CN105157712A - Vehicle routing planning method and planning system - Google Patents

Abstract

The invention applies to the field of vehicle routing planning and provides a vehicle routing planning method and planning system. The method comprises steps as follows: an initial solution of a vehicle routing scheme is established on the basis of a minimum cost insertion method based on user demands; initial parameters of the initial solution are set; whether the initial solution meets the end condition of a stage is judged, the optimal solution is updated according to a judgement result, corresponding initial parameters of the optimal solution are updated, and the optimal solution is updated according to the judgement result as follows: if the initial solution doesn't meet the end condition, the optimal solution is acquired with a variable neighborhood search algorithm, and if the initial solution meets the end condition, the initial solution is the optimal solution, and restarting is performed. Through implementation of the embodiment of the invention, the routing planning procedure can be simplified.

Description

A kind of planing method of vehicle route and planning system

Technical field

The invention belongs to vehicle path planning field, particularly relate to a kind of planing method and planning system of vehicle route.

Background technology

Along with the arrival of economic globalization, manufacturing industry, retail trade and ecommerce have expedited the emergence of the fast development of modern logistics.The widespread use of logistics information technology, the construction of management information system, Logistics Information Platform, logistics distribution system advances fast.Vehicle path planning is the important step of urban logistics distribution, is also basis and the important module of logistics distribution information system foundation.

Logistics distribution center is often faced with the constraint of various goods stock resource, an actual constraint has limited various, the vehicle number of often kind of vehicle fixes the capacity limit with the vehicle of every type, this problem is the Multi-types vehicle routine problem (HeterogeneousFixedFleetVehicleRoutingProblem typically with fixed vehicle number, HFFVRP), and the arrangement vehicle route of how economical rationality in logistics distribution, most important for development low cost, high efficiency material flow industry.

Tabu search algorithm (TabuSearch is called for short TS) is the one overall situation Stepwise optimization algorithm proposed in 1986 by Glover, simulates the mental process of the mankind.Tabu search algorithm have employed storage organization and taboo rule in order to avoid roundabout search, and introduces special pardon criterion and absolve by some excellent conditions avoided, thus ensure that the variation of search, is conducive to realizing global optimum.Tabu search algorithm has been applied to many field of engineering technology, comprises production scheduling, machine learning, computer network Route Selection, circuit design and neural network etc.But this algorithm is parameter-dependent, lack dirigibility, variation search and concentrate that the balance of searching for is poor, arithmetic speed is slower.

Variable neighborhood search algorithm (VariableNeighborhoodSearch is called for short VNS) is a kind of meta-heuristic algorithm proposed first in 1997 by Hansen and Mladenovic.It constantly can adjust neighbour structure collection to expand hunting zone in search procedure, produces locally optimal solution, and then again expands hunting zone to explore new locally optimal solution by this locally optimal solution by systematically changing neighbour structure collection.The basic operation of variable neighborhood search algorithm comprises the selection of search order between the structure of neighbour structure collection, neighbour structure, the design of neighbour structure search mechanisms, Local Search design and stopping criterion design etc.Variable neighborhood search algorithm simple structure, has good local search ability and versatility, but variable neighborhood search algorithm is based on neighbour structure collection, is easily absorbed in locally optimal solution.

Summary of the invention

Given this, the invention provides a kind of planing method and planning system of vehicle route, to solve the technical matters of prior art path planning complexity.

The embodiment of the present invention is achieved in that a kind of planing method of vehicle route, said method comprising the steps of:

Minimum cost insertion method based on user's request builds the initial solution of vehicle route scheme;

The initial parameter of described initial solution is set;

Judge whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprise: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

The embodiment of the present invention also provides a kind of planning system of vehicle route, and described system comprises:

Initial solution construction unit, for building the initial solution of vehicle route scheme based on the minimum cost insertion method of user's request;

Initial parameter setting unit, for arranging the initial parameter of the initial solution that described initial solution construction unit builds;

Optimum solution acquiring unit, for judging whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprising: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

The embodiment of the present invention, stage of algorithm and cycle are divided, accommodation is carried out according to the scale of problem solving and network structure, improve the dirigibility of vehicle path planning, the Regulation mechanism of auto-adaptive parameter is incorporated in tabu search algorithm, make the restriction of Tabu Length, neighborhood and neighbor operator perform the parameters such as frequency automatically to regulate according to the change of solution space, effectively balance the centrality in algorithm search and diversity; Construct sequentially variable neighborhood search algorithm and nested type variable neighborhood search algorithm and be respectively used to the search of neighbour structure in algorithm major cycle and the further improvement to best feasible solution, improve the global optimizing ability of vehicle route, shorten search time, make the vehicle route arrangement searched more save logistics transportation cost, and meet the real-time response requirement in application.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

