2011_08_01-Market-Eval-Bus-Plan-Drive-China

合集下载

运筹学英汉词汇ABC

运筹学英汉词汇(0,1) normalized ――0-1规范化Aactivity ――工序additivity――可加性adjacency matrix――邻接矩阵adjacent――邻接aligned game――结盟对策analytic functional equation――分析函数方程approximation method――近似法arc ――弧artificial constraint technique ――人工约束法artificial variable――人工变量augmenting path――增广路avoid cycle method ――避圈法Bbackward algorithm――后向算法balanced transportation problem――产销平衡运输问题basic feasible solution ――基本可行解basic matrix――基阵basic solution ――基本解basic variable ――基变量basic ――基basis iteration ――换基迭代Bayes decision――贝叶斯决策big M method ――大M 法binary integer programming ――0-1整数规划binary operation――二元运算binary relation――二元关系binary tree――二元树binomial distribution――二项分布bipartite graph――二部图birth and death process――生灭过程Bland rule ――布兰德法则branch node――分支点branch――树枝bridge――桥busy period――忙期Ccapacity of system――系统容量capacity――容量Cartesian product――笛卡儿积chain――链characteristic function――特征函数chord――弦circuit――回路coalition structure――联盟结构coalition――联盟combination me――组合法complement of a graph――补图complement of a set――补集complementary of characteristic function――特征函数的互补性complementary slackness condition ――互补松弛条件complementary slackness property――互补松弛性complete bipartite graph――完全二部图complete graph――完全图completely undeterministic decision――完全不确定型决策complexity――计算复杂性congruence method――同余法connected component――连通分支connected graph――连通图connected graph――连通图constraint condition――约束条件constraint function ――约束函数constraint matrix――约束矩阵constraint method――约束法constraint ――约束continuous game――连续对策convex combination――凸组合convex polyhedron ――凸多面体convex set――凸集core――核心corner-point ――顶点（角点）cost coefficient――费用系数cost function――费用函数cost――费用criterion ; test number――检验数critical activity ――关键工序critical path method ――关键路径法（CMP ）critical path scheduling ――关键路径cross job ――交叉作业curse of dimensionality――维数灾customer resource――顾客源customer――顾客cut magnitude ――截量cut set ――截集cut vertex――割点cutting plane method ――割平面法cycle ――回路cycling ――循环Ddecision fork――决策结点decision maker决――策者decision process of unfixed step number――不定期决策过程decision process――决策过程decision space――决策空间decision variable――决策变量decision决－－策decomposition algorithm――分解算法degenerate basic feasible solution ――退化基本可行解degree――度demand――需求deterministic inventory model――确定贮存模型deterministic type decision――确定型决策diagram method ――图解法dictionary ordered method ――字典序法differential game――微分对策digraph――有向图directed graph――有向图directed tree――有向树disconnected graph――非连通图distance――距离domain――定义域dominate――优超domination of strategies――策略的优超关系domination――优超关系dominion――优超域dual graph――对偶图Dual problem――对偶问题dual simplex algorithm ――对偶单纯形算法dual simplex method――对偶单纯形法dummy activity――虚工序dynamic game――动态对策dynamic programming――动态规划Eearliest finish time――最早可能完工时间earliest start time――最早可能开工时间economic ordering quantity formula――经济定购批量公式edge ――边effective set――有效集efficient solution――有效解efficient variable――有效变量elementary circuit――初级回路elementary path――初级通路elementary ――初等的element――元素empty set――空集entering basic variable ――进基变量equally liability method――等可能性方法equilibrium point――平衡点equipment replacement problem――设备更新问题equipment replacing problem――设备更新问题equivalence relation――等价关系equivalence――等价Erlang distribution――爱尔朗分布Euler circuit――欧拉回路Euler formula――欧拉公式Euler graph――欧拉图Euler path――欧拉通路event――事项expected value criterion――期望值准则expected value of queue length――平均排队长expected value of sojourn time――平均逗留时间expected value of team length――平均队长expected value of waiting time――平均等待时间exponential distribution――指数分布external stability――外部稳定性Ffeasible basis ――可行基feasible flow――可行流feasible point――可行点feasible region ――可行域feasible set in decision space――决策空间上的可行集feasible solution――可行解final fork――结局结点final solution――最终解finite set――有限集合flow――流following activity ――紧后工序forest――森林forward algorithm――前向算法free variable ――自由变量function iterative method――函数迭代法functional basic equation――基本函数方程function――函数fundamental circuit――基本回路fundamental cut-set――基本割集fundamental system of cut-sets――基本割集系统fundamental system of cut-sets――基本回路系统Ggame phenomenon――对策现象game theory――对策论game――对策generator――生成元geometric distribution――几何分布goal programming――目标规划graph theory――图论graph――图HHamilton circuit――哈密顿回路Hamilton graph――哈密顿图Hamilton path――哈密顿通路Hasse diagram――哈斯图hitchock method ――表上作业法hybrid method――混合法Iideal point――理想点idle period――闲期implicit enumeration method――隐枚举法in equilibrium――平衡incidence matrix――关联矩阵incident――关联indegree――入度indifference curve――无差异曲线indifference surface――无差异曲面induced subgraph――导出子图infinite set――无限集合initial basic feasible solution ――初始基本可行解initial basis ――初始基input process――输入过程Integer programming ――整数规划inventory policy―v存贮策略inventory problem―v货物存储问题inverse order method――逆序解法inverse transition method――逆转换法isolated vertex――孤立点isomorphism――同构Kkernel――核knapsack problem ――背包问题Llabeling method ――标号法latest finish time――最迟必须完工时间leaf――树叶least core――最小核心least element――最小元least spanning tree――最小生成树leaving basic variable ――出基变量lexicographic order――字典序lexicographic rule――字典序lexicographically positive――按字典序正linear multiobjective programming――线性多目标规划Linear Programming Model――线性规划模型Linear Programming――线性规划local noninferior solution――局部非劣解loop method――闭回路loop――圈loop――自环（环）loss system――损失制Mmarginal rate of substitution――边际替代率Marquart decision process――马尔可夫决策过程matching problem――匹配问题matching――匹配mathematical programming――数学规划matrix form ――矩阵形式matrix game――矩阵对策maximum element――最大元maximum flow――最大流maximum matching――最大匹配middle square method――平方取中法minimal regret value method――最小后悔值法minimum-cost flow――最小费用流mixed expansion――混合扩充mixed integer programming ――混合整数规划mixed Integer programming――混合整数规划mixed Integer ――混合整数规划mixed situation――混合局势mixed strategy set――混合策略集mixed strategy――混合策略mixed system――混合制most likely estimate――最可能时间multigraph――多重图multiobjective programming――多目标规划multiobjective simplex algorithm――多目标单纯形算法multiple optimal solutions ――多个最优解multistage decision problem――多阶段决策问题multistep decision process――多阶段决策过程Nn- person cooperative game ――n人合作对策n- person noncooperative game――n人非合作对策n probability distribution of customer arrive――顾客到达的n 概率分布natural state――自然状态nature state probability――自然状态概率negative deviational variables――负偏差变量negative exponential distribution――负指数分布network――网络newsboy problem――报童问题no solutions ――无解node――节点non-aligned game――不结盟对策nonbasic variable ――非基变量nondegenerate basic feasible solution――非退化基本可行解nondominated solution――非优超解noninferior set――非劣集noninferior solution――非劣解nonnegative constrains ――非负约束non-zero-sum game――非零和对策normal distribution――正态分布northwest corner method ――西北角法n-person game――多人对策nucleolus――核仁null graph――零图Oobjective function ――目标函数objective( indicator) function――指标函数one estimate approach――三时估计法operational index――运行指标operation――运算optimal basis ――最优基optimal criterion ――最优准则optimal solution ――最优解optimal strategy――最优策略optimal value function――最优值函数optimistic coefficient method――乐观系数法optimistic estimate――最乐观时间optimistic method――乐观法optimum binary tree――最优二元树optimum service rate――最优服务率optional plan――可供选择的方案order method――顺序解法ordered forest――有序森林ordered tree――有序树outdegree――出度outweigh――胜过Ppacking problem ――装箱问题parallel job――平行作业partition problem――分解问题partition――划分path――路path――通路pay-off function――支付函数payoff matrix――支付矩阵payoff――支付pendant edge――悬挂边pendant vertex――悬挂点pessimistic estimate――最悲观时间pessimistic method――悲观法pivot number ――主元plan branch――方案分支plane graph――平面图plant location problem――工厂选址问题player――局中人Poisson distribution――泊松分布Poisson process――泊松流policy――策略polynomial algorithm――多项式算法positive deviational variables――正偏差变量posterior――后验分析potential method ――位势法preceding activity ――紧前工序prediction posterior analysis――预验分析prefix code――前级码price coefficient vector ――价格系数向量primal problem――原问题principal of duality ――对偶原理principle of optimality――最优性原理prior analysis――先验分析prisoner’s dilemma――囚徒困境probability branch――概率分支production scheduling problem――生产计划program evaluation and review technique――计划评审技术(PERT) proof――证明proper noninferior solution――真非劣解pseudo-random number――伪随机数pure integer programming ――纯整数规划pure strategy――纯策略Qqueue discipline――排队规则queue length――排队长queuing theory――排队论Rrandom number――随机数random strategy――随机策略reachability matrix――可达矩阵reachability――可达性regular graph――正则图regular point――正则点regular solution――正则解regular tree――正则树relation――关系replenish――补充resource vector ――资源向量revised simplex method――修正单纯型法risk type decision――风险型决策rooted tree――根树root――树根Ssaddle point――鞍点saturated arc ――饱和弧scheduling (sequencing) problem――排序问题screening method――舍取法sensitivity analysis ――灵敏度分析server――服务台set of admissible decisions(policies) ――允许决策集合set of admissible states――允许状态集合set theory――集合论set――集合shadow price ――影子价格shortest path problem――最短路线问题shortest path――最短路径simple circuit――简单回路simple graph――简单图simple path――简单通路Simplex method of goal programming――目标规划单纯形法Simplex method ――单纯形法Simplex tableau――单纯形表single slack time ――单时差situation――局势situation――局势slack variable ――松弛变量sojourn time――逗留时间spanning graph――支撑子图spanning tree――支撑树spanning tree――生成树stable set――稳定集stage indicator――阶段指标stage variable――阶段变量stage――阶段standard form――标准型state fork――状态结点state of system――系统状态state transition equation――状态转移方程state transition――状态转移state variable――状态变量state――状态static game――静态对策station equilibrium state――统计平衡状态stationary input――平稳输入steady state――稳态stochastic decision process――随机性决策过程stochastic inventory method――随机贮存模型stochastic simulation――随机模拟strategic equivalence――策略等价strategic variable, decision variable ――决策变量strategy (policy) ――策略strategy set――策略集strong duality property ――强对偶性strong ε-core――强ε-核心strongly connected component――强连通分支strongly connected graph――强连通图structure variable ――结构变量subgraph――子图sub-policy――子策略subset――子集subtree――子树surplus variable ――剩余变量surrogate worth trade-off method――代替价值交换法symmetry property ――对称性system reliability problem――系统可靠性问题Tteam length――队长tear cycle method――破圈法technique coefficient vector ――技术系数矩阵test number of cell ――空格检验数the branch-and-bound technique ――分支定界法the fixed-charge problem ――固定费用问题three estimate approach一―时估计法total slack time――总时差traffic intensity――服务强度transportation problem ――运输问题traveling salesman problem――旅行售货员问题tree――树trivial graph――平凡图two person finite zero-sum game二人有限零和对策two-person game――二人对策two-phase simplex method ――两阶段单纯形法Uunbalanced transportation problem ――产销不平衡运输问题unbounded ――无界undirected graph――无向图uniform distribution――均匀分布unilaterally connected component――单向连通分支unilaterally connected graph――单向连通图union of sets――并集utility function――效用函数Vvertex――顶点voting game――投票对策Wwaiting system――等待制waiting time――等待时间weak duality property ――弱对偶性weak noninferior set――弱非劣集weak noninferior solution――弱非劣解weakly connected component――弱连通分支weakly connected graph――弱连通图weighed graph ――赋权图weighted graph――带权图weighting method――加权法win expectation――收益期望值Zzero flow――零流zero-sum game――零和对策zero-sum two person infinite game――二人无限零和对策。

