CN104992242A - Method for solving logistic transport vehicle routing problem with soft time windows - Google Patents

Abstract

The invention discloses a method for solving a logistic transport vehicle routing problem with soft time windows. According to the method, for the purpose of solving the problem of the logistics transport vehicle routing problem with the soft time windows on the basis of real-time traffic information, a time window punishment mechanism is employed and a mathematic model is established; and the model is solved by use of a self-adaptive chaotic ant colony algorithm, and the searching optimization capability of the algorithm is improved through self-adaptive updating of algorithm information elements and chaotic self-adaptive adjustment of algorithm parameters. According to the invention, the method better matches logistics distribution in realistic production life, the problem is solved by use of the self-adaptive chaotic ant colony algorithm, the optimization search capability is better, a search process is effectively prevented from partial optimum, the diversity of solutions and the global searching optimization capability are improved, the global updating strategy is improved, an elite strategy is introduced, and positive feedbacks of information elements released by high-quality ants are properly improved; and the upper limits and lower limits of the information elements and the information element increments are arranged so that overlarge differences of the information elements on a path are reduced, and the classic vehicle routing searching optimization problem is solved by use of the self-adaptive chaotic ant colony algorithm.

Description

A kind of method solving band weak rock mass transport truck routing problem

Technical field

The invention belongs to technical field of intelligent traffic, be specifically related to a kind of method solving band weak rock mass transport truck routing problem.

Background technology

In actual life, vehicle can be subject to the impact of the many factors such as such as traffic administration, traffic jam, driving restriction, peak period on and off duty under steam, thus result in each Link Travel Time in vehicle travel and also correspondingly change, cause vehicle different in the travel speed in each time period or single section, this will have an impact to whole logistics distribution process.Tsing-Hua University Xu Jie (Xu Jie in recent years, Jiang Yongheng, Huang Dexian. the Vehicle Routing Problems based on Real-time Traffic Information is studied [J]. computing machine and applied chemistry, 2009,26 (9): 1093-1096) have studied the VRP problem based on single home-delivery center of Real-time Traffic Information and the constraint of free window, Dalian University of Technology Lv Jun space (Lv Junyu. based on the optimization method [D] of real-time traffic prediction. Dalian University of Technology, 2013) predict based on Real-time Traffic Information, and have studied control and optimization method; Jilin University in Yao (in Yao. based on Used in Dynamic Traffic Assignment modeling and the realization [D] of traveler behavior. Changchun: Jilin University, 2014) carry out dynamic vehicles selection according to traveler for different destination behaviors, carry out modeling and realization according to Used in Dynamic Traffic Assignment.

Self-adaptation Chaos Ant Colony Optimization is that the ergodicity of chaos, randomness, initial value sensitivity and chaotic disturbance operator are introduced ant group algorithm, by Chaos Variable linear mapping to optimized variable interval, adjust automatically, can effectively avoid being absorbed in local optimum in search procedure, improve the diversity and global optimizing ability of separating, utilize this algorithm can solve routing problem better.But prior art is not enough to the research of goods' transportation routing problem, is not very identical with the logistics distribution in real productive life, Optimizing Search ability and global optimizing ability poor.

Summary of the invention

For the deficiencies in the prior art part, technical matters to be solved of the present invention is to provide a kind of method solving band weak rock mass transport truck routing problem, more identical with the logistics distribution in real productive life, self-adaptation Chaos Ant Colony Optimization is used to solve this problem, there is better Optimizing Search ability, effectively avoid search procedure to be absorbed in local optimum, improve the diversity and global optimizing ability of separating.

In order to solve the problems of the technologies described above, the present invention can be realized by following technical measures: a kind of method solving band weak rock mass transport truck routing problem, comprises the following steps:

S1, obtain band weak rock mass VRP problem basic parameter based on Real-time Traffic Information: the position of logistics distribution center, the geographic position of client, client cargo demand, the time windows constraints of client and the information of vehicle;

S2, in problem, client has weak rock mass requirement: the time window of deliver goods is [et _i, lt _i], distribution vehicle is et in the service time the earliest of dispensing point i _i, arriving distribution time is the latest lt _i, late or early to all producing rejection penalty; Waiting cost per hour is c ₁, deferred charges per hour is c ₂; Vehicle Speed is determined by different time sections, 1 dispatching cycle is divided into some discrete segment periods, vehicle is change in the travel speed of each period, the inner travel speed in each interval is certain, current time travel speed can only be known by dispatching center, future time travel speed and present speed same treatment;

S3, be defined as follows variable:

S4, set up the mathematical model based on the band weak rock mass transport truck routing problem of Real-time Traffic Information according to following constraint condition:

(1) 1 parking lot, N number of client, customer demand is determined;

(2) vehicle of the same race, undercapacity;

(3) each car has maximum dispensing distance restraint and load-carrying constraint;

(4) closed type vehicle route;

(5) time window of client is known, and weak rock mass limits, and the time window of each client requires known;

(6) target is solved: in urban dynamic traffic Information Network, find one group of optimum logistics distribution path serving all clients, reach distribution cost minimum;