Fig. 1 is the process flow diagram of the vehicle path planning method that the embodiment of the present invention provides;

Fig. 2 is touring Distribution path put by vehicle that the embodiment of the present invention provides schematic diagram different client;

Fig. 3 is the schematic diagram of the implementation of the order variable neighborhood search algorithm that the embodiment of the present invention provides;

Fig. 4 is the schematic diagram of neighborhood search operator between 4 kinds of paths providing of the embodiment of the present invention;

Fig. 5 is the schematic diagram of neighborhood search operator in 2 kinds of paths providing of the embodiment of the present invention;

Fig. 6 is the schematic diagram of the implementation of the nested type variable neighborhood search algorithm that the embodiment of the present invention provides;

Fig. 7 is the structural drawing of the vehicle path planning system that the embodiment of the present invention provides.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

In order to technical solutions according to the invention are described, be described below by specific embodiment.

Embodiment one

Be illustrated in figure 1 the process flow diagram of the vehicle path planning method that the embodiment of the present invention provides, said method comprising the steps of:

Step S101, the minimum cost insertion method based on user's request builds the initial solution of vehicle route scheme.

In embodiments of the present invention, the minimum cost insertion method based on user's request is first adopted to build the initial solution of vehicle route scheme.The step of the initial solution of described structure vehicle route scheme, comprising:

1, metric solution vector encoding method encodes solution vector is adopted.

In embodiments of the present invention, solution vector x is expressed as π _jsequence j=1,2 ..., m, the wherein sum of various types of vehicles that has for home-delivery center of m, π _jfor the set that node each in vehicle route sorts successively, vehicle route is expressed as the sequence of touring node successively, and first node and last node of sequence are fixed as home-delivery center's point respectively.As in Fig. 2, the solution vector of vehicle delivery process can be expressed as: π ₁=0,5,2, Isosorbide-5-Nitrae, 8,0}, π ₂={ 0,6,3,7,9,11,0}, π ₃={ 0,10,14,13,12,0}.

2, Stochastic choice m client's point, is initialised to described solution vector by described m client's point and produces m bar Distribution path R _i(i=1,2 ..., m).

In embodiments of the present invention, i is put for each client undetermined, if the path total cost inserting this client point is minimum and demand that is that increase does not exceed the residual capacity in path, then select the feasible path R with greatest residual capacity _k, calculate and insert cost and client is put i and be inserted into the path R meeting cost minimization _krelevant position.

Described client puts i at path R _kon insertion cost be calculated as follows:

$S_i^k=(d_{i_1,i}+d_{i,i_2}-d_{i_1,i_2})\×v_k---\left(1\right)$

Wherein, i ₁and i ₂continue before insertion position node and descendant node respectively, v _kfor the variable cost of vehicle k.

3, when there is the client undetermined not inserting path and putting, j is put for arbitrary client undetermined ₁, select the path R with greatest residual capacity _k1, by j ₁be inserted into path R _k1on make to insert the position of cost minimization.

In embodiments of the present invention, in order to strengthen the diversity of algorithm, algorithm allows to occur infeasible solution, and what namely can produce vehicle capacity runs counter to operation.To algorithm search to infeasible solutions give certain rejection penalty according to the amount of running counter to, use punishment objective function f (x ') carry out evaluate candidate solution x ', f (x ') comprises former target expense and rejection penalty two parts, and computing method are:

$f\left(x^{\′}\right)=\underset{k=1}{\overset{m}{\Σ}}(v_k+{\mathrm{Pl}}_k)---\left(2\right)$

Wherein, P is penalty term, l _kfor the load-carrying amount of exceeding of vehicle k.