A-T Controls 31 Series 3-Way Flanged Direct Mount

31Series-2R-20230502Copyright 2013 A-T Controls, Inc.PneumaticElectricSee automated data sheets for pre-sized assembliesEasy to Automate!This 3-way ball valve offers diverting and mixing flow patterns, often eliminating the need for two valves. The full port design is easily automated and isavailable with various seat materials. The bolts on side flanges make for easy seat changes as necessary to accomodate your service. Available in L or T port conﬁgurations.Cincinnati, Ohio FAX (513) 247-5462********************3-Way 150# Flanged Direct Mount Ball ValveShaded boxes indicate standard flow path from factory unless otherwise specified by customer. Automated assemblies rotate counter-clockwise standard from the factory when energized. For L-Ports that would be L1 to L2 and T3 to T4 for T-ports. By specifying FCCW or DAR actuators, multiple flow patterns can be achieved to meet process requirements.FLOW PATTERNSCRN3-Way Flanged Ball Valve Full Port, L or T Option ANSI Class 150TFM™ is a trademark of Dyneon™, a 3M Company.Kalrez is a registered trademark of DuPont Performance Elastomers.# 1-1/2” THRU 3” = Qty. 12 pcs | 4” = Qty. 18 pcs | 6” = Qty. 30 pcs^ 1-1/2” THRU 3” = 4 pcs, 4” = 6 pcs., 6” = 12 pcsDIMENSIONS For ANSI Class 150 (IN)NOTE: At temperature, valves are limited by either the valve body/end cap pressure ratings, seat pressure ratings, or packing/stem seal/gaskets;whichever is lower.Published torquesare based on full differential pressure with clean water. Consult the Application Sizing Guide for assistance with sizing actuators.Cincinnati, Ohio FAX (513) 247-5462********************Actuators are sized based on clean/clear fluid.SERIES 31 3-Way 150# Flanged Direct MountFlanged 3-WayClass 150 Double Acting Assembly For operating temperatures in excess of 175° F with Buna-N seals in the actuator, an extended bracket is required. FKM seals in the actuator require an extended bracket for more than 300° F. Please consult factory for sizing information.Actuators are sized for clean liquid surface with a specific gravity of 1 and at 70° F.SAMPLE PART #31C-FX-150/2R3D-_ _ _(2) Valve Series (4) End Connection (6) Valve Size(5) Seat, Lining, & TrimMaterial(7) Actuator(8) Accessories/Options(3) Body/Ball/StemMaterial(9) Accessories(10) Special Designation Refer to Series 31 IOM for all repair kit, seat and gasket part numbers.See the last page of catalog for How To Order detail and options.See Automated Part Number Matrix for complete part number and options.31Series-2R-20230502Copyright 2013 A-T Controls, Inc.Flanged 3-WaySERIES 313-Way 150# Flanged Direct MountCincinnati, Ohio FAX (513) 247-5462********************Class 150 Spring Return AssemblySpecify flow pattern when ordering. See automated part number matrix for complete part number and options.Note: A number following the actuator model (Ex: 2R6S4), indicates the number of springs per side. For a standard (5 spring per side) actuator, the spring designation is omitted from the automated part number.Actuators are sized for clean liquid surface with a specific gravity of 1 and at 70° F.For operating temperatures in excess of 175° F with Buna-N seals in the actuator, an extended bracket is required. FKM seals in the actuator require an extended bracket for more than 300° F. Please consult factory for sizing information.SAMPLE PART #31C-FX-150/2R6S4-_ _ _(2) Valve Series (4) End Connection (6) Valve Size(5) Seat, Lining, & TrimMaterial(7) Actuator(8) Accessories/Options(3) Body/Ball/StemMaterial(9) Accessories(10) Special Designation Refer to Series 31 IOM for all repair kit, seat and gasket part numbers.See the last page of catalog for How To Order detail and options.See Automated Part Number Matrix for complete part number and options.31Series-2R-20230502Copyright 2013 A-T Controls, Inc.Flanged 3-WayCincinnati, Ohio 45246 FAX (513) 247-5462********************NOTE: Heater and thermostat standard (2) auxiliary switches standardOther options available - call for detailsActuators are sized based on clean/clear fluid.SERIES 31 3-Way 150# Flanged Direct MountClass 150 Electric Assembly Actuators are sized for clean liquid surface with a specific gravity of 1 and at 70° F.For operating temperatures in excess of 158° F with an electric actuator, an extended bracket is required. Please consult factory for sizing information.SAMPLE PART #31C-FX-150/WEC1-_ _ _(2) Valve Series (4) End Connection (6) Valve Size(5) Seat, Lining, & TrimMaterial(7) Actuator(8) Accessories/Options(3) Body/Ball/StemMaterial(9) Accessories(10) Special Designation Refer to Series 31 IOM for all repair kit, seat and gasket part numbers.See the last page of catalog for How To Order detail and options.See Automated Part Number Matrix for complete part number and options.31Series-2R-20230502Copyright 2013 A-T Controls, Inc.17-4 PH ® is a registered trademark of AK Steel Corporation.Chemraz® is a registered trademark of Greene, Tweed & Co.Markez® is a registered trademark of Marco Rubber & Plastic Products Inc.Perlast® is a registered trademark of Precision Polymer Engineering Limited.TFM TM is a trademark of Dyneon TM, a 3M Company.HOW TO ORDER: Manual ValvesSAMPLE PART #31C-F1-0200-XXX-_ _ _(2) Valve Series (4) End Connection(5) Valve Size(6) Seat, Lining, & TrimMaterial(7) Special Designation (8) Additional Specials(9) Special Designation(3) Body/Ball/StemMaterial(10) O-RingDesignation (11) AdditionalSpecialsRefer to Series 31 IOM for all repair kit, seat and gasket part numbers.31Series-2R-20230502Copyright 2013 A-T Controls, Inc.SAMPLE PART #31C-FX-200/2R3D-_ _ _(2) Valve Series (4) End Connection (6) Valve Size(5) Seat, Lining, & TrimMaterial(7) Actuator(8) Accessories/Options(3) Body/Ball/StemMaterial(9) Accessories(10) Special Designation Refer to Series 31 IOM for all repair kit, seat and gasket part numbers.HOW TO ORDER: Automated ValvesCincinnati, Ohio FAX (513) 247-5462********************31Series-2R-20230502Copyright 2013 A-T Controls, Inc.。

遗传算法旅行商问题c语言代码

遗传算法是一种模拟自然选择过程的优化算法，可以用于解决各种复杂的组合优化问题。

其中，旅行商问题是一个经典的组合优化问题，也是一个典型的NP难题，即寻找最优解的时间复杂度是指数级的。

在本文中，我们将讨论如何使用遗传算法来解决旅行商问题，并给出相应的C语言代码实现。

我们将介绍旅行商问题的数学模型，然后简要介绍遗传算法的原理，最后给出C语言代码实现。

旅行商问题是指一个旅行商要拜访n个城市，恰好拜访每个城市一次，并返回出发城市，要求总路程最短。

数学上可以用一个n*n的距离矩阵d[i][j]表示城市i到城市j的距离，问题可以形式化为求解一个排列p={p1,p2,...,pn}，使得目标函数f(p)=Σd[p[i]][p[i+1]]+d[p[n]][p[1]]最小。

这个问题是一个组合优化问题，其搜索空间是一个n维的离散空间。

遗传算法是一种基于生物进化过程的优化算法，主要包括选择、交叉、变异等操作。

在使用遗传算法解决旅行商问题时，可以将每个排列p看作一个个体，目标函数f(p)看作个体的适应度，通过选择、交叉和变异等操作来搜索最优解。

以下是遗传算法解决旅行商问题的C语言代码实现：1. 我们需要定义城市的距离矩阵和其他相关参数，例如城市的数量n，种裙大小pop_size，交叉概率pc，变异概率pm等。

2. 我们初始化种裙，即随机生成pop_size个排列作为初始种裙。

3. 我们进入遗传算法的迭代过程。

在每一代中，我们首先计算种裙中每个个体的适应度，然后通过选择、交叉和变异操作来更新种裙。

4. 选择操作可以采用轮盘赌选择法，即根据个体的适应度来进行选择，适应度越高的个体被选中的概率越大。

5. 交叉操作可以采用部分映射交叉方法，即随机选择两个个体，然后随机选择一个交叉点，将交叉点之后的基因片段进行交换。

6. 变异操作可以采用变异率为pm的单点变异方法，即随机选择一个个体和一个位置，将该位置的基因值进行随机变异。

7. 我们重复进行迭代操作，直到达到停止条件（例如达到最大迭代次数或者适应度达到阈值）。

八数码问题启发函数和代价函数

八数码问题启发函数和代价函数八数码问题作为经典的搜索问题，其解决过程中启发函数和代价函数的选择对搜索效率有着重要的影响。

本文将针对八数码问题中启发函数和代价函数的选择进行探讨，并分析它们在搜索过程中的作用和影响。

一、启发函数的选择启发函数是在搜索过程中用来评估节点的“接近程度”的函数，它可以指导搜索算法朝着离目标更近的方向前进，从而提高搜索效率。

在八数码问题中，常用的启发函数有误放置数目、曼哈顿距离和线性冲突等。

1. 误放置数目误放置数目是指当前状态与目标状态中不同数字的个数，它可以作为启发函数来评估当前状态与目标状态的“距离”。

当误放置数目越小，说明当前状态距离目标状态越近，因此误放置数目可以作为一种简单而有效的启发函数。

2. 曼哈顿距离曼哈顿距离是指当前状态的每个数字到目标状态的正确位置之间的曼哈顿距离之和。

曼哈顿距离可以更准确地评估当前状态与目标状态的“距离”，因此在某些情况下，比误放置数目更适合作为启发函数。

3. 线性冲突线性冲突是指在某一行或某一列中有两个数字的目的位置相互交叉，这种情况下移动其中一个数字就会导致另一个数字也需要移动。

线性冲突可以影响搜索的效率，因此考虑线性冲突可以使启发函数更精确地评估当前状态与目标状态的“距离”。

二、代价函数的选择代价函数是指在搜索过程中用来评估节点的“代价”的函数，它可以指导搜索算法在选择候选节点时进行排序，从而提高搜索效率。

在八数码问题中，常用的代价函数有实际代价和估计代价等。

1. 实际代价实际代价是指从初始状态到当前状态的实际代价，它可以作为代价函数来评估当前状态的“代价”。

通过记录从初始状态到当前状态的实际代价，搜索算法可以更准确地评估每个候选节点的“代价”，从而更有针对性地选择下一个节点。

2. 估计代价估计代价是指从当前状态到目标状态的估计代价，它可以作为代价函数来评估当前状态的“代价”。

估计代价通常是通过启发函数来估计的，因此选择合适的启发函数对于估计代价的准确性非常重要。

驱动调试笔记

sensorlist.cpp
MTKCameraHardware.cpp
***********************************************************
CUSTOM_HAL_CAMERA = camera
CUSTOM_HAL_IMGSENSOR = gc2015_yuv
G-sensor
-------------------------------------------------------------------------------------------------------------
int hwmsen_gsensor_add(struct sensor_init_info* obj)
-------------------------------------------------------------------------------------------------------------
TouchPanel
-------------------------------------------------------------------------------------------------------------
board_init
-------------------------------------------------------------------------------------------------------------
\mediatek\platform\mt6573\kernel\core\mt6573_devs.c