The adaptive updates of S5, design ant group algorithm pheromones improves optimizing ability, the noise immunity of algorithm, and then avoids algorithm to be absorbed in local optimum, is embodied in:

S5-1, improve overall update strategy, introduce elitism strategy, Pheromone update is carried out in the path being greater than the average overall metric function value of current iteration to overall metric function value after each algorithm circulation, and Pheromone update is not carried out to the path being less than the average overall metric of current iteration, overall situation update rule is only for optimum solution path, and rule is as follows:

In above formula, Q represents pheromones intensity, and its value is constant, L _gbrepresent the current global optimum's path found;

The bound of S5-2, configuration information element and pheromones increment crosses big-difference, by suitable adjustment, by the pheromones intersity limitation on each paths built in interval [τ with what reduces pheromones on path _min, τ _max] in, when each lastest imformation element, pheromones increment is limited in interval [Δ τ simultaneously _min, Δ τ _max], according to above-mentioned improvement and optimization, total expression formula of pheromones adaptive updates is:

In (4) formula, pheromones increment

The chaos self-adaptative adjustment of S6, design ant group algorithm calculating parameter, improves algorithm to the optimizing ability of this mathematical model and counting yield; When the optimizing of ant group algorithm every generation iteration starts, α gets initial value is Arbitrary Digit between 1 to 5, and the initial value that ρ gets is the Arbitrary Digit between 0.1-0.5, then carries out chaos self-adaptative adjustment by as shown in the formula to heuristic factor α and pheromones volatility coefficient ρ:

S7, utilize self-adaptation Chaos Ant Colony Optimization to seek the optimum Distribution path of the band weak rock mass transport truck routing problem based on Real-time Traffic Information, concrete steps are as follows:

S7-1, initialization self-adaptation Chaos Ant Colony Optimization parameter: maximum iteration time NC _max, ant number m, ρ (0), α (0), Q, τ, Δ τ, iteration count NC=0;

S7-2, ant are from the position of home-delivery center;

S7-3, to ant k from 1 to m, repeat S7-4 and S7-6 step, until ant k has traveled through all dispensing points.The method solving band weak rock mass transport truck routing problem of the present invention, for the band weak rock mass transport truck routing problem based on Real-time Traffic Information, adopts time window penalty mechanism, sets up its mathematical model; Use self-adaptation Chaos Ant Colony Optimization to solve this model, improved the optimizing ability of algorithm by the adaptive updates of algorithm information element and the chaos self-adaptative adjustment of algorithm parameter.Simulation analysis shows, the present invention has the overall situation and local optimal searching ability preferably, has higher efficiency and stability when solving the transport truck routing problem based on the band weak rock mass of Real-time Traffic Information.Logistics distribution in the present invention and real productive life is more identical, using self-adaptation Chaos Ant Colony Optimization to solve this problem, because of having better Optimizing Search ability, effectively avoiding search procedure to be absorbed in local optimum, improves diversity and the global optimizing ability of solution.

As the preferred implementation solving the method for band weak rock mass transport truck routing problem of the present invention, in described step S4, set up Related Mathematical Models according to constraint condition, comprise further:

S4-1, mathematicization constraint condition:

$\underset{n=1}{\overset{N}{\Σ}}{d}_{ij}^k\≤D_{max},i,j=1,2,...,N;k=1,2,...,m---\left(7\right)$

et _i≤T _i≤lt _ii＝1,2,...,N；T ₀＝0 (8)

$\begin{array}{c}\underset{i=1}{\overset{N}{\Σ}}{g}_i\underset{j=0}{\overset{N}{\Σ}}{x}_{jik}\left(t\right)\≤G\\ \mathrm{\α}\≤t\≤\mathrm{\β}\end{array}---\left(9\right)$

$\underset{i=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{jk},i,j=1,2,...,N;k=1,2,...,m---\left(10\right)$

$\underset{j=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{jk},i,j=1,2,...,N;k=1,2,...,m---\left(11\right)$

$\underset{i=0}{\overset{N}{\Σ}}\underset{j=0}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)\≤N,k=1,2,...,m---\left(12\right)$

x _ijk(t)＝1 i，j＝1，2，…，N；k＝1，2，…，m (13)

y _ik＝1 i，j＝1，2，…，N；k＝1，2，…，m (14)

$t_{ij}=\frac{d_{ij}}{v_{ij}\left(t\right)},i,j=1,2,...,N---\left(15\right)$

T _always=t _0i+ t _ij+ ...+t _k0i, j, k=1,2 ..., N (16)

Formula (7) is the constraint of vehicle operating range, and wherein n is client's number that vehicle k serves, and is N to the maximum; Formula (8) is vehicle arrival client i time windows constraints, T _ifor the time of vehicle point of arrival i; Formula (9) represents that the goods weight that each car transports can not exceed vehicle load quantitative limitation; Formula (10) and (11) represent the relation between Two Variables; Formula (12) represents that client's sum of each car of guarantee is less than or equal to total client's number; Formula (13) directly drives to client j for its service after representing vehicle service client i; Formula (14) represents that each client can only be served by 1 car and each client can be served; Formula (15) t _ijrepresent the running time from i to j; Formula (16) represents the T.T. in subpath driving process;