Algorithm also adopts long-term memory, objective function f simultaneously ₁(x ') is:

$\left\{\begin{array}{cc}{f}_1\left(x^{\′}\right)=f\left(x^{\′}\right)& \⇐f\left(x^{\′}\right)\≤f\left(x\right)\\ f_1\left(x^{\′}\right)=f\left(x^{\′}\right)+\mathrm{\β}D\sqrt{rn}\mathrm{\ρ}/K& \⇐f\left(x^{\′}\right)>f\left(x\right)\end{array}\right.---\left(3\right)$

Wherein x is current solution, and β is zooming parameter, and D is the maximum absolute difference that f (x) produces in two continuous print iteration steps, and ρ is that client puts the number of times be moved, and K is algorithm iteration number of times up to the present.

Step S102, arranges the initial parameter of described initial solution.

In embodiments of the present invention, the step arranging the initial parameter of initial solution comprises:

1, the initial parameter in stage is set and the initial parameter in cycle is set.

Wherein, the initial parameter in stage comprises: the iterations in a stage is restricted to K _l, zooming parameter β, Tabu Length θ, neighborhood restriction δ.The initial parameter in cycle comprises: the controling parameters σ that feasible and infeasible solution space boundary shakes, and maximum permission solution does not improve number of times K _bL, neighborhood search operator performs frequency F _k, k=1,2 ..., 6.

2, the taboo list of the initial parameter in described stage and the taboo list of the initial parameter in described cycle is emptied.

Step S103, judge whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprising: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

In embodiments of the present invention, after the initial parameter being provided with initial solution, judge whether initial solution meets the end condition in stage, if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then initial solution be set to optimum solution and restart, wherein determination methods uses prior art, does not repeat at this.Described variable neighborhood search algorithm obtains optimum solution, comprising:

1, execution sequence variable neighborhood search algorithm, the result of calculation according to described variable neighborhood search algorithm obtains optimum solution.

In embodiments of the present invention, described order variable neighborhood search algorithm, flow process as shown in Figure 3, performs neighbor operator N successively according to given order _k, k=1,2 ..., 6, if algorithm is absorbed in local optimum, will by execution N _kjump to and perform N _k+1, otherwise will from first operator N ₁the operator of all uses starts to re-execute, until all cannot improve current optimum solution.

To each operator N _kalso define one and perform frequency F _k, k=1,2 ..., 6, as F ₃=10, F ₅=∞ then represents that every iteration performs a N 10 times ₃, do not perform operator N ₅.The execution frequency of operator will adjust according to solution procedure, thus improve dirigibility and the diversity of neighborhood search.

Described neighbor operator to comprise between path search operators two class in search operators and path, and its concrete action mode is respectively as shown in Figure 4 and Figure 5, as follows:

(1) N ₁: 1-0 exchanges.

1-0 exchange is that the client between a kind of path puts moving method, and move also referred to as Swap/Shift, client o'clock is inserted into another paths from a paths by it.

(2) N ₂: 1-1 exchanges.

1-1 exchanges and belongs to mobile operator between a kind of path, is also that a kind of Swap/Shift moves, and realizes the client's point in a path to put mutual exchange with the client in another paths.

(3) N ₃: 2-opt* exchanges.

It is search operators between a kind of path that 2-opt* exchanges, and be the expansion that 1-1 exchanges, the afterbody in two different paths exchanges by mutually.

(4) N ₄: Cross exchanges.

It is neighborhood search method between a kind of path that Cross exchanges.First the arc of two neighboring customer points in the neighboring customer point of two in one paths and an other paths is deleted by it, and then by they cross connections.

(5)N ₅：Exchange。

Exchange is the neighborhood search operator in a kind of path, is the 1-1 switching method in path.It realizes the exchange operation of two different client's points in a paths.

(6)N ₆：Reverse。

Reverse belongs to neighborhood transform method in path.This operator selects the parton road of a paths, is then reversed on this sub-road.

In order to reduce the number of times of potential movement in each iteration, the saving algrithm execution time, guarantee to search better solution simultaneously, the restriction of δ neighborhood search is performed to each neighbor operator, here the δ neighborhood definition of node i is the set of δ the node nearest apart from this node, and namely operator only operates the δ neighborhood of node.

2, by nested type variable neighborhood search algorithm, described optimum solution is improved, and upgrade initial parameter corresponding to described optimum solution.

In embodiments of the present invention, described nested type variable neighborhood search algorithm, performs flow process as shown in Figure 6, is the neighbor operator N to current solution x _r(x) (r=1,2 ..., 6) and each candidate solution x ' execution that produces becomes neighborhood search computing, and selecting of operator is random to increase the diversity of searching for, until N _rx the solution in () cannot be improved till.