Perseus randomized point-based

UniversiteitvanAmsterdamIAS technical report IAS-UVA-04-02Perseus:randomized point-basedvalue iteration for POMDPsMatthijs T.J.Spaan and Nikos VlassisInformatics InstituteFaculty of ScienceUniversity of AmsterdamThe NetherlandsPartially observable Markov decision processes(POMDPs)form an attractive andprincipled framework for agent planning under uncertainty.Point-based approx-imate techniques for POMDPs compute a policy based on aﬁnite set of pointscollected in advance from the agent’s belief space.We present a randomizedpoint-based value iteration algorithm called Perseus.The algorithm performsapproximate value backup stages,ensuring that in each backup stage the valueof all points in the belief set is improved(or at least does not decrease).Con-trary to other point-based methods,Perseus backs up only a(random)subsetof belief points—the key observation is that a single backup may improve thevalue of many points in the set.We show how the same idea can be extended todealing with continuous action spaces.Experimental results show the potentialof Perseus in large scale POMDP problems.IASintelligent autonomous systemsPerseus:randomized point-based value iteration for POMDPs ContentsContents1Introduction1 2Partially observable Markov decision processes22.1Exact value iteration (4)2.2Approximate value iteration (6)3Randomized point-based backup stages63.1Perseus (7)3.2Discussion (8)3.3Extension to planning with continuous actions (9)4Related work10 5Experiments115.1Discrete action spaces (11)5.1.1Benchmark mazes (11)5.1.2Tag (12)5.2Continuous action spaces (13)5.2.1Active localization (14)5.2.2Arbitrary heading navigation (15)6Conclusions16 Intelligent Autonomous SystemsInformatics Institute,Faculty of ScienceUniversity of AmsterdamKruislaan403,1098SJ AmsterdamThe NetherlandsTel(fax):+31205257461(7490) http://www.science.uva.nl/research/ias/Corresponding author:Matthijs T.J.Spaantel:+31205257524mtjspaan@science.uva.nlhttp://www.science.uva.nl/~mtjspaan/Copyright IAS,2004Section1Introduction11IntroductionA major goal of Artiﬁcial Intelligence(AI)is to build intelligent agents[29].The agent,whether physical or simulated,should be able to autonomously perform a given task.An intelligent agent is often characterized by its sense–think–act loop:it uses sensors to observe the environment, considers this information to decide what to do,and executes the chosen action.The agent inﬂuences its environment by acting and can detect the eﬀect of its actions by sensing:the environment closes the loop.In this work we are interested in computing a plan that maps sensory input to the optimal action to execute for a given task.We consider types of domains in which an agent is uncertain about the exact eﬀect of its actions.Furthermore,it cannot determine with full certainty the state of the environment with a single sensor reading,i.e.,the environment is only partially observable to the agent.Planning under these kinds of uncertainty is a challenging problem as it requires reason-ing over all possible futures given all possible histories.Partially observable Markov decision processes(POMDPs)provide a rich mathematical framework for acting optimally in such par-tially observable and stochastic environments[3,32,15,13].The POMDP deﬁnes a sensor model specifying the probability of observing a particular sensor reading in a speciﬁc state and a stochastic transition model which captures the uncertain outcome of executing an action.The agent’s task is deﬁned by the reward it receives at each time step and its goal is to maximize the discounted cumulative reward.Assuming discrete models,the POMDP framework allows for capturing all uncertainty introduced by the transition and observation model by deﬁning and operating on the belief state of an agent.A belief state is a probability distribution over all states and summarizes all information regarding the past.Using belief states allows one to transform the original discrete state POMDP into a contin-uous state Markov decision process(MDP),which can in turn be solved by corresponding MDP techniques[6].However,the optimal value function in a POMDP exhibits particular structure (it is piecewise linear and convex)that one can exploit in order to facilitate the solving.Value iteration,for instance,is a method for solving POMDPs that builds a sequence of value function estimates which converge to the optimal value function for the current task[32].The value function is parametrized by aﬁnite number of hyperplanes,or vectors,over the belief space,and which partition the belief space in aﬁnite amount of regions.Each vector maximizes the value function in a certain region,and with each vector an action is associated which is the optimal action to take for beliefs in its puting the next value function estimate—looking one step deeper into the future—requires taking into account all possible actions the agent can take and all subsequent observations it may receive.Unfortunately,this leads to an exponential growth of vectors with the planning horizon.Many of the computed vectors will be useless in the sense that their maximizing region is empty,but identifying and subsequently pruning them is an expensive operation.Exact value iteration algorithms[32,11,13]search in each value iteration step the complete belief simplex for a minimal set of belief points that generate the necessary set of vectors for the next horizon value function.This typically requires solving a number of linear programs and is therefore costly in high dimensions.In[38]it was argued that value iteration still converges to the optimal value function if exact value iteration steps are interleaved with approximate value iteration steps in which the new value function is an upper bound to the previously computed value function.This results in a speedup of the total algorithm,however,linear programming is again needed in order to ensure that the new value function is an upper bound to the previous one over the complete belief simplex.In general,computing exact solutions for POMDPs is an intractable problem[20,16],calling for approximate solution techniques[15,12].In practical tasks one would like to compute solutions only for those parts of the belief simplex that are reachable,i.e.,that can be actually encountered by interacting with the environment.2Perseus:randomized point-based value iteration for POMDPs This has recently motivated the use of approximate solution techniques for POMDPs which focus on the use of a sampled set of belief points on which planning is performed[12,22,27,21,37,33], a possibility already mentioned in[15].The idea is that instead of planning over the complete belief space of the agent(which is intractable for large state spaces),planning is carried out only on a limited set of prototype beliefs that have been sampled by letting the agent interact (randomly)with the environment.PBVI[21],for instance,builds successive estimates of the value function by updating the value and its gradient only at the points of a(dynamically growing)belief set.In this work we describe Perseus,a randomized point-based value iteration algorithm for POMDPs[37,33].Perseus operates on a large set of beliefs which are gathered by simulating random interactions of the agent with the POMDP environment.On this belief set a number of backup stages are performed.The algorithm ensures that in each backup stage the value of all points in the belief set is improved(or at least does not decrease).Contrary to other point-based methods,Perseus backs up only a(random)subset of belief points—the key observation is that a single backup may improve the value of many points in the set.This allows us to compute value functions that consist of only a small number of vectors(relative to the belief set size),leading to signiﬁcant speedups.We evaluate the performance of Perseus on benchmark problems from literature,and it turns out to be very competitive to state-of-the-art methods in terms of solution quality and computation time.We extend Perseus to compute plans for agents which have a continuous set of actions at their disposal.Examples include navigating to an arbitrary location,or rotating a pan-and-tilt camera at any desired angle.Almost all work on POMDP solution techniques targets discrete action spaces;an exception is the application of a particleﬁlter to a continuous state and action space[35].We report on experiments in an abstract active localization domain in which an agent can control its range sensors to inﬂuence its localization estimate,and on results from a navigation task involving a mobile robot with omnidirectional vision in a perceptually aliased oﬃce environment.The rest of the paper is structured as follows:in Section2we review the POMDP framework from an AI perspective.We discuss exact methods for solving POMDPs and their tractability problems.Next we outline a class of approximate value iteration algorithms,the so called point-based techniques.In Section3we describe and discuss the Perseus algorithm,as well as the extension to continuous action spaces.Related work on approximate techniques for POMDP planning is discussed in Section4.We present experimental results from several problem domains in Section5.Finally,Section6wraps up with some conclusions.2Partially observable Markov decision processesA partially observable Markov decision process(POMDP)models the repeated interaction of an agent with a stochastic environment,parts of which are hidden from the agent’s view.The agent’s goal is to perform a task by choosing actions which fulﬁll the task best.Put otherwise, the agent has to compute a plan that optimizes the given performance measure.We assume that time is discretized in time steps of equal length,and at the start of each step the agent has to execute an action.At each time step the agent also receives a scalar reward from the environment,and the performance measure directs the agent to maximize the cumulative reward it can gather.The reward signal allows one to deﬁne a task for the agent,e.g.,one can give the agent a large positive reward when it accomplishes a certain goal and a small negative reward for each action leading up to it.In this way the agent is steered towardﬁnding the plan which will let it accomplish its goal as fast as possible.The POMDP framework models stochastic environments in which an agent is uncertain about the exact eﬀect of executing a certain action.This uncertainty is captured by a proba-Section2Partially observable Markov decision processes3 bilistic transition model as is the case in a fully observable Markov decision process(MDP)[34,6]. An MDP deﬁnes a transition model which speciﬁes the probabilistic eﬀect of how each action changes the state.Extending the MDP setting,a POMDP also deals with uncertainty resulting from the agent’s imperfect sensors.It allows for planning in environments which are only par-tially observable to the agent,i.e.,environments in which the agent cannot determine with full certainty the true state of the environment.In general the partial observability stems from two sources:(1)multiple states lead to the same sensor reading,in case the agent can only sense a limited part of the environment,and(2)its sensor readings are noisy:observing the same state can result in diﬀerent sensor readings.The partially observability can lead to“perceptual aliasing”:the problem that diﬀerent parts of the environment appear similar to the agent’s sensor system,but require diﬀerent actions.The POMDP represents the partial observability by a probabilistic observation model,which relates possible observations to states.More formally,a POMDP assumes that at any time step the environment is in a state s∈S,the agent takes an action a∈A and receives a reward r(s,a)from the environment as a result of this action,while the environment switches to a new state s according to a known stochastic transition model p(s |s,a).The Markov property entails that s only depends on the previous state s and the action a.The agent then perceives an observation o∈O,that may be conditional on its action,which provides information about the state s through a known stochastic observation model p(o|s,a).All sets S,O,and A are assumed discrete andﬁnite here (but we will generalize to continuous A in Section3.3).In order for an agent to choose its actions successfully in partially observable environments some form of memory is needed,as the observations the agent receives do not provide an unique identiﬁcation of s.Given the transition and observation model the POMDP can be transformed to a belief-state MDP:the agent summarizes all information about its past using a belief vector b(s).The belief b is a probability distribution over S,which forms a Markovian signal for the planning task[3].All beliefs are contained in the simplex∆,which means we can represent a belief using|S|−1numbers.Each POMDP problem assumes an initial belief b0,which for instance can be set to a uniform distribution over all states(representing complete ignorance regarding the initial state of the environment).Every time the agent takes an action a and observes o,its belief is updated by Bayes’rule:b o a(s )=p(o|s ,a)p(o|a,b)s∈Sp(s |s,a)b(s),(1)where p(o|a,b)= s ∈S p(o|s ,a) s∈S p(s |s,a)b(s)is a normalizing constant.As we discussed above,the goal of the agent is to choose actions which fulﬁll its task as good as possible,i.e.,to compute an optimal plan.Such a plan is called a policyπ(b)and maps beliefs to actions.Note that,contrary to MDPs,the policyπ(b)is a function over a continuous set of probability distributions over S.The quality of a policy is rated by a performance measure, i.e.,by an optimality criterion.A common criterion is the expected discounted future reward E[ ∞t=0γt r(s t,π(b t))],whereγis a discount rate,0≤γ<1.The discount rate ensures aﬁnite sum and is usually chosen close to1.A policy which maximizes the optimality criterion is called an optimal policyπ∗;it speciﬁes for each b the optimal action to execute at the current step, assuming the agent will also act optimal at future time steps.A policy can be deﬁned by a value function V n which determines the expected amount of future discounted reward V n(b)the agent can gather in n steps from every belief b.The value function of an optimal policy is characterized by the optimal value function V∗which satisﬁes the Bellman optimality equation V∗=HV∗,orV∗(b)=maxa∈As∈Sr(s,a)b(s)+γ o∈O p(o|a,b)V∗(b o a) ,(2)4Perseus:randomized point-based value iteration for POMDPs with b o a given by(1),and H is the Bellman backup operator[4].When(2)holds for every b∈∆we are ensured the solution is optimal.V∗can be approximated by iterating a number of stages,as we will see in the next section,at each stage considering a step further into the future.For problems with aﬁnite planning horizon V∗will be piecewise linear and convex(PWLC)[30],and for inﬁnite horizon(non-episodic)tasks V∗can be approximated arbitrary well by a PWLC value function.We parametrize such a value function V n by aﬁnite set of vectors(hyperplanes){αi n},i=1,...,|V n|.Additionally,with each vector an action a(αi n)∈A is associated,which is the optimal one to take in the current step. Each vector deﬁnes a region in the belief space for which it is the maximizing element of V n. These regions form a partition of the belief space,induced by the piecewise linearity of the value function.Examples of a value function for a two state POMDP are shown in Fig.1(a)and1(d). Given a set of vectors{αi n}|V n|i=1in V n,the value of a belief b is given byV n(b)=max{αi n}ib·αi n,(3) where(·)denotes inner product.The gradient of the value function at b is given by the vectorαb n=arg max{αin }ib·αi n,and the policy at b is given byπ(b)=a(αb n).2.1Exact value iterationComputing an optimal plan for an agent means solving the POMDP,and a classical method for POMDP solving is value iteration.In the POMDP framework,value iteration involves approximating V∗by applying the exact dynamic programming operator H above,or some approximate operator˜H,to an initially piecewise linear and convex value function V0.For H,and for many commonly used˜H,the produced intermediate estimates V1,V2,...will also be piecewise linear and convex.The main idea behind many value iteration algorithms for POMDPs is that for a given value function V n and a particular belief point b we can easily compute the vectorαb n+1of HV n such thatαb n+1=arg max{αi n+1}ib·αi n+1(4)where{αi n+1}|HV n|i=1is the(unknown)set of vectors for HV n.We will denote this operationαb n+1=backup(b).It computes the optimal vector for a given belief b by back-projecting all vectors in the current horizon value function one step from the future and returning the vector that maximizes the value of b.In particular,deﬁning r a(s)=r(s,a)and using(1),(2),and(3) we have:V n+1(b)=maxa b·r a+γop(o|a,b)V n(b o a) (5)=maxa b·r a+γop(o|a,b)max{αi n}isb o a(s )αi n(s ) (6)=maxa b·r a+γomax{αi n}isp(o|s ,a) s p(s |s,a)b(s)αi n(s ) (7)=maxa b·r a+γomax{g i a,o}ib·g i a,o ,(8)whereg i a,o(s)= s p(o|s ,a)p(s |s,a)αi n(s ).(9)Section2Partially observable Markov decision processes5 Applying the identity max j b·αj=b·arg max j b·αj in(8)twice,we can compute the vector backup(b)as follows:backup(b)=arg max{g b a}a∈Ab·g b a,where(10)g b a=r a+γ o arg max{g i a,o}ib·g i a,o.(11)Although computing the vector backup(b)for a given b is straightforward,locating the (minimal)set of points b required to compute all vectors∪b backup(b)of HV n is very costly.As each b has a region in the belief space in which itsαb n is maximal,a family of algorithms tries to identify these regions[32,11,13].The corresponding b of each region is called a“witness”point,as it testiﬁes to the existence of its region.Another set of exact POMDP value iteration algorithms do not focus on searching in the belief space,but instead consider enumerating all possible vectors of HV n and then pruning useless vectors[18,10].As an example of exact value iteration let us consider the most straightforward way of computing HV n due to Monahan[18].This involves calculating all possible ways HV n could be constructed,exploiting the known structure of the value function.We operate independent of a particular b now so(11)can no longer be applied.Instead we have to include all ways of selecting g i a,o for all o:HV n= a,i{g i a},with{g i a}= o r a+γg i a,o ,(12)where denotes the cross-sum operator.1Unfortunately,at each stage a number of vectors exponential in|O|are generated:|A||V n||O|.The regions of many of the generated vectors will be empty and these vectors as such are useless,but identifying and subsequently pruning them requires linear programming and is therefore costly in high dimensions.In[38]an alternative approach to exact value iteration was proposed,designed to speed up each exact value iteration step.It turns out that value iteration still converges if exact value update steps are interleaved with approximate update steps in which a new value function V n+1 is computed from V n such thatV n(b)≤V n+1(b)≤HV n(b),for all b∈∆.(13)This additionally requires that the value function is appropriately initialized,which is triviallyrealized by choosing V0to be a single vector with all its components equal to11−γmin s,a r(s,a).Such a vector represents the minimum of cumulative discounted reward obtainable in the POMDP,and is guaranteed to be below V∗.In[38],V n+1is computed by backing up all witness points of V n for a number of steps.As we saw above,backing up a set of belief points is a relatively cheap operation.Thus,given V n,a number of vectors of HV n are created by applying backup to the witness points of V n,and then a set of linear programs are solved to ensure that V n+1(b)≥V n(b),∀b∈∆.This is repeated for a number of steps,before an exact value update step takes place.The authors demonstrate experimentally that a combination of approximate and exact backup steps can speed up exact value iteration.In general,however,computing optimal planning solutions for POMDPs is an intractable problem for any reasonably sized task[20,16].This calls for approximate solution techniques. We will describe next a recent line of research on approximate POMDP algorithms which focus on planning on aﬁxed set of belief points.1Cross-sum of sets{R i}is deﬁned as: k i=1R i=R1⊕R2⊕...⊕R k,with P⊕Q={p+q|p∈P,q∈Q}.6Perseus:randomized point-based value iteration for POMDPs2.2Approximate value iterationThe major cause of intractability of exact POMDP solution methods is their aim of computing the optimal action for every possible belief point in∆.For instance,if we use(12)we end up with a series of value functions whose size grows exponentially in the planning horizon.A natural way to sidestep this intractability is to settle for computing an approximate solution by considering only aﬁnite set of belief points.The backup stage reduces to applying(10)aﬁxed number of times,resulting in a small number of vectors(bounded by the size of the belief set). The motivation for using approximate methods is their ability to compute successful policies for much larger problems,which compensates for the loss of optimality.Such approximate POMDP value iteration methods operating on aﬁxed set of points are explored in[15]and in subsequent works[12,22,21,37,33].In[21]for instance,an approximate backup operator˜H PBVI is used instead of H,that computes in each value backup stage the set˜HPBVI V n= b∈B backup(b)(14)using aﬁxed set of belief points B.The general assumption underlying these so-called point-based methods is that by updating not only the value but also its gradient(theαvector)at each b∈B,the resulting policy will generalize well and be eﬀective for most beliefs encountered by the agent.Whether or not this assumption is realistic depends on the POMDP’s structure and the contents of B,but the intuition is that in many problems the set of‘reachable’beliefs forms a low dimensional manifold in the belief simplex,and thus it can be covered densely enough by a relatively small number of belief points.Crucial to the control quality of the computed approximate solution is the makeup of B.A number of schemes to build B have been proposed.For instance,one could use a regular grid on the belief simplex,computed,e.g.,by Freudenthal triangulation[15].Other options include taking all extreme points of the belief simplex or use a random grid[12,22].An alternative scheme is to include belief points that can be encountered by simulating the POMDP:we can generate trajectories through the belief space by sampling random actions and observations at each time step[15,12,22,21,37,33].This sampling scheme focuses the contents of B to be beliefs that can actually be encountered while experiencing the POMDP model.The PBVI algorithm[21]is an instance of such a point-based POMDP algorithm.PBVI starts by selecting a small set of beliefs B0,performs a number of backup stages(14)on B0, expands B0to B1by sampling more beliefs,performs again a series of backups,and repeats this process until a satisfactory solution has been found(or the allowed computation time expires). The set B t+1grows by simulating actions for every b∈B t,maintaining only the new belief points that are furthest away from all other points already in B t+1.This scheme is a heuristic to let B t cover a wide area of the belief space,but comes at a cost as it requires computing distances between all b∈B t.By backing up all b∈B t the PBVI algorithm generates at each stage approximately|B t|vectors,which can lead to performance problems in domains requiring large B t.In the next section we will present a point-based POMDP value iteration method which does not require backing up all b∈B.We compute backups for a subset of B only,but seeing to it that the computed solution will be eﬀective for B.As a result we limit the growth of the number of vectors in the successive value function estimates,leading to signiﬁcant speedups.3Randomized point-based backup stagesWe have introduced the POMDP framework which models agents inhabiting stochastic envi-ronments that are partially observable to them,and discussed exact and approximate methodsSection3Randomized point-based backup stages7 for computing successful plans for such agents.Below we describe Perseus,an approximate solution method capable of computing competitive solutions in large POMDP domains.3.1PerseusPerseus is an approximate point-based value iteration algorithm for POMDPs[37,33].The value update scheme of Perseus implements a randomized approximate backup operator˜HPerseus that improves(instead of maximizes)the value of all belief points in B.Such an operator can be very eﬃciently implemented in POMDPs given the shape of the value function. The key idea is that in each value backup stage we can improve the value of all points in the belief set by only updating the value and its gradient of a randomly selected subset of the points. In each backup stage,given a value function V n,we compute a value function V n+1that improves the value of all b∈B,i.e.,we build a value function V n+1=˜H Perseus V n that upper bounds V n over B(but not necessarily over∆which would require linear programming):V n(b)≤V n+1(b),for all b∈B.(15) Weﬁrst let the agent randomly explore the environment and collect a set B of reachable belief points.We initialize the value function V0as a single vector with all its components equalto11−γmin s,a r(s,a)as in[38].Starting with V0,Perseus performs a number of value functionupdate stages until some convergence criterion is met.Each backup stage is deﬁned as follows, where˜B is the set of non-improved points:Perseus randomized backup stage:V n+1=˜H Perseus V n1.Set V n+1=∅.Initialize˜B to B.2.Sample a belief point b uniformly at random from˜B and computeα=backup(b).3.If b·α≥V n(b)then addαto V n+1,otherwise addα =arg maxαi∈Vnb·αi to V n+1.pute˜B={b∈B:V n+1(b)<V n(b)}.If˜B=∅then stop,else go to2.Often,a small number of vectors will be suﬃcient to improve V n(b)∀b∈B,especially in theﬁrst steps of value iteration.The idea is to compute these vectors in a randomized greedy manner by sampling from˜B,an increasingly smaller subset of B.We keep track of the set of non-improved points˜B consisting of those b∈B whose new value V n+1(b)is still lower than V n(b).At the start of each backup stage,V n+1is set to∅which means˜B is initialized to B, indicating that all b∈B still need to be improved in this backup stage.As long as˜B is not empty,we sample a point b from˜B and computeα=backup(b).Ifαimproves the value of b (i.e.,if b·α≥V n(b)in step3),we addαto V n+1and update V n+1(b)for all b∈B by computing their inner product with the newα.The hope is thatαimproves the value of many other points in B,and all these points are removed from˜B.As long as˜B is not empty we continue sampling belief points from it and try to add theirαvectors.To ensure termination of each backup stage we have to enforce that˜B shrinks when adding vectors,i.e.,that eachαactually improves at least the value of the b that generated it.If not (i.e.,b·α<V n(b)in step3),we ignoreαand insert a copy of the maximizing vector of b from V n in V n+1.Point b is now considered improved and is removed from˜B in step4,together with any other belief points which had the same vector as maximizing one in V n.This procedure ensures that˜B shrinks and the backup stage will terminate.A pictorial example of a backup stage is presented in Fig.1.。

