点的当前流程如图1所示。
区域A:
o乘客随机到达检查站,并等待队列,直到安全人员可以检查他们的身份证明和登机文件。
区域B:
o然后乘客移动到打开的筛选线的后续队列;根据机场的预期活动水平,或多或少的线路可能开放。
o一旦乘客到达这个队列的前面,他们准备所有的物品用于X射线检查。乘客必须用液体去除鞋子,皮带,夹克,金属物体,电子产品和容器,将它们放置在单独的X射线箱中;笔记本电脑和一些医疗设备也需要从其袋中取出并放置在单独的容器中。
o他们的所有物品,包括包含上述物品的箱子,由传送带通过X光机移动,其中一些物品被标记,供安全人员(D区)进行额外的搜索或筛选。
o同时乘客通过毫米波扫描仪或金属探测器进行处理。
o未能通过此步骤的乘客接受安全官员(D区)的轻击检查。
C区:
o乘客然后前进到X射线扫描仪另一侧的传送带,收集他们的物品并离开检查站区域。
图1:TSA安全筛选过程的图示。
大约45%的乘客报名参加一个称为预检查信任旅行者的计划。这些乘客支付85美元,接受背景调查,并享受五年的独立筛选程序。尽管事实上更多的乘客使用预检查过程,但是每三条常规车道通常有一个预检查车道打开。预检查乘客和他们的行李经过相同的筛选过程,经过一些修改,以加快筛选。预检查乘客还必须移除扫描用的金属和电子物品以及任何液体,但不需要去除鞋子,皮带或灯罩;他们也不需要从他们的包里删除他们的电脑。
收集了关于乘客如何进行安全检查过程的每个步骤的数据。
您的特定任务是:
a.开发一个或多个模型,允许您通过安全检查点探索乘客流,并识别瓶颈。清楚地确定当前流程中存在哪些问题区域。
b.对当前流程开发两个或多个潜在修改,以提高旅客吞吐量并减少等待时间的差异。对这些更改进行建模,以演示修改如何影响流程。
c.众所周知,世界上不同的地方都有自己的文化规范,塑造了地方社会互动的规则。考虑这些文化规范如何影响你的模型。例如,美国人以深为尊重和优先考虑别人的个人空间而闻名,在别人的面前“切割,或者理解为剪切”当作是一种社会歧视。同时,瑞士人以集体效率为重点,中国人以优先个人效率而闻名。考虑文化差异如何影响乘客的过程通过检查点作为敏感性分析的方式。您应用于敏感性分析的文化差异可以基于真实的文化差异,或者您可以模拟与任何特定文化(例如,较慢的旅行者)无关的不同旅行者风格。安全系统如何以加快乘客吞吐量并减少差异的方式来适应这些差异?
d.根据您的模型为安全管理器提出政策和程序建议。这些策略可以是全球适用的,或者可以针对特定文化和/或旅行者类型来定制。
除了开发和实施您的模型来解决这个问题,您的团队还应该验证您的模型,评估优势和弱点,并提出改进建议(未来工作)。
####个人思路 仅供参考
个人觉得是一个排队论中的多服务排队系统方面的问题【 Multi-Server and Multi-Queue Systems】,但是特殊的,需要评估风险因子
参考文献:
Fair Operation of Multi-Server and Multi-Queue Systems.pdf
Airport Gate Scheduling for Passengers, Aircraft, and.pdf
A Discrete Event Simulation to model Passenger Flow in the Airport Terminal .pdf
Optimizing the Aviation Checkpoint Process to Enhance Security and Expedite Screening.pdf