It is pointed out that if reach the centre position in cycle, namely then adjust θ, δ and F _k(k=1,2 ..., 6).

The embodiment of the present invention, stage of algorithm and cycle are divided, accommodation is carried out according to the scale of problem solving and network structure, improve the dirigibility of vehicle path planning, the Regulation mechanism of auto-adaptive parameter is incorporated in tabu search algorithm, make the restriction of Tabu Length, neighborhood and neighbor operator perform the parameters such as frequency automatically to regulate according to the change of solution space, effectively balance the centrality in algorithm search and diversity; Construct sequentially variable neighborhood search algorithm and nested type variable neighborhood search algorithm and be respectively used to the search of neighbour structure in algorithm major cycle and the further improvement to best feasible solution, improve the global optimizing ability of vehicle route, shorten search time, make the vehicle route arrangement searched more save logistics transportation cost, and meet the real-time response requirement in application.

Embodiment two

Be illustrated in figure 7 the structural drawing of the vehicle path planning system that the embodiment of the present invention provides, for convenience of explanation, the part relevant to the embodiment of the present invention be only shown, comprise:

Initial solution construction unit 701, for building the initial solution of vehicle route scheme based on the minimum cost insertion method of user's request.

In embodiments of the present invention, the minimum cost insertion method based on user's request is first adopted to build the initial solution of vehicle route scheme.Described initial solution construction unit 701 build initial solution, is specially:

1, metric solution vector encoding method encodes solution vector is adopted.

In embodiments of the present invention, solution vector x is expressed as π _jsequence j=1,2 ..., m, the wherein sum of various types of vehicles that has for home-delivery center of m, π _jfor the set that node each in vehicle route sorts successively, vehicle route is expressed as the sequence of touring node successively, and first node and last node of sequence are fixed as home-delivery center's point respectively.As in Fig. 2, the solution vector of vehicle delivery process can be expressed as: π ₁=0,5,2, Isosorbide-5-Nitrae, 8,0}, π ₂={ 0,6,3,7,9,11,0}, π ₃={ 0,10,14,13,12,0}.

2, Stochastic choice m client's point, is initialised to described solution vector by described m client's point and produces m bar Distribution path R _i(i=1,2 ..., m).

In embodiments of the present invention, i is put for each client undetermined, if the path total cost inserting this client point is minimum and demand that is that increase does not exceed the residual capacity in path, then select the feasible path R with greatest residual capacity _k, calculate and insert cost and client is put i and be inserted into the path R meeting cost minimization _krelevant position.

Described client puts i at path R _kon insertion cost be calculated as follows:

$s_i^k=(d_{i_1,i}+d_{i,i_2}-d_{i_1,i_2})\×v_k---\left(1\right)$

Wherein, i ₁and i ₂continue before insertion position node and descendant node respectively, v _kfor the variable cost of vehicle k.

3, when there is the client undetermined not inserting path and putting, j is put for arbitrary client undetermined ₁, select the path with greatest residual capacity by j ₁be inserted into path on make to insert the position of cost minimization.

In embodiments of the present invention, in order to strengthen the diversity of algorithm, algorithm allows to occur infeasible solution, and what namely can produce vehicle capacity runs counter to operation.To algorithm search to infeasible solutions give certain rejection penalty according to the amount of running counter to, use punishment objective function f (x ') carry out evaluate candidate solution x ', f (x ') comprises former target expense and rejection penalty two parts, and computing method are:

$f\left(x^{\′}\right)=\underset{k=1}{\overset{m}{\Σ}}(v_k+{\mathrm{Pl}}_k)---\left(2\right)$

Wherein, P is penalty term, l _kfor the load-carrying amount of exceeding of vehicle k.

Algorithm also adopts long-term memory, objective function f simultaneously ₁(x ') is:

$\left\{\begin{array}{cc}{f}_1\left(x^{\′}\right)=f\left(x^{\′}\right)& \⇐f\left(x^{\′}\right)\≤f\left(x\right)\\ f_1\left(x^{\′}\right)=f\left(x^{\′}\right)+\mathrm{\β}D\sqrt{rn}\mathrm{\ρ}/K& \⇐f\left(x^{\′}\right)>f\left(x\right)\end{array}\right.---\left(3\right)$

Wherein x is current solution, and β is zooming parameter, and D is the maximum absolute difference that f (x) produces in two continuous print iteration steps, and ρ is that client puts the number of times be moved, and K is algorithm iteration number of times up to the present.