《高铁票价定价模型分析国内外文献综述3700字》

高铁票价定价模型研究国内外文献综述1国内研究现状国内有些学者聚焦对不同交通方式之间票价的影响因素开展研究。

刘莉文&张明[13]在梳理高速铁路和高速公路在各自因素条件下的经济运输距离，在此基础上制定不同经济运输距离条件下的运输资源优化策略；陶莉[14]比较分析交通运输行业不同运输方式的优劣势，并以京沪高速铁路为案例对象，结合高铁价格比较模型，得出了短途、中长途、长途等不同铁路运输方式之间的价格比较关系及相应的优势领域，指出高铁票价直接影响高速铁路作用的发挥和使命的实现。

王欢[15]在进行问卷调查的基础上，详细研究了不同收入群体在铁路交通运输客流高峰时期的弹性需求规律，进而制定了差异化的定价策略，并针对中长途客运范围内民航对高铁的影响制定合理的票价。

李旭峰，等[16]在统一计量企业以及社会属性等影响因素的条件下，制定了客运专线的客票定价体系，有助于缓解铁路客运压力。

张一腾、王小平[17]通过分析线路同一OD间的各次列车上座率，根据列车之间的相互替代性并结合交通出行乘客对于时间、价格的需求特点，在列车整体期望收益最大化为目标的约束条件下，建立了各次列车综合收益最大化的动态定价模型，从而最大限度地吸引客流，增加运输密度。

在铁路票价定价模型方面，邢泽邦，等[18]以京津城际铁路为案例对象，构建普速铁路，城际铁路以及高速铁路等运输方式的广义成本模型，并基于2012-2020年的数据对京津城际铁路各种运输方式的分担率和未来趋势进行计算和预测。

张睿, 马瑜, 赵冰茹,等[19]通过SP调查问卷的形式，详细梳理了交通出行乘客对高铁、民航的不同需求，利用Logit模型分析了高铁、民航两种交通出行方式在票价、发车频率、发车时刻等因子的变化规律，明确了高铁、民航两种交通出行方式分时段发车频率的确定方法，从而促进高铁、民航运能资源的最优配置，提高综合交通运输体系的资源利用率。

宋丹丹[20]利用系统动力学方法高铁票价的影响因素以及定价机制开展了详细研究。

车辆路径问题的一种先寻路后分组算法

顾客点分配到不同的路径中。首先创建一个长度为Ｌ
传算法的主要区别在变异操作上，差分进化的变异是基于染色体的差异向量进行的。由于性能卓越，近年来差
分进化算法在众多领域逐渐引起了越来越多的关注。
禁忌搜索（ｃｔｒｅｒ）Ｇｏｅ为了改进局部搜ＳａｅＳａｃ是ｌｒｔｈｖ
大多数已发表的ＶＰ研究都集中在设计高效可行的启Ｒ
发式算法上。然而这些算法的优势与缺点并存，比如一
Ｐｉｓｒ在将其应用在混合进化算法中时证明：ｎ必然存在一
条哈密尔顿回路，将其切割成ＶＰ的最优解Ｊ因可Ｒ。
解空问内向不能改进解的方向探索，从而能够跳出局部最优。同时通过禁忌表这种记忆方式阻止搜索的频繁回溯。禁忌搜索在解决ＶＰ上的出色表现使其经常成Ｒ为同类问题其他算法性能测试的参照基准。本文所提出的混合算法基于一个高效的两阶段改
进策略，的编码形式在ＴＰ解和ＶＰ解之间反复转解ＳＲ换，用Ｌｈｅ编码形式的加减操作来描述候选解之采ｅｍｒ
Ｖｏ．Ｎｏ２１２３．Ａｐ．０１ｒ２０
文章编号：６３１４（０００－２５０１７－５９２１）２０４－４
车辆路径问题的一种先寻路后分组算法
单琨，向晓林
（四川大学工商管理学院，成都６０６）１０５
些启发式算法为了得到高质量的解，通常在计算和调整参数上花费大量的时间。为了扬长避短，探索混合算法并设计不同算法间的组合框架显得尤为重要。作为一种基于群体智能的优化算法，差分进化（ｉＤｆ－

马拉车算法——精选推荐

马拉车算法Manacher Algorithm算法，俗称马拉车算法，其时间复杂为O(n)。

该算法是利⽤回⽂串的特性来避免重复计算的。

在时间复杂度为O(n^2)的算法中，我们在遍历的过程要考虑到回⽂串长度的奇偶性，⽐如说“abba”的长度为偶数，“abcba”的长度为奇数，这样在寻找最长回⽂⼦串的过程要分别考奇偶的情况，是否可以统⼀处理了？⼀）第⼀步是改造字符串S，变为T，其改造的⽅法如下：在字符串S的字符之间和S的⾸尾都插⼊⼀个“#”，如：S=“abba”变为T="#a#b#b#a#" 。

我们会发现S的长度是4，⽽T的长度为9，长度变为奇数了！！S的长度为奇数时，S=“abcba”，变化为T=“#a#b#c#b#a#”，T的长度为11，所以我们发现其改造可以将字符串的长度变为奇数，这样就可以统⼀的处理奇偶的情况了。

⼆）第⼆步，为了改进回⽂相互重叠的情况，我们将改造完后的T[ i ] 处的回⽂半径存储到数组P[ i ]中(P[ i ]表⽰以字符T[ i ]为中⼼的最长回⽂字串的最端右字符到T[ i ]的长度)，这样最后遍历数组P，取其中最⼤值即可。

举⼀个简单的例⼦感受⼀下：数组P有⼀性质，P[ i ] - 1就是该回⽂⼦串在原字符串S中的长度，⾄于证明，⾸先在转换得到的字符串T中，所有的回⽂字串的长度都为奇数，那么对于以T[ i ]为中⼼的最长回⽂字串，其长度就为2*P[ i ]-1, 经过观察可知，T中所有的回⽂⼦串，其中分隔符的数量⼀定⽐其他字符的数量多1，也就是有P[ i ]个分隔符，剩下P[ i ]-1个字符来⾃原字符串，所以该回⽂串在原字符串中的长度就为P[ i ]-1。

另外，由于第⼀个和最后⼀个字符都是#号，且也需要搜索回⽂，为了防⽌越界，我们还需要在⾸尾再加上⾮#号字符，实际操作时我们只需给开头加上个⾮#号字符，结尾不⽤加的原因是字符串的结尾标识为'\0'，等于默认加过了。

Diffusive Logistic Model Towards Predicting

arXiv:1108.0442v1 [ce social networks have recently become an effective and innovative channel for spreading information and inﬂuence among hundreds of millions of end users. Many prior work have carried out empirical studies and proposed diffusion models to understand the information diffusion process in online social networks. However, most of these studies focus on the information diffusion in temporal dimension, that is, how the information propagates over time. Little attempt has been given on understanding information diffusion over both temporal and spatial dimensions. In this paper, we propose a Partial Differential Equation (PDE), speciﬁcally, a Diffusive Logistic (DL) equation to model the temporal and spatial characteristics of information diffusion in online social networks. To be more speciﬁc, we develop a PDE-based theoretical framework to measure and predict the density of inﬂuenced users at a given distance from the original information source after a time period. The density of inﬂuenced users over time and distance provides valuable insight on the actual information diffusion process. We present the temporal and spatial patterns in a real dataset collected from Digg social news site, and validate the proposed DL equation in terms of predicting the information diffusion process. Our experiment results show that the DL model is indeed able to characterize and predict the process of information propagation in online social networks. For example, for the most popular news with 24,099 votes in Digg, the average prediction accuracy of DL model over all distances during the ﬁrst 6 hours is 92.08%. To the best of our knowledge, this paper is the ﬁrst attempt to use PDE-based model to study the information diffusion process in both temporal and spatial dimensions in online social networks.

《车辆路径问题》课件

满载率和里程利用率是衡量运输效率的重要指标。通过提高满载率和里程利用率，可以降低单位里程的成本，实现成本优化。
组合运输是指将多个需求点或货物组合在一起进行运输，以提高满载率和里程利用率。组合运输可以降低单位里程的成本，实现成本优化。
不同的运输方式和运输路线会有不同的成本。在成本优化中需要考虑选择合适的运输方式和路线，以降低总成本。
背景
随着物流配送行业的快速发展，VRP已成为提高物流效率、降低运输成本的关键问题。
问题的起源和重要性
起源
VRP最早由Dantzig和Ramser于 1959年提出，是运输问题的一个变种。
重要性
VRP在实际生活中广泛应用于快递配送、货物运输、公共交通路线规划等领域，对于提高物流效率和客户满意度具有重要意义。
05
车辆路径问题的实际应用案例
物流配送
物流配送是车辆路径问题最常见的应用场景之一。
例如，在电商物流中，车辆路径的优化可以减少配送时间，提高客户满意度。
通过优化车辆路径，降低运输成本，提高配送效率，满足客户对时效性的要求。
公共交通规划
公共交通规划中，车辆路径问题用于优化公交线路、出租，快速找到问题的近似最优解。
近似算法
设计具有多项式时间复杂度的近似算法，在可接受的时间内获得
近似最优解。
THANKS FOR WATCHING
感谢您的观看
《车辆路径问题》ppt 课件
目录
• 车辆路径问题简介 • 车辆路径问题的基本模型 • 车辆路径问题的求解方法 • 车辆路径问题的优化策略 • 车辆路径问题的实际应用案例 • 未来研究方向和展望
01
车辆路径问题简介