S4-2, set up band weak rock mass logistics transportation scheduling mathematic model based on Real-time Traffic Information, as shown in (17) formula:

$\begin{array}{c}\min z=c_x*\underset{t=\mathrm{\α}}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}{x}_{ijk}\left(t\right)+c_r*\underset{t=t_k}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}\[r_{ijk}\left(t\right)-t\]x_{ijk}\left(k\right)+c_1*\underset{i=1}{\overset{N}{\Σ}}\max \{({\mathrm{et}}_i-T_i),0\}\\ +c_2*\underset{i=1}{\overset{N}{\Σ}}\max \{(T_i-{\mathrm{lt}}_i),0\}+c_w*\underset{k=1}{\overset{m}{\Σ}}(t_k-t_0)\end{array}---\left(17\right)$

Objective function Z is total transport cost objective function, and α, β are initial, the end time of dispatching cycle, comprise five parts altogether: Section 1 is the payment for initiation use of vehicle, c _xfor vehicle launch unit costs; Section 2 is run cost, c _rfor vehicle travels unit costs; Section 3 and Section 4 are time window rejection penalty, wherein c ₁and c ₂represent respectively and dispatch buses earlier and the late penalty coefficient arriving dispensing place, Section 5 is driver's expense, c _wfor driver's unit cost, T _irepresent the time arriving client i, solving target is make total distribution cost of vehicle minimum.

As the preferred implementation solving the method for band weak rock mass transport truck routing problem of the present invention, also comprise after described step S7-3:

S7-4, wherein allow mP _mant carries out Chaos Search by ant group algorithm self-adaptation is chaotization, remaining m (1-P _m) ant dispensing point target of selecting next step to allow;

$p_{ij}^m\left(k\right)=\left\{\begin{array}{c}{\[{\mathrm{\τ}}_{ij}\left(k\right)\]}^{\mathrm{\α}}/\underset{s\∈{\mathrm{allowed}}_i}{\Σ}{\[{\mathrm{\τ}}_{is}\left(k\right)\]}^{\mathrm{\α}}(j\∈{\mathrm{allowed}}_i)\\ 0\end{array}\right.---\left(18\right)$

If S7-5 provides and delivers, target j meets the constraint condition of model, then a kth ant is moved to dispensing point j; Dispensing target j is joined in the goal set taboo list set of having accessed;

S7-6, judge whether to be absorbed in local optimum, carry out the chaos self-adaptative adjustment of parameter by formula (5) and (6), and carry out Pheromone update according to formula (3) and (4) and restricted information element increment bound;

S7-7, obtain current all efficient solutions and find the optimum solution of minimum deflection;

If S7-8 is in the continuous T time iteration of setting, the optimal path that algorithm obtains obviously does not become excellent, then carry out the overall situation according to formula (3) to the pheromones on optimum solution path to upgrade and disturbance, otherwise overall situation renewal is carried out to the pheromones on optimum solution path;

If S7-9 does not reach maximum iterations, then forward step S7-2 to; Otherwise export and obtain optimum solution, termination algorithm, draw distribution route.

Implement the technical scheme solving the method for band weak rock mass transport truck routing problem of the present invention and there is following beneficial effect: the method solving band weak rock mass transport truck routing problem of the present invention, for the band weak rock mass transport truck routing problem based on Real-time Traffic Information, adopt time window penalty mechanism, set up its mathematical model; Use self-adaptation Chaos Ant Colony Optimization to solve this model, improved the optimizing ability of algorithm by the adaptive updates of algorithm information element and the chaos self-adaptative adjustment of algorithm parameter.Simulation analysis shows, the present invention has the overall situation and local optimal searching ability preferably, has higher efficiency and stability when solving the transport truck routing problem based on the band weak rock mass of Real-time Traffic Information.Logistics distribution in the present invention and real productive life is more identical, using self-adaptation Chaos Ant Colony Optimization to solve this problem, because of having better Optimizing Search ability, effectively avoiding search procedure to be absorbed in local optimum, improves diversity and the global optimizing ability of solution.The present invention sets up a kind of mathematical model that more can show logistics transportation practical problems, and propose a kind of self-adaptation Chaos Ant Colony Optimization problem is solved, adaptive updates by algorithm information element: improve overall update strategy, introduce elitism strategy, suitably improve the positive feedback of the pheromones of high-quality ant release; The bound of configuration information element and pheromones increment, to reduce the big-difference excessively of pheromones on path, utilizes self-adaptation Chaos Ant Colony Optimization to solve classical vehicle route optimization problem.

Above-mentioned explanation is only the general introduction of technical solution of the present invention, in order to technological means of the present invention can be better understood, and can be implemented according to the content of instructions, and can become apparent to allow above and other object of the present invention, feature and advantage, below in conjunction with preferred embodiment, and coordinate accompanying drawing, be described in detail as follows.

Accompanying drawing explanation

The invention will be further described to utilize accompanying drawing, but the embodiment in accompanying drawing does not