Initial parameter setting unit 702, for arranging the initial parameter of the initial solution that described initial solution construction unit 701 builds.

In embodiments of the present invention, initial parameter setting unit 702 arranges initial parameter, is specially:

1, the initial parameter in stage is set and the initial parameter in cycle is set.

Wherein, the initial parameter in stage comprises: the iterations in a stage is restricted to K _l, zooming parameter β, Tabu Length θ, neighborhood restriction δ.The initial parameter in cycle comprises: the controling parameters σ that feasible and infeasible solution space boundary shakes, and maximum permission solution does not improve number of times K _bL, neighborhood search operator performs frequency F _k, k=1,2 ..., 6.

2, the taboo list of the initial parameter in described stage and the taboo list of the initial parameter in described cycle is emptied.

Optimum solution acquiring unit 703, for judging whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprising: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

In embodiments of the present invention, after the initial parameter being provided with initial solution, judge whether initial solution meets the end condition in stage, if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then initial solution be set to optimum solution and restart, wherein determination methods uses prior art, does not repeat at this.Described optimum solution acquiring unit 703 obtains optimum solution, is specially:

1, execution sequence variable neighborhood search algorithm, the result of calculation according to described variable neighborhood search algorithm obtains optimum solution.

In embodiments of the present invention, described order variable neighborhood search algorithm, flow process as shown in Figure 3, performs neighbor operator N successively according to given order _k, k=1,2 ..., 6, if algorithm is absorbed in local optimum, will by execution N _kjump to and perform N _k+1, otherwise will from first operator N ₁the operator of all uses starts to re-execute, until all cannot improve current optimum solution.

To each operator N _kalso define one and perform frequency F _k, k=1,2 ..., 6, as F ₃=10, F ₅=∞ then represents that every iteration performs a N 10 times ₃, do not perform operator N ₅.The execution frequency of operator will adjust according to solution procedure, thus improve dirigibility and the diversity of neighborhood search.

Described neighbor operator to comprise between path search operators two class in search operators and path, and its concrete action mode is respectively as shown in Figure 4 and Figure 5, as follows:

(1) N ₁: 1-0 exchanges.

1-0 exchange is that the client between a kind of path puts moving method, and move also referred to as Swap/Shift, client o'clock is inserted into another paths from a paths by it.

(2) N ₂: 1-1 exchanges.

1-1 exchanges and belongs to mobile operator between a kind of path, is also that a kind of Swap/Shift moves, and realizes the client's point in a path to put mutual exchange with the client in another paths.

(3) N ₃: 2-opt* exchanges.

It is search operators between a kind of path that 2-opt* exchanges, and be the expansion that 1-1 exchanges, the afterbody in two different paths exchanges by mutually.

(4) N ₄: Cross exchanges.

It is neighborhood search method between a kind of path that Cross exchanges.First the arc of two neighboring customer points in the neighboring customer point of two in one paths and an other paths is deleted by it, and then by they cross connections.

(5)N ₅：Exchange。

Exchange is the neighborhood search operator in a kind of path, is the 1-1 switching method in path.It realizes the exchange operation of two different client's points in a paths.

(6)N ₆：Reverse。

Reverse belongs to neighborhood transform method in path.This operator selects the parton road of a paths, is then reversed on this sub-road.

In order to reduce the number of times of potential movement in each iteration, the saving algrithm execution time, guarantee to search better solution simultaneously, the restriction of δ neighborhood search is performed to each neighbor operator, here the δ neighborhood definition of node i is the set of δ the node nearest apart from this node, and namely operator only operates the δ neighborhood of node.

2, by nested type variable neighborhood search algorithm, described optimum solution is improved, and upgrade initial parameter corresponding to described optimum solution.

In embodiments of the present invention, described nested type variable neighborhood search algorithm, performs flow process as shown in Figure 6, is the neighbor operator N to current solution x _r(x) (r=1,2 ..., 6) and each candidate solution x ' execution that produces becomes neighborhood search computing, and selecting of operator is random to increase the diversity of searching for, until N _rx the solution in () cannot be improved till.

It is pointed out that if reach the centre position in cycle, namely then adjust θ, δ and F _k(k=1,2 ..., 6).