Franchise partner slection

Franchise partner selection:perspectives of franchisors and franchiseesMaureen Brookes and Levent AltinayDepartment of Hospitality,Leisure and Tourism Management,Oxford Brookes University,Oxford,UKAbstractPurpose–This paper aims to identify the partner selection criteria employed both by franchisors and franchisees in master franchise agreements and evaluate how different selection criteria interact within the selection process and inﬂuence the decisions taken.Design/methodology/approach–A single embedded case study of an international hotelﬁrm was the focus of the enquiry.Interviews and document analysis were used as the data collection techniques.Findings–Theﬁndings reveal that the establishment of franchise partnership involves a mutual and careful evaluation between franchisors and franchisees to assess whether their partnership criteria are compatible.The partner selection approach determines the extent of importance attached to different task-and partner-related selection criteria.In addition,the study identiﬁes the role that different selection criteria play at different stages of the process.Research limitations/implications–Theﬁndings are based on a single case study in the international hotel industry and therefore may not be generalisable to otherﬁrms or industry sectors.Moreover,the study comprised master franchise agreements,and this contextual variable may impact on theﬁndings determined.Practical implications–This paper illuminates the challenges both international franchisors and franchisees face in selecting their partners and proposes that both franchisors and franchisees should employ clearly deﬁned selection criteria,utilise a deﬁned selection process and choose their selection approach carefully in recruiting partners.Originality/value–This paper cross-fertilises the strategic alliance and franchise literature to evaluate the interplay of partner selection criteria, process,selection approach and international franchise recruitment.Theﬁndings contribute to the understanding of a largely neglected area,franchise partner selection and recruitment,by taking a holistic approach and incorporating the views of both franchisors and franchisees.Keywords Master franchising,Franchisor,Franchisee,Task and partner selection criteria,Process,Selection approach,Franchising,Selection Paper type Research paperAn executive summary for managers and executive readers can be found at the end of this article.1.IntroductionThe high cost and risk of developing new products and penetrating new markets is a major force that drives organisations to work in partnership(Hitt et al.,2000). While there has been an increase in a wide variety of inter-ﬁrm alliances,business format franchising has emerged as a powerful form of collaboration,expanding faster and more vigorously than other forms in international service industries (Doherty and Alexander,2004;Alon,2006)and in the hospitality industry in particular(Altinay and Wang,2006). Recent research conducted by Franchise Direct(2009)[1], identiﬁes that77per cent of the top100global franchises (Franchise Direct,2009)are within service industries, including hospitality,cleaning and maintenance,professional and retail services.Within hospitality,quick service restaurants represent20per cent,and hotels a further9per cent of the total.For both these industry sectors,master franchise agreements in particular have become a popular international market entry mode.In any type of business format franchising,a franchisor grants a franchisee the rights to use its brand name,product and business system in a speciﬁed manner for a speciﬁc period of time(Felstead,1993,p.58).Franchisees gain access to a proven brand concept and business system and franchisors gain access to the franchisees’local market knowledge.While these complementary beneﬁts help to explain the popularity of franchising,they also underpin the main issue of contention in franchise systems,i.e.achieving an appropriate balance between franchisor control to maintain brand uniformity and integrity,and franchisee autonomy to respond to local market demands(Bradach,1997;Sorenson and Sorensen,2001;Weavin and Frazer,2007a,b).These tensions are compounded in geographically dispersed and differentiated markets(Cox and Mason,2007)and are therefore exacerbated in international franchise systems.They are also potentially more prominent in master franchise agreements.These differ from other types of business format franchising as they entitle the franchisee the rights to open franchised hotel units and to grant these rights to third parties as a sub-franchisor(Connell,1999;Quinn and Alexander,The current issue and full text archive of this journal is available at /0887-6045.htmJournal of Services Marketing25/5(2011)336–348q Emerald Group Publishing Limited[ISSN0887-6045] [DOI10.1108/08876041111149694]Received:May2009 Revised:October2009 Accepted:December20092002).As such,they are quite distinct inter-ﬁrm agreements, where a degree of operational control is devolved to the master franchisee(Ryans et al.,1999;Brookes and Roper, 2008).Partner selection is paramount to the success of master franchise agreements as failing to select the“right”partner could lead to a divergence of goals between franchisors and franchisees.That is,franchisees will behave in an opportunistic fashion and pursue their own interest at the expense of those of the franchisor(Brickley and Dark,1987; Elango and Fried,1997;T aylor,2000).As this has a cost to the brand and other franchisees,there is a need to prevent franchisees behaving opportunistically in order to maintain brand uniformity and protect a franchisor’s brand name and image(Fladmoe-Lindquist,2000).One of the most efﬁcient ways of reducing this risk is to select franchisee partners who will adopt more of a system-wide perspective for their individual activities and contribute to the attainment of system-wide goals(Altinay and Wang,2006;T aylor,2000). Despite this importance,partner selection within international franchising remains a relatively unexplored area(Doherty,2009).Research that has been undertaken does reveal the importance of both the partner selection process and partner selection criteria(see,for example,Altinay,2006; Doherty,2009).However,these studies have tended to look at partner selection from the viewpoint of the franchisor only and the value of examining the process and criteria from both franchisor and franchisee perspectives has been identiﬁed (Doherty,2009).Furthermore,the distinct inter-organisational characteristics of master franchise agreements highlight the relevance of examining partner selection from both of these perspectives,an approach more frequently adopted within joint venture studies.This paper therefore aims to identify the partner selection criteria employed both by franchisors and franchisees in international master franchise agreements.It draws on both the extant joint venture and franchise literature to examine partner selection criteria and the processes employed.After presenting the research design,the paper evaluates how different selection criteria interact within the master franchise partner selection process and inﬂuence the decisions taken.2.Partner selectionThe importance of partner selection in international joint ventures(IJVs)is well recognised(Glaister and Buckley, 1997;Grifﬁth et al.,1998;Al-Khalifa and Peterson,1999; Hitt et al.,2000)and empirical studies have revealed a range of relevant partner selection criteria.In his study of IJV partnerships of Britishﬁrms in Pakistan and India,T omlinson (1970)identiﬁed six categories of partner selection criteria of different degrees of importance.Favourable past association was found to be the most important category;resources, facilities,partner status and forced choice,the next most important category;and local identity,the least signiﬁcant category.In a later study,however,T omlinson and Thompson (1977)found a difference between the selection criteria used by Canadian and Mexican joint venture partners.While ﬁnancial resources were important to bothﬁrms,Canadian ﬁrms prioritised compatibility in business,similar objectives, ability to negotiate with government and common ethics.In contrast,technology and experience applying it,international prestige and experience,commitment,sound management and the ability to communicate with Mexicans were revealed as the key traits sought by the Mexicanﬁrms.Y et a further study by Awadzi et al.(1988)of US IJV partnership identiﬁed four key criteria as partners’resource contribution,past association between partners,relatedness of partners’business and relatedness of foreign partners and IJV business.The researchers concluded that the more resources an organisation contributes to the partnership,the greater likelihood that it would be selected as a partner. Recognising the range of different criteria used,Geringer (1991)made a landmark contribution to understanding partner selection by developing a typology based on two types of criteria;namely task and partner related.T ask-related criteria include patents,technical knowledge,experience of management,access to marketing and distribution systems, andﬁnancial resources–in other words the operational skills and resources a joint venture requires to be competitive (T atoglu,2000).In contrast,partner-related criteria comprise the variables which are speciﬁc to the character,culture and history of the partners,for example past association,national or corporate culture,organisational size or structure and the compatibility or trust between the partners’management teams(Glaister and Buckley,1997).As such,partner-related criteria are concerned with the effectiveness of cooperation between IJV partners(Al-Khalifa and Peterson,1999).T esting Geringer’s(1991)typology in a study of IJVs between UKﬁrms and European,US and Japanese partners, Glaister and Buckley(1997)identiﬁed the most important task-related criteria as access to local market and cultural knowledge,distribution channels and links with major buyers and the most important partner-related criteria as trust between management teams,relatedness of partner business and reputation.The authors concluded that IJV partners should possess both types of criteria.However,in a study of 42IJVs in Bahrain,Al-Khalifa and Peterson(1999) concluded that partner-related criteria are the dominant criteria,particularly for largerﬁrms with more IJV experience. In particular,their study determined that reputation, experience and personal knowledge of the partner organisations and the personal characteristics of the CEO are important factors in IJV partner selection.T atoglu(2000) also found a greater reliance on partner-related criteria using Geringer’s(1991)typology to investigate IJVs between local ﬁrms in Turkey and Western partners.His study further determined that the most important partner-related criteria are trust between top management teams and reputation of partners,and that favourable past association,size and international experience were relatively unimportant.He also includes partner’s knowledge of the local market as key partner-related criteria.While T atoglu(2000)also identiﬁes the most important task-related selection criteria as access to knowledge of local markets and culture,he concludes that that task criteria will be“somewhat speciﬁc to the underlying purpose of the IJV”(p.144).In other words,ﬁrms will seek complementary resources and skills relevant to the joint venture.Similarly,Al-Khalifa and Peterson(1999)advise that ﬁrms seek compatibility in relation to partner-related criteria. Compatibility of organisational cultures,goals and works systems is also well recognised as an essential ingredient of successful alliance agreements(see,for example,Kanter, 1994;Buono,1997;Kauser and Shaw,2004).Although relatively scarce in comparison,empirical studies on partner selection in franchise agreements have also yielded a range of partner selection criteria.For example, Jambulingam and Nevin(1999)identiﬁedﬁnancial capability,experience and management skills,demographic characteristics and attitudes towards business dimensions as important selection criteria used by franchisors.Clarkin and Swavely’s(2006)study of1,043multi sector North American franchisors identiﬁed the importance of franchisees’personal characteristics,includingﬁnancial strength,attitudes and personality,psychological proﬁling,formal education,general business and industry speciﬁc experience.Similarly,Hsu and Chen(2008)identiﬁed the operating ability,ﬁnancial capability,experience and personality as selection criteria for retail franchisors in T aiwan.The authors conclude however, thatﬁnancial and business ability are the key traits for franchisors to determine.Within international franchise agreements,Doherty and Alexander(2004)examined partner need recognition,selection search and evaluation from a relationship marketing perspective.The authors found that the“right chemistry”(p.1224)between partners was important to the retail franchisors in their study,irrespective of other speciﬁc criteria used.However,Choo et al.(2007) found three key criteria used by international fast food retail franchisors in Singapore to be:1ﬁnancial strength;2ability to secure prime retail space;and3knowledge of the local market.These studies suggest that both partner-related characteristics and task-related characteristics as deﬁned by Geringer(1991) are used to variable extents in the franchise selection process, as did Altinay(2006)in his study of international hotel franchising.However,Altinay(2006)contributes further to our understanding of international franchisee partner selection by identifying that the emphasis placed on these criteria varies in different stages of the decision making process.Greater emphasis is placed on partner-related criteria during the early stages of the selection process,and the partner-related traits of the franchisees are used by franchisor members to determine whether potential franchisees have the ability and the background to meet the task-related criteria. Franchisor members thoroughly analyse their organisation’s expectations and compare the franchisees’current and future task-related capabilities to those deemed necessary for franchise success during the selection process.Altinay (2006)concludes that there are three important contextual variables which have a bearing on the franchisee selection criteria employed:1the strategic context of the organisation;2different country markets;and3the nature of the business itself(franchise partnership). Furthermore,given the socio-cultural differences between country markets,the study emphasises the importance of having a selection process in place as well as clear criteria in order to make sound selection decisions.Doherty(2009)also recognises the importance of both the process of partner selection and the criteria used in international retail franchising and reports that franchisors adopt both strategic and opportunistic approaches to the process.The key difference between these two approaches is whether it is the franchisor that initiates the process(strategic) or the franchisee that approaches the franchisor (opportunistic).Her study reveals that in strategic partner selection,there is a deﬁned process whereby appropriate potential franchise partners are identiﬁed after a market has been selected and speciﬁc selection criteria are used to short list and then select the franchise partner.Selection criteria identiﬁed includeﬁnancial stability,business know how,local market knowledge,a shared understanding of the brand and strategic direction of the business,and chemistry between the franchisor and franchisee.In other words,a mix of task and partner-related criteria.In opportunistic approaches, franchisors in the study followed a set process for setting out their terms and conditions and for the potential franchisee to submit a business development plan for approval.The criteria used to evaluate partners therefore,appears to relate predominantly toﬁnancial criteria and in these approaches, partner selection precedes market selection.Despite the contribution of these studies,the need for further partner selection research is well recognised.Altinay and Miles(2006)note that employing strict selection criteria and a deﬁned selection process can help to control the behaviour of franchisees prior to the establishment of the partnership and thus help in their integration into the wider franchise system.Partner selection criteria can also be used to train and educate franchisees to ensure goal congruence before the partnership is established and can help to develop relationships between potential partners during the negotiations process(Altinay,2006).Effective management of relationships is well recognised as a critical ingredient of effective alliances in general(Kanter,1994;Buono,1997; Kauser and Shaw,2004)and franchise partnerships in particular(Hopkinson and Hogarth-Scott,1999;Quinn, 1999;Doherty and Alexander,2004;Clarkin and Swavely, 2006;Weavin and Frazer,2007a,b).Relationship management betweenﬁrms is frequently referred to as social or relational control as it contributes to the development of shared norms and organisational practices and serves to better coordinate inter-ﬁrm activities(Buono, 1997;Child and McGrath,2001;Weavin and Frazer,2007a, b).These processes can also help to break down barriers between organisations and increase the permeability between ﬁrms(Martinez and Jarillo,1989;Dess et al.,1995)to help achieve partnership goals.The review of the literature reveals that a wide range of partner and task-related criteria have been identiﬁed in both IJV and franchise research.T able I provides a summary of the speciﬁc criteria and highlights the different importance of criteria identiﬁed in the studies reviewed.Furthermore,T able I reveals that empirical studies undertaken within franchising have examined partner selection from the perspective of the franchisor only.There remains a gap in our understanding of the criteria used by each partner and the combined impact that the franchisor and franchisee might have on the selection process.As Clarkin and Swavely(2006,p.133)advise,“despite franchising’s ubiquity,how franchisors and franchisees select each other remains a largely unexplored topic”.Furthermore,there is a need for further research to determine how different selection criteria are used by both partners throughout the partner selection process and whether these are inﬂuenced by contextual variables and the selection approach used.Given these gaps,this paper examines partner selection within international master franchise agreements,from the perspective of both by franchisors and franchisees.It evaluates the importance ofTable I Keyﬁndings from empirical studies in partner selectionContext of study Authors(year)Examined from Partner selection criteria identiﬁed KeyﬁndingsUK IJVs in India and Pakistan Tomlinson(1970)Perspective of UK based parentcompanySix criteria(in order of importance):1favourable past association2resources3facilities4partner status5forced choice6local identityIn addition to ranking the criteria,threekey contextual variables were alsoidentiﬁed:1parent size2nature of business3motivation behind IJV formationIJVs between Canadian and Mexicanﬁrms TomlinsonandThompson(1977)Perspective of both IJV partners Canadianﬁrms(ﬁve criteria):1ﬁnancial status2compatibility in business3similar objectives4ability to negotiate with government5common ethicsMexican Firms(six criteria):1ﬁnancial resources2technological resources andexperience3international prestige andexperience4commitment to JV5sound management6ability to communicateDifferent criteria and priority of criteria byCanadian and Mexicanﬁrms.Financialresources important toﬁrms from bothnational contextsUS IJVs with European, Canadian and Japanese partners(manufacturing)Awadzi et al.(1988)Perspective of US partner Four criteria(in order of importance):1resource contribution2past association between partners3relatedness of partners’business4relatedness of foreign partners’andIJV businessCooperation between JV partners andcomplementarity of resources enhance theIJV performanceUS-based IJVs (manufacturing)Geringer(1991)Perspective of US-based parentcompanyTwofold typology of criteria identiﬁed:1task criteria2partner criteriaTask criteria relate to skills and resources;partner-criteria concerns effectiveness ofcooperationThe relative importance of speciﬁc task-related criteria is determined by thestrategic context of the proposed IJVUK multi-sector IJVs with European,US and Japanese partners Glaister andBuckley(1997)Perspective of UK partners In order of importance,12partner-relatedcriteria:1trust between top managementteams2relatedness of partner’s business3reputation4ﬁnancial status5complementarity of resources6marketing and distribution systems7size8international experience9technological experience10management in depth11degree of favourable pastassociation12ability to negotiate withgovernment12task-related criteria:1knowledge of local market2distribution channels3links with major buyers4knowledge of local culture5technology6the product itself7knowledge of production processes8capital9regulatory permits10labour11local brand names12materials/natural resourcesBoth task and partner criteria areimportant to partner selection andpartners should possess both types ofcriteria(continued)Context of study(year)Examined from Partner selection criteria identiﬁed KeyﬁndingsManufacturing IJVs in Bahrain Al-Khalifaand Peterson(1999)Perspective of IJV CEOs in Bahrain,from range of unspeciﬁed countries14criteria(in order of importance):1reputation in local market2ﬁnancial status3similar goals,objectives andaspirations4enthusiasm and commitment5contacts in local market6compatibility of organisations7knowledge of local/hostmarket8ability to cover territory9prior trade relationship10technical competence11adequate staff to marketeffectively12recommendations by bank,government,etc.13previous JV success14prior JV experiencePartner-related criteria are the dominantcriteria in the partner selection process.Companies with multi-country JVexperience place more emphasis onpartner-related criteriaUS franchises fromﬁve industrial sectors Jambulingamand Nevin(1999)Perspective of the franchisor,but datacollected from franchisees,some ofwhich were master franchiseesImportant criteria include franchiseeattitudes towards business,includingperceived innovativeness and personalcommitment to the businessControlling the quality of the franchiseesusing appropriate selection criteria canimprove the efﬁciency of the contractualrelationship for the franchisor.Franchiseeattitudes are more important thantraditional selection criteria typically usedby franchisorsTurkish mutli-sector IJVs with European and US partners Tatogolu(2000)Perspective of foreign partnerresponsible for Turkish operationIn order of importance,15partner-relatedcriteria:1knowledge of local market2trust between management teams3reputation of partner4ability to negotiate withgovernment5compatibility of managementteams6quality of management team7ﬁnancial resources8size9favourable past association10established marketing anddistribution11relatedness of business12complentarity of resources13experience of technology14ability to raise funds15International experienceNine task-related criteria:1access to knowledge of localmarket23access to knowledge of localculture4access to regulatory permits5access to labour6access to capital7access to natural resources8access to technology9access to productThere is a greater reliance on partner-related criteria;task criteria are speciﬁc tothe underlying purpose of the IJVUK-based fashion retailers with international franchisees Doherty andAlexander(2004)Perspective of UK franchisor Main criteria identiﬁed1ﬁnancial stability2business know-how3partner’s local marketAlso important:1like-minded partners2chemistry between partnersFinancial stability is a crucial factor inpotential partner identiﬁcation,butmutual attraction and relationshippotential are important when makingpartner choices(continued)speciﬁc selection criteria at the different stages of the partner selection process and assesses how different contextual variables including type of selection approach used(strategic or opportunistic)and the nature of business inﬂuence partner selection in master franchise partnerships.The following section reports on the design of the study.3.Research designIn order to examine partner selection from franchisor and franchisee perspectives,a single,embedded qualitative case study strategy was employed.Doherty and Alexander(2004) advocate that this approach provides an alternative to the numerous positivistic and quantitative franchise studies. Doherty(2007,2009)argues qualitative studies are increasingly employed within international franchise research so that a better understanding of operational issues can be determined.In addition,they provide an opportunity to obtain rich data(Moore et al.,2004)illustrating real-life organisational experiences.The case comprises a US hotel franchisor and its two international master franchisees,referred to in this study as the Franchisor,Franchisee A and Franchisee B.While both franchisees are from Europe,they are headquartered in different countries and have different geographical territories covered in their franchise agreements.There is also some discrepancy between the market level of the portfolios,and while recognised by the franchise partners,this did not prevent the formation of the franchise partnerships.The study explores the dyadic partner selection process and the criteria used between the Franchisor and each Franchisee.The territory of the case is bound(Miles and Huberman,1999) within one international hotel brand within the Franchisor’s multi-branded portfolio.T able II provides a proﬁle of the operating characteristics of the franchiseﬁrms in the study. Primary data was collected using semi-structured key informant interviews,a practice frequently used in organisational studies as it provides an economical approach to gaining“global”data on organisations(Bryman and Burgess,1994,p.49).The interviews sought to identify the partner selection process and the speciﬁc criteria used from the perspectives of the franchisor and each franchisee for each franchise agreement as advised by Doherty(2009).The interviews examined each partnership from the point of consideration of a franchise until the contract signature. Multiple interviews were conducted with corporate levelContext of study(year)Examined from Partner selection criteria identiﬁed KeyﬁndingsCase study of one UK international hotel franchisor Altinay(2006)Perspective of UK franchisor Five criteria identiﬁed:1general background2ﬁnancial strength3expertise4partner strategy/rationale forpartnership5howﬁnance projectGreater emphasis is placed on partner-related criteria during the early stages ofthe selection process.Three contextualvariables identiﬁed include strategiccontext,different country markets andnature of businessUS fast food franchises in Singapore Choo et al.(2007)Perspective of US franchisor Three criteria identiﬁed:1ﬁnancial strength to launch andgrow brand2access to prime real estate sites3local knowledge to adapt brand tosuit marketThree key criteria identiﬁed for East AsianmarketsMulti-sector North American franchisors Clarkin andSwavely(2006)Perspective of NA franchisor,drawson secondary dataFour criteria identiﬁed(in order ofimportance):1personal interview2ﬁnancial net worth3general business experience4psychological proﬁle,i.e.formaleducation and speciﬁc industryexperienceAttitudes and personality are importancecriteria in franchisee selectionCase study of retail franchisor in Taiwan Hsu andChen(2008)Perspective of franchisor In order of importance,personal criteria:1personal background2ﬁnancial situation3business abilityStore location:1consumer purchasing power2footfall3parking conveniencePersonal condition(criteria)of franchiseeis more important than store location,andﬁnancial and business ability are ofparticular importanceUK international retail franchisors Doherty(2009)Perspective of UK franchisor Five criteria used in strategic approach:1ﬁnancial stability2business know-how(preferably inretail)3local market knowledge4shared understanding of brand andstrategic future5shared chemistryOne criterion,i.e.ﬁnancial capability,usedin opportunistic approachBoth selection criteria and processimportant,but selection criteria differ asto whether a strategic or opportunisticapproach is adopted.In opportunisticapproaches,partner selection precedesmarket selection。

Intelligent Transportation Systems

Intelligent Transportation Systems Intelligent Transportation Systems Intelligent Transportation Systems (ITS) refer to the application of advanced technologies in the field of transportationto improve safety, efficiency, and sustainability. ITS encompasses a wide range of solutions, including traffic management systems, electronic toll collection, and vehicle-to-infrastructure communication. In this essay, we will explore thebenefits and challenges of ITS, as well as its potential impact on the future of transportation. One of the key benefits of ITS is its ability to improve traffic flow and reduce congestion. Advanced traffic management systems can monitor and control the flow of vehicles, optimizing traffic signal timings and providingreal-time traffic information to drivers. This not only reduces travel time for commuters but also decreases fuel consumption and air pollution. Additionally, electronic toll collection systems enable seamless and efficient payment on highways, eliminating the need for toll booths and reducing traffic congestion at toll plazas. Furthermore, ITS has the potential to enhance road safety throughthe implementation of technologies such as collision avoidance systems and automated vehicle control. These systems can detect and respond to potential hazards on the road, significantly reducing the number of accidents and fatalities. Vehicle-to-infrastructure communication can also improve safety by enabling vehicles to receive real-time information about road conditions and potential dangers. In addition to improving safety and efficiency, ITS can also have a positive impact on the environment. By reducing traffic congestion and optimizing traffic flow, ITS helps to minimize fuel consumption and greenhouse gas emissions. Furthermore, the integration of electric and autonomous vehicles into ITS can further reduce the environmental impact of transportation. However, despite its numerous benefits, the implementation of ITS also poses several challenges. One of the main challenges is the high cost of deploying and maintaining advanced technologies. Additionally, the interoperability of different ITS solutions andthe standardization of communication protocols are important issues that need tobe addressed to ensure the seamless integration of ITS systems. Another challenge is the potential impact of ITS on employment in the transportation sector. The automation of certain transportation tasks, such as toll collection and vehiclecontrol, could lead to job displacement for workers in these fields. It is crucial for policymakers to consider the social and economic implications of ITS and to develop strategies to mitigate any negative effects on employment. In conclusion, Intelligent Transportation Systems have the potential to revolutionize the way we travel, offering numerous benefits in terms of safety, efficiency, and sustainability. However, the successful implementation of ITS requires addressing various challenges, such as cost, interoperability, and employment impact. With careful planning and investment, ITS has the power to transform the future of transportation for the better.。