The embodiment of the present invention, stage of algorithm and cycle are divided, accommodation is carried out according to the scale of problem solving and network structure, improve the dirigibility of vehicle path planning, the Regulation mechanism of auto-adaptive parameter is incorporated in tabu search algorithm, make the restriction of Tabu Length, neighborhood and neighbor operator perform the parameters such as frequency automatically to regulate according to the change of solution space, effectively balance the centrality in algorithm search and diversity; Construct sequentially variable neighborhood search algorithm and nested type variable neighborhood search algorithm and be respectively used to the search of neighbour structure in algorithm major cycle and the further improvement to best feasible solution, improve the global optimizing ability of vehicle route, shorten search time, make the vehicle route arrangement searched more save logistics transportation cost, and meet the real-time response requirement in application.

One of ordinary skill in the art will appreciate that the unit included by above-described embodiment two is carry out dividing according to function logic, but be not limited to above-mentioned division, as long as corresponding function can be realized; In addition, the concrete title of each functional unit, also just for the ease of mutual differentiation, is not limited to protection scope of the present invention.

Those of ordinary skill in the art it is also understood that, the all or part of step realized in above-described embodiment method is that the hardware that can carry out instruction relevant by program has come, described program can be stored in a computer read/write memory medium, described storage medium, comprises ROM/RAM, disk, CD etc.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (8)

1. a planing method for vehicle route, is characterized in that, said method comprising the steps of:

Minimum cost insertion method based on user's request builds the initial solution of vehicle route scheme;

The initial parameter of described initial solution is set;

Judge whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprise: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

2. the method for claim 1, is characterized in that, the step of the initial solution of described structure vehicle route scheme, comprising:

Adopt metric solution vector encoding method encodes solution vector;

Stochastic choice m client's point, is initialised to described solution vector by described m client's point and produces m bar Distribution path R _i(i=1,2 ..., m);

When there is the client undetermined not inserting path and putting, j is put for arbitrary client undetermined ₁, select the path R with greatest residual capacity _k1, by j ₁be inserted into path R _k1on make to insert the position of cost minimization.

3. the method for claim 1, is characterized in that, the described step arranging the initial parameter of initial solution comprises:

The initial parameter in stage is set and the initial parameter in cycle is set;

Empty the taboo list of the initial parameter in described stage and the taboo list of the initial parameter in described cycle.

4. the method as described in any one of claim 1 ~ 2, is characterized in that, described variable neighborhood search algorithm of crossing obtains optimum solution, comprising:

Execution sequence variable neighborhood search algorithm, the result of calculation according to described variable neighborhood search algorithm obtains optimum solution;

By nested type variable neighborhood search algorithm, described optimum solution is improved, and upgrade initial parameter corresponding to described optimum solution.

5. a planning system for vehicle route, is characterized in that, described system comprises:

Initial solution construction unit, for building the initial solution of vehicle route scheme based on the minimum cost insertion method of user's request;

Initial parameter setting unit, for arranging the initial parameter of the initial solution that described initial solution construction unit builds;

Optimum solution acquiring unit, for judging whether described initial solution meets the end condition in stage, optimum solution is upgraded according to judged result, and upgrade initial parameter corresponding to described optimum solution, described according to judged result renewal optimum solution, comprising: if do not met, then obtain optimum solution by variable neighborhood search algorithm, if met, then described initial solution be set to optimum solution and restart.

6. system as claimed in claim 5, it is characterized in that, described initial solution construction unit build initial solution, is specially:

Adopt metric solution vector encoding method encodes solution vector;

Stochastic choice m client's point, is initialised to described solution vector by described m client's point and produces m bar Distribution path R _i(i=1,2 ..., m);

When there is the client undetermined not inserting path and putting, j is put for arbitrary client undetermined ₁, select the path R with greatest residual capacity _k1, by j ₁be inserted into path R _k1on make to insert the position of cost minimization.

7. system as claimed in claim 5, it is characterized in that, initial parameter setting unit arranges initial parameter, is specially:

The initial parameter in stage is set and the initial parameter in cycle is set;

Empty the taboo list of the initial parameter in described stage and the taboo list of the initial parameter in described cycle.

8. system as claimed in claim 5, is characterized in that, described optimum solution acquiring unit obtains optimum solution, is specially:

Execution sequence variable neighborhood search algorithm, the result of calculation according to described variable neighborhood search algorithm obtains optimum solution;

By nested type variable neighborhood search algorithm, described optimum solution is improved, and upgrade initial parameter corresponding to described optimum solution.