智能交通英语词汇

JJitney 随停公车；简便公车Job Mix Formula 工地拌杂公式Joint Opening 开口宽度；Joint Operation of Transport 联运Jointed Concrete Pavement JCP 接缝式混凝土铺面Jointed Reinforced Concrete Pavement JRCP 接缝式钢筋混凝土铺面Joule 焦耳〔能量单元〕Journal Resistance 轴颈阻力Jumbo 钻堡Junction 路口Junction box 汇流井KKalman filter algorithm 卡门滤波法Kalman filter, Kalman filtering 卡门滤波Keep right sign 靠右标识表记标帜Key count station 关键查询拜访站Kinematic Viscosity 动黏度Kink 纽结Kiss-and-Ride 泊车转乘Kneading Action 辗挤作用Knot 节〔等于每小时1.85公里〕Knowledge Base 常识库Kurtosis 峰度LLag 间距Lag time 延迟时间Land Access 可及性Land Expropriation 地盘征收Land Transportation 陆路运输Land use 地盘使用；地盘操纵Landfill site 掩埋场Landscape design 景不雅设计Landslide/Slump 坍方Lane, traffic lane 车道Lane 巷道Lane allocation, Lane layout 车道配置Lane Balance 车道平衡Lane change 变换车道Lane control 车道管制Lane distribution 车道分布Lane Group 车道群Lane headway 车道行进间距Lane Line 车道线Lane Reduction Transition Line 路宽渐变线Lane residual width 巷道残剩宽度Lane Width 车道宽度Lane-direction control signal 车道行车标的目的管制号志Large Network Grouping 大型网络群组划分Large Passenger Vehicle 大客车Large-area Detector 大区域侦测器Latent travel demand 潜在旅次需求Lateral Acceleration 横向加速率Lateral clearance 侧向净距Lateral Collision Avoidance 侧向防撞Lateral Separation 摆布隔离Lateral shift 侧向位移Laws of randomness 随机定理Lay-Down Vehicle Days 停驶延日车数Leading 绿灯早开Leading & lagging design 早开迟闭设计Leading Car 带领车Leading design 早开设计Learning Permit 学习驾照Lease 租赁Left turn 左转Left turn accel-decel & storage lane island 左转加减速－停储车道式分向岛Left turn acceleration lane island 左转加速车道式分向岛Left turn crossing 左转交叉穿越Left turn lane 左转车道Left turn maneuver 左转运行Left turn on red 在红灯时段内进行左转运行Left turn waiting zone 左转待转区Left turning vehicle 左转车辆Leg 路肢Legibility 公认性；易读性Length 长度Length of Economical Haul 经济运距Length of grade 坡长Length of lane change operation 变换车道作业的长度Length of Superelevation Runoff 超高渐变长度Length of time parked 泊车时间的久暂Level 水准仪；横坑；程度面Level Crossing 平面交叉Level of illumination 敞亮度水准Level of Service 效劳水准Level or flat terrain 平原区Level surface 水准面Leveling Course 整平层License Plate 汔车号牌License Plate Method 牌照法License Suspension 吊扣驾照License Termination 裁撤License Plate 汽车号牌License plate method 车辆牌照记录法；车辆牌照法License Plate Recognition 车牌辨识License renewal method 换照法Life cycle assessments LCA 生命周期评估Lift-On／Lift-Off 吊上吊下式Light Characteristics 灯质Light List 灯质表Light Motorcycle 轻型机踏车Light on method 亮灯法Light Phase 灯相Light Rail Rapid Transit LRRT 轻轨捷运Light Rail Transit LRT 轻轨运输Light Truck 小货车Lighting System 灯光系统Limited purpose (parking) survey 局部目的〔泊车〕查询拜访Line Capacity 路线容量Line marking 标线Linear Referencing System 线性参考系统Linear Shrinkage 线收缩Linear-Induction Motor LIM 线性感应马达Linear-Synchronous Motor LSM 线性同步马达Link arrival rate 路段流量达到率Link flow 路段流量Link performance function 道路绩效函数Linked or coordinated signal system 连锁号志系统Lip Curb 边石Liquidate 变成液体：偿还：破产Liquidated Damage 违约罚金Load Equivalent Factor LEF 荷重当量系数；载重当量因素Load Factor 负荷指数Load limit 载重限制Load Safety Factor 载重安然因素Loading 载重Loading & unloading 装卸Loading & unloading zone 上下搭客区段或装卸货物区段Loading Island 搭客上下车的车站岛Local Area Network LAN 局域网络Local Controller 路口控制器Local street 地域性街道Local traffic 地域性交通Local transmission network 区域传输网络Localizer 摆布定位台Location file 地址档案Location of stop 站台设置位置Locked Joint 连锁接头Log Likelihood Function 对数概似函数Logical Architecture 逻辑架构Logit Model 罗吉特模式Logo 标识表记标帜；商标Long Loop 长线圈Long tunnel 长地道Long tunnel system 长地道系统Long vehicle tunnel 车行长地道Long-chord 长弦Longitude 经度Longitudinal Collision Avoidance 纵向防撞Longitudinal distribution of vehicle 车辆的纵向分布Longitudinal Drain 纵向排水Longitudinal Grade 纵坡度Longitudinal Joint 纵向接缝Longitudinal Separation 前后隔离Longitudinal Slope For Grade Line 纵断坡度Longitudinal ventilation 纵流式通风Longitudinal Warping 纵向扭曲Long-Range Planning 长程规划Long-term scour depth 持久冲刷深度Loop 环道〔公路方面〕；回路〔电路方面〕Loop Detector 环路型侦测器Loop inductance 感应回路Los Angeles Abrasion Test 洛杉矶磨耗试验Lost Time 损掉时间Louvers of Daylight Screening Structure 遮阳隔板Low beam 近灯Low Heat 低热Low or first gear 低速檔Low Pressure Sodium Lamp 低压钠气灯Low relative speed 低相对速率Low Truss 低架式Lumen 流明Luminance 辉度Luminaire 灯具Luminous Efficiency of a Source 光源效应Luminous flux 光流；光束Luminous Intensity 光度；光强度Lux 勒克斯MMacro or mass analysis 汇总阐发；宏不雅阐发Macroscopic 巨不雅Magnetic detector 电磁〔磁性〕侦测器Magnetic Levitation Maglev 磁浮运输系统Magnetic loop detector 磁圈侦测器Magnitude 规模Mainline 主线Maintenance Factor 维护系数Maintenance Work 养护工程Major flow 主要车流Major parking survey 主要泊车查询拜访Major Phase 主要时相Management Information Base MIB 网管信息库〔办理讯息库〕Management Information System MIS 办理信息系统Maneuverability 运行性Manhole 人孔；井Man-machine driving behavior 人机驾驶行为Man-machine interaction 人机互动Manual counts 人工查询拜访法Map Matching Method 地图配对法Map scale 图比例尺Marginal vehicle 边际车辆Marker 标物；标识表记标帜Market Package 产物组合Market Segment 市场区隔Marking 标线Marshaling Yard 货柜堆积场Marshall Test 马歇尔试验Mass Diagram 土积图Mass transportation 群众运输Match Fund 配合款Master 主路口Master controller 主要〔总枢纽〕控制器Master Node 主控制点Master Plan 主计画Material handling 物料搬运Maturity 成熟程度Max out 绿灯时间完全使用之现象Maximum allowable gradient 最大容许坡度Maximum Allowable Side Friction Factor 最大容许侧向摩擦系数Maximum Arterial Flow Method 最大干道流量法Maximum capacity 最大容量Maximum Density 最大密度Maximum flow rate 最大流率Maximum Grade 最大坡度Maximum individual delay 最大个别延滞Maximum Likelihood Function 最大要似法Maximum Load Section MLS 最大承载区间Maximum Peak Hour Volume 最尖峰小时交通量Maximum possible rate of flow 最大可能车流率Maximum queue 最大等待量〔车队长度〕Maximum Theoretical Specific Gravity 最大理论比重Mean Absolute Value of Error MAE 平均绝对误差Mean deviation 平均差Mean difference 均互差Mean Square Error MSE 平均平方误差Mean variance 离均差Mean velocity 平均速度Measure of Effectiveness MOE 绩效评估指针Mechanic License 技工执照Mechanical counter 机械式计数器Mechanical garage 机械式泊车楼〔间〕Mechanical Kneading Compactor 揉搓夸压机Mechanical Load-Transfer Devices 机械传重设备Mechanical parking 机械泊车Mechanical power 机械动力Median 中央分隔带；中央岛Median (50th percentile) speed 中位数速率；第50百分位数速率Median Bus Lane 设于道路中央之公车专用车道Median conflicts 中央冲突点Median Curing Cutback Asphalt MC 中凝油溶沥青Median island 中央岛Million Vehicles Kilometer MVK 百万车公里Median opening 中央分向岛缺〔开〕口Medium Distribution 中分布Memorandum of Understanding MOU 备忘录Mental factor 精神因素Mercury Vapor Lamp 水银蒸汽灯Merge 合并；并流；进口匝道；并入Merging area 并流区域Merging behavior 并入行为Merging conflicts 并流冲突点Merging maneuver 并流运行Merging of traffic 交通汇流Merging point 并流点Merging traffic 调集交通Message Set 讯息集Metadata 诠释资料Metered Freeway Ramp 匝道仪控Metering rate 仪控率Metropolitan Planning Area 大城市规划区Micro or spot analysis 重点阐发；微不雅阐发Microwave Beacon 微波信号柱Mid-block (bus) stop 街廓中段公车站台Mid-block delay 街廓中段延滞Mid-Block Flow 半途转入流量Middle Ordinate 中距Mide-Block 街廓中央停靠方式Mile Per Gallon MPG 每加仑油量可跑的英哩数Milled Materials 刨除料Million Vehicles Kilometer MVK 百万车公里Mini car 迷你车Minibus 小型公车Minimum Curve Length 曲线之最短Minimum Design for Turning Roadway 转向道之最小设计Minimum Grade 最小纵断坡度Minimum Green Time 最短绿灯时间Minimum Phase Time 最短时相时间Minimum running time 最短行车时间Minimum separation 最小间距Minimum sight triangle 最小视界三角形Minimum speed limit 最低速率Minimum turning radius 最小转弯半径Minimum-speed curve 最低速率曲线Minor flow 次要车流Minor Phase 次要时相Minor street 次要道路Mix Design 配合设计Mixed fleet 混合车队Mixed fleet operation 混合车队营运Mixed flow, Mixed traffic, Mixed traffic flow 混合车流Mobile Communications 步履通讯Mobile Data 步履数据Mobile Data Network 步履数据网络Mobile Plant 移动式厂拌；活动式厂拌Mobile radio unit 车装式无线机Mobility 可行性Modal Split 运具分配Mode 运具Model traffic ordinance 榜样交通条例Modem 通讯解调器Modular 模块化Modulus of Elasticity 弹性系数Modulus of Rupture 破裂模数Modulus of Subgrade Reaction 路基反响系数Moment of inertia 惯性力矩Monitoring 监测Monorail 单轨铁路Mortality 死亡数Motivation 动机Motor License 行车执照Motor transport service 汽车运输业Motor vehicle code 机动车辆尺度Motorcycle lane 机车道Motorcycle user 机车使用者Motorcycle waiting zone 机车停等区Motorcyclist 机车驾驶人Motorcyclist Safety 机车骑士安然Mountable Curb or Rolled Curbs 可越式绿石Mountain road 山区道路Mountain terrain 山岭区Mounting height 装设高度Movement Distribution 流向分配Movement-Oriented 流动导向Moving belt 输送带Muck 碴Mud 泥浆Multi-Trip Ticket 多程票或回数票Multicommodity flow problem 多商品流量问题Multicommodity network flow problem 多重货物网络流动问题Multi-function Alarm Sign 多功能警示标识表记标帜Multilane highway 多车道公路Multi-lane rural highway 多车道郊区公路Multilayer 多层Multileg Interchange 多路立体交叉Multileg Intersection 多路交叉Multimodal 多运具的Multimode mixed traffic 多车种车流Multi-parameter detector 多参数侦测器Multi-path effect 多路径效应Multipath Traffic Assignment Model 多重路线指派模式Multiperiod 多时段Multiple commodity network flow problem 多重商品网络流动问题Multiple cordon survey 多环周界查询拜访Multiple network flow problem with side constraint 含额外限制多重网络流动问题Multiple turning lane 多线转向车道Multiple use area 多用途空间Multiple user classes 多种用路人Multiple-ride-ticket 回数票Multistories or multifloor garage 高楼泊车间Multi-Tasking 多任务作业NNaphtha 石脑油National Cooperative Highway Research Program NCHRP 美国公路合作研究组织群National freeway 国道National Freeway Construction & Management Fund 国道公路建设办理基金National freeway network 国道路网National System Architecture 国家级架构National System of Interstate and Defense Highway 洲际国防公路Native Asphalt 天然沥育Natural (normal) distribution 自然〔常态〕分布Natural Disaster Traffic Management 自然灾害交通办理Natural moisture content 自然含水量Natural path 自然迹线Natural Rubber Latex 天然橡胶流质Natural Subgrade 天然路基Natural ventilation 自然通风Natural ventilation effect 自然通风效应Navigation 引导；导航Near-Side 路口近端Near-side bus stop 近端公车停靠站台Neon regulatory sign 霓虹式禁制标识表记标帜Net Tractive Effort 净牵引力Net Weight 净重Network 网络New Jersey concrete barrier 纽泽西混凝土护栏Night visibility 夜间可见〔视〕性Night vision 夜间视力Nighttime driving 夜间驾驶Nitrogen Oxides NOX 氧化碳No left turn 不准左转；请勿左转No parking 禁止泊车No parking on yellow line 禁止泊车黄线No passing zone 不准超车区段No passing zone marking 不准超越地带标线No right turn 不准右转；请勿右转No standing on red line 禁止临时泊车红线No turning 不准转向运行No U-turn 不准回转；请勿回转Noise 噪音Noise analysis 噪音阐发Noise barrier, Sound insulating wall 隔音墙Noise control 噪音控制Noise induced annoyance 噪音干扰Noise intensity 噪音强度Noise Level 噪音水准Noise pollution 噪音污染Noise prevention 噪音防制Noise sensitive area 噪音敏感地域Nomograph 换算图Non-collision accident 非碰撞性闯祸Non-Cutoff 无遮被型Non-Cutoff type NC/O 无遮蔽型Nonfatal Injury 非致命性的伤害Nonhomogeneous flow 不同流向的车流；非均质车流Non-Motorist 非机动车使用者〔如行人〕Non-Overlap 非重叠时相Nonpassing Sight Distance 不超车视距Non-Recurrent Congestion 非重现性交通壅塞Non-road user 非用路者Nonskid Surface Treatment 防滑处置Nonsynchronous controller 异步控制器No-Passing Zone Markings 禁止超车路段线Normal Crown NC 正常路拱Not approved 认定不该开发Novelty 别致性Number of accident 闯祸件数Number of conflict 冲突点数目Number of fatality 死亡人数Number of injury 受伤人数Number of Operating Vehicles 营业车辆数Number of parking spaces 泊车车位数Number of Passengers 客运人数Number of Registered Vehicle 车辆登记数Numbers 要求之重复Numerical speed limit 数值速率限制Nurture room 育婴室加载数据网站地图| 设为首页| 参加保藏首页┆资讯┆技术┆ITS人物┆产物┆公司┆解决方案┆应用案例┆ITS文库┆下载┆视频┆问吧┆论坛首页交通规划城市规划智能交通公共交通物流运输交通经济静态交通组织办理交通统计3S技术企业办理交通办理新闻搜索产物搜索企业搜索软件搜索视频搜索热门标签：智能交通(13) 电子差人(9) ETC(7) 不泊车收费系统(5) 英语词汇(3) 交通变乱(2) 当前位置：智能交通不雅察→ITS文库→交通办理→交通控制交通控制英语词汇大全ITS不雅察来源：ITSobserve 日期：2021-04-02 交通新闻投稿智能交通论坛智能交通不雅察网保举您阅读:交通控制英语词汇大全Stop-line——泊车线A congested link——阻塞路段Weighting factor——权重因子Controller——控制器Emissions Model——排气仿真the traffic pattern——交通方式Controller——信号机Amber——黄灯Start-up delay——启动耽搁Lost time——损掉时间Off-peak——非颠峰期The morning peak——早颠峰Pedestrian crossing——人行横道Coordinated control systems——协调控制系统On-line——实时Two-way——双向交通Absolute Offset——绝对相位差Overlapping Phase——搭接相位Critical Phase——关键相位Change Interval——绿灯间隔时间Flow Ratio——流量比Arterial Intersection Control 干线信号协调控制Fixed-time Control——固定式信号控制Real-time Adaptive Traffic Control——实时自适应信号控制Green Ratio——绿信比Through movement——直行车流Congestion——阻塞，拥挤The percentage congestion——阻塞率The degree of saturation——饱和度The effective green time——有效绿灯时间The maximum queue value——最大排队长度Flow Profiles——车流图示Double cycling——双周期Single cycling——单周期Peak——颠峰期The evening peak periods——晚颠峰Siemens——西门子Pelican——人行横道Fixed time plans——固定配时方案One-way traffic——单向交通Green Ratio——绿信比Relative Offset——相对相位差Non-overlapping Phase——非搭接相位Critical Movement——关键车流Saturation Flow Rate——饱和流率Isolated Intersection Control——单点信号控制〔点控〕Area-wide Control——区域信号协调控制Vehicle Actuated (VA)——感应式信号控制The Minimum Green Time——最小绿灯时间Unit Extension Time——单元绿灯耽误时间The Maximum Green Time——最大绿灯时间Opposing traffic——对向交通〔车流〕Actuation——Control——感应控制方式Pre-timed Control——定周期控制方式Remote Control——有缆线控方式Self－Inductfanse——环形线圈检测器Signal——spacing——信号间距Though-traffic lane——直行车道Inbound——正向Outbound——反向第一章交通工程——Traffic Engineering运输工程——Transportation Engineering铁路交通——Rail Transportation航空交通——Air Transportation水上交通——Water Transportation管道交通——Pipeline Transportation交通系统——Traffic System交通特性——Traffic Characteristics人的特性——Human Characteristics车辆特性——Vehicular Characteristics交通流特性——Traffic Flow Characteristics道路特性——Roadway Characteristics交通查询拜访——Traffic Survey交通流理论——Traffic Flow Theory交通办理——Traffic Management交通环境庇护——Traffic Environment Protection 交通设计——Traffic Design交通统计学——Traffic Statistics交通心理学——Traffic Psychology汽车力学——Automobile Mechanics交通经济学——Traffic Economics汽车工程——Automobile Engineering人类工程——Human Engineering环境工程——Environment Engineering自动控制——Automatic Control应用数学——Applied Mathematics电子计算机——Electric Computer第二章公共汽车——Bus无轨电车——Trolley Bus有轨电车——Tram Car大客车——Coach小轿车——Sedan载货卡车——Truck拖挂车——Trailer平板车——Flat-bed Truck动力特性——Driving Force Characteristics牵引力——Tractive Force空气阻力——Air Resistance滚动阻力——Rolling Resistance坡度阻力——Grade Resistance加速阻力——Acceleration Resistance附出力——Adhesive Force汽车的制动力——Braking of Motor Vehicle自行车流特性——Bicycle flow Characteristics 驾驶员特性——Driver Characteristics刺激——Stimulation感觉——Sense判断——Judgment步履——Action视觉——Visual Sense听觉——Hearing Sense嗅觉——Sense of Smell味觉——Sense of Touch视觉特性——Visual Characteristics视力——Vision视野——Field of Vision色彩感觉——Color Sense眩目时的视力——Glare Vision视力恢复——Return Time of Vision动视力——Visual in Motion亮度——Luminance照度——Luminance反响特性——Reactive Characteristics刺激信息——Stimulant Information驾驶员疲劳与兴奋——Driving Fating and Excitability 交通量——Traffic Volume交通密度——Traffic Density地址车速——Spot Speed瞬时车速——Instantaneous Speed时间平均车速——Time mean Speed空间平均车速——Space mean speed车头时距——Time headway车头间距——Space headway0交通流模型——Traffic flow model自由行驶车速——Free flow speed阻塞密度——Jam density速度－密度曲线——Speed-density curve流量－密度曲线——Flow-density curve最正确密度——Optimum concentration流量——速度曲线——Flow-speed curve最正确速度——Optimum speed持续流——Uninterrupted traffic间断流——Interrupted traffic第三章交通查询拜访阐发——Traffic survey and analysis交通流查询拜访——Traffic volume survey车速查询拜访——Speed survey通行能力查询拜访——Capacity survey车辆耗油查询拜访——Energy Consumption Survey 居民出行查询拜访——Trip Survey车辆出行查询拜访——Vehicle Trip Survey泊车场查询拜访——Parking Area Survey交通变乱查询拜访——Traffic Accident Survey交通噪声查询拜访——Traffic Noise Survey车辆废气查询拜访——Vehicle Emission Survey平均日交通量——Average Daily Traffic(ADT)周平均日交通量——Week Average Daily Traffic月平均日交通量——Month Average Daily Traffic年平均日交通量——Annual Average Daily Traffic颠峰小时交通量——Peak hour Volume年最大小时交通量——Highest Annual Hourly Volume年第30位最高小时交通量——Thirtieth Highest Annual Hourly Volume 颠峰小时比率——Peak Ratio时间变化——Time Variation空间变化——Spatial Variation样本选择——Selection Sample样本大小——Size of Sample自由度——Freedom车速分布——Speed Distribution组中值——Mid-Class Mark累计频率——Cumulative Frequency频率分布直方图——Frequency Distribution Histogram85%位车速——85% Percentile Speed限制车速——Regulation Speed效劳程度——Level of Service牌照对号法——License Number Matching Method跟车测速——Car Following Method浮动车测速法——Moving Observer Speed Method通行能力查询拜访——Capacity Studies饱和流量——Saturation Flow第四章泊松分布——Poisson Distribution交通特性的统计分布——Statistical Distribution of Traffic Characteristics 驾驶员处置信息的特性Driver Information Processing Characteristics跟车理论——Car Following Theory交通流模拟——Simulation of Traffic Flow间隔分布——Interval Distribution二项分布——Binomial Distribution拟合——Fitting移位负指数分布——Shifted Exponential Distribution排队论——Queuing Theory运筹学——Operations Research加速骚扰——Acceleration Noise泊车波——Stopping Wave起动波——Starting Wave第五章城市交通规划——Urban Traffic Planning地盘操纵——Land-Use可达性——Accessibility起讫点查询拜访——Origin –Destination Survey出行端点——Trip End期望线——Desire Line主流倾向线——Major Directional Desire Line查询拜访区境界线——Cordon Line分隔查核线——Screen Line样本量——Sample Size出行发生——Trip Generation出行发生——Trip Production出行吸引——Trip Attraction发生率法——Generation Rate Method回归发生模型——Regression Generation Model类型发生模型——Category Generation Model出行分布——Trip Distribution此刻型式法——Present Pattern Method重力模型法——Gravity Model Method行程时间模型——Travel Time Model彼此影响模型——Interactive Model分布系数模型——Distribution Factor Model交通方式划分——Model Split , Mode Choice转移曲线——Diversion Curve交通量分配——Traffic Assignment最短路径分配〔全有全无〕Shortest Path Assignment(All-or-Nothing)多路线概率分配Probabilistic Multi-Route Assignment线权——Link Weight点权——Point Weight费用——效益阐发——Cost –benefit Analysis现值法——Present Value Method第六章交通安然——Traffic Safety交通变乱——Traffic Accident交通死亡变乱率——Traffic Fatal-Accident Rate交通法规——Traffic Law多发变乱地址——High accident Location交通条例——Traffic Regulation交通监视——Traffic Surveillance变乱陈述——Accident Report抵触触犯形式——Collision Manner财富损掉——Property Damage变乱档案——Accident File变乱报表——Accident Inventory固定目标——Fixed Object变乱率——Accident Ratelxy变乱数法——Accident Number Method 质量控制法——Quality Control Method 人行横道——Pedestrian Crosswalk 行人过街道信号——Pedestrian Crossing Beacon 人行天桥——Passenger Foot-Bridge 人行地道——Passenger Subway 栅栏——Gate 立体交叉——Underpass(Overpass) 标线——Marking 无信号控制交叉口——Uncontrolled Intersection 让路标识表记标帜——Yield Sign 泊车标识表记标帜——Stop Sign 渠化交通——Channelization traffic 单向交通——One-Way 禁止转弯——No Turn Regulation 禁止进入——No-Entry 禁止超车——Prohibitory Overtaking 禁止泊车——Prohibitory Parking 禁止通行——Road Closed 安然带——Life Belt 第七章交通控制与办理——Traffic Control and Management 交通信号——Traffic Signal 单点按时信号——Isolated Pre-timed Signal 信号相位——Signal Phase 周期长度——Cycle Length 绿信比——Split 优先控制——Priority Control 耽搁——Delay 流量比——Flow Ratio 有效绿灯时间——Effective Green Time 损掉时间——Loss Time 绿灯间隔时间——Intergreen Interval 信号配时——Signal Timing (or Signal Setting) 交通感应信号——Traffic Actuated Signal 城市交通控制系统——Urban Traffic Control System 联动控制——Coordinated Control 区域控制——Area Control 时差——Offset同时联动控制——Simultaneous Coordinated Control交变联动控制——Alternate Coordinated Control绿波带——Green Wave持续通行联动控制——Progressive Coordinated Control中心控制器——Master Controller局部控制器——Local——Controller实时——Real Time联机——On-line脱机——Off-line登山法——Hill-Climbing“小型高效〞区域控制系统——Compact Urban Traffic Control System道路控制系统——Corridor Control System交通仿真——Traffic Simulation时间扫描法——Time Scanning事件扫描法——Event ScanningTags：交通控制英语词汇ITS不雅察[本日：3 本周：4 本月：4 总数：137 ] [返回上一页] [打印] 0好的评价如果您觉得此文库好，就请您0%(0)差的评价如果您觉得此文库差，就请您100%(1)出格声明：本站文章版权归文章原始作者所有，转载文章必需先获得作者同意，请务必注明出处和原始作者。

管理预测与决策课程设计计算机据俱乐部仓库(CCW)问题

课程：管理预测与决策课程设计题目：计算机据俱乐部仓库（CCW）问题姓名：韩喜娟学号：09611204专业：计算机信息管理时间：2011年10月26号目录一、计算机俱乐部仓库的机制二、计算机俱乐部仓库面临的问题三、解决问题的预测模型1、25%规则2、季节调整后的上期值预测3、平均值预测方法4、移动平均预测法5、指数平滑预测法6、线性回归因果预测四、小结一、计算机俱乐部仓库的机制计算机俱乐部仓库（Computer Club Warehouse）通过与客户电话下单确定价格（以及网上和传真下单）的方式销售各种计算机产品。

其产品包括台式计算机、笔记本电脑、外围设备、附属硬件、备用品、软件及与电脑相关的家具。

公司每年几次将产品目录寄给用户及大量的未来客户，还通过电脑杂志发行微型目录。

这些目录明确的告知用户使用800免费电话下单。

这些电话被接入公司的呼叫中心。

二、计算机俱乐部仓库面临的问题呼叫中心从不关闭。

在繁忙的时段，它被大量的代理人挤满她们唯一的工作是通过电话接受并处理顾客订单。

新的代理人在开始工作前接受为期一周的培训，这项培训将重点放在如何高效、周到的处理顶单上。

一个代理人处理每个电话的平均期望时间不超过5分钟。

纪录被保留下来，没有在试用期达到目标的代理人将不再续聘。

尽管代理人的收入不低，工作带来的厌倦及时间压力造成了相当高的人员流动率。

呼叫中心为了接入电话提供了大量的电话线路，如果在电话到来时代理人正忙，电话会进入等待队列。

如果所有的线路都在使用，电话会响起忙音。

尽管一些遇到忙音或等待时间过长而挂断电话的用户会再次拨打电话直至拨通，但是许多客户并不会这样做的。

因此拥有足够的值班代理人来使这种情况出现的次数最少是很重要的。

另一方面，由于代理人的劳动成本过高CCW试图避免有过多的代理人工作，造成他们大量的闲暇时间。

于是获取代理人需求的预测成了公司的当务之急。

一、解决问题的预测模型1、25%规则除了圣诞节期间业务猛增之外，其他时间的业务相对稳定，于是有：二季度预测量=一季度呼叫量三季度预测量=二季度呼叫量四季度预测量=1.25（三季度呼叫量）MAD是预测误差的简称：MAD=预测误差之和/预测次数MSE是预测误差平方的均值：MSE=预测误差的平方和/预测次数其中MAD表示绝对差异。

最小生成树算法在旅行商问题中的应用_李萍（1）

最⼩⽣成树算法在旅⾏商问题中的应⽤_李萍（1）* 收稿⽇期:2011-10-06,修回⽇期:2011-11-29**　李　萍,⼥,1975年⽣,研究⽣,讲师,研究⽅向:数据挖掘。

⽂章编号:1003-5850(2012)01-0062-02最⼩⽣成树算法在旅⾏商问题中的应⽤李　萍,王春红,王⽂霞,任姚鹏(运城学院计算机科学与技术系,⼭西　运城　044000)摘　要:如何在n 个顶点之间的1/2(n -1)!巡回路径中选择距离最短的,这是⼀个典型的组合优化问题,也是解决旅⾏商问题的根本。

在最⼩⽣成树的基本思想上进⾏了改进,成功地解决了旅⾏商问题。

关键词:最⼩⽣成树,旅⾏商问题,回路,连通图中图分类号:T P301.6 ⽂献标识码:AApplication of Minimum Cost Spanning Tree toTraveling Salesman ProblemLI Ping ,W AN G Chun-hong ,W AN G W en-xia,REN Yao-peng(Y uncheng Univ ersity Department of Computer Science and T echnology ,Yuncheng 044000,China )Abstract :It is a ty pical co mbinatorial optimizatio n problem a nd the fundament of solving trav eling salesman problem that how to find the shor test loo p fro m half o f facto rial of n -1betw een n vertex .In the tex t th ro ugh im po rting the basic idea of minimum cost spanning tree ,the trav eling salesma n problem is solv ed successfully .Key words :minimum cost spanning tree ,trav eling salesman problem ,loo p ,co nnected g raph旅⾏商问题(Trav eling Salesman Problem,简称为TSP)就是给定n 个城市,⼀个旅⾏商从其中的某⼀城市出发,不重复地⾛完其余n -1个城市并回到原点,在所有可能中求出路径长度最短的⼀条巡回路径。

1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。

• The figures on the foregoing slide only show domestic sales figures China, according to the 2008 statistics the elevator im- and export market are as follows: - EXPORT: 37,500 sets 15.5 % of total production domestic + export - IMPORT: 1,764 sets 0.9 % of total production domestic + export • While the export business of Chinese made elevators can be assumed to increase in the future, import business had been on a minimum level 2008 already, was probably even lower in 2010 and will trend to 0 in the future. • Although some major components of a Chinese elevator like brake and encoder of the traction machine as well as other high tech parts of the lift are still partly imported, the Chinese elevator industry is able to cover more than 99 % of the local demand. • The Chinese government planning of „building housings for the people to live in“ is still running and supported. The state plan foresees the formation of further 10-20 large urban agglomerations like the Yangtze and the Pearl River delta in the next 10-20 years. • Therefore – naturally depending on the worldwide economic and political development – a continous growth in the Chinese real estate and thus elevator market can be expected in the next years.
2. Market Segments:
Residential building: Commercial (office) building: Hotel / Hospital: Public / shopping building:
Share 70 15 10 5 % % % %
3. Niche / Special Applications:
Ziehl-Abegg China
Elevator Market Evaluation + Business Plan China Drive Business
Shanghai, 2011-01-15 – 2011-07-15
William Wang
Sales Manager
Daniel Haitzler
6
Elevator Market Evaluation China Drive Business
01.08.2011
Version 2.0 – Oktober 2007
3. Market Players and Volume (2)
Business Share MRL drives Market Players in 2010:
3. Market Players and Volume (1)
China Market Sold Elevator Sets:
Elevator Company SMEC Hitachi XiZi Otis KONE OTIS Tianjin Thyssen Krupp Schindler Shanghai Yungtay Giant KONE Toshiba CANNY Sigma Fujitec Langfang Suzhou Express SJEC Eidenberg Xi Ji Shenlong Shenyang BLT Others TOTAL 2009 28,800 27,000 21,600 19,800 10,350 9,900 9,000 9,450 9,000 8,100 7,200 5,400 8,550 6,400 5,400 4,500 4,320 3,935 3,240 27,735 229,680 2010 35,150 32,400 26,600 23,760 12,420 11,880 10,800 15,000 10,800 9,720 8,640 7,000 7,350 7,200 7,000 5,400 5,184 4,722 4,700 41,074 287,100 Dev. % 22% 20% 23% 20% 20% 20% 20% 59% 20% 20% 20% 30% -14% 17% 30% 20% 20% 20% 45% 48% 25%
MRL applications: Super high rise 3-4 m/s Villa / Home lift:
Share 23 % 1% 2%
5
Elevator Market Evaluation China Drive Business
01.08.2011
Version 2.0 – Oktober 2007
Project Manager
Ziehl-Abegg Mechanical and Electrical Equipment (Shanghai) Co., Ltd. No.65, Hong Mu Dan Road, Xinbang Town, Songjiang District Shanghai, 201605 Tel. +86 21 57893291 Fax +86 21 57893932 daniel.haitzler@
2
Elevator Market Evaluation China Drive Business
01.08.2011
1. Facts and Figures about the real estate and elevator market in China (1)
Version 2.0 – Oktober 2007
• In the last 15 % years China‘s real estate market has become the country‘s pillar industry with an average annually investment increase of 24.1 % which is much higher than the average annually GDP increase of 13.2 %. • Statistics show that the elevator market development is closely linked to the development of the real estate market with a correlation of the factor 0.8. • Within 30 years China‘s elevator industry has increased from the annual output of around 2,200 sets with a retain number of 10,000 sets at the beginning of the 80‘s to the annual output of nearly 300,000 sets with a retain number of more than 1,200,000 in 2010:
4
Elevator Market Evaluation China Drive Business
01.08.2011
2. Market Definition and Categorisation for Ziehl-Abegg drives (1)
Version 2.0 – Oktober 2007
The existing and potential market for the ZA drive business can be categorised in: 1. Product range:
Business share: Lifting weight: Suspension: Speed: Passenger Lift 95 % 320 – 3,200 kg 2:1 1 – 2.5 m/s Goods Lift 5% 1,500 – 6,500 kg 2:1 & 4:1 0.5 – 1 m/s
3
Elevator Market Evaluation China Drive Business
01.08.2011
1. Facts and Figures about the real estate and elevator market in China (2)
Version 2.0 – Oktober 2007
Hitachi 11.3% XiZi Otis 9.3%
KONETianjin 4.3% Thyssen Krupp 4.1% Schindler 3.8% Shanghai Yungtay 5.2% Others 31.3% Giant KONE 3.8% Toshiba 3.4% CANNY 3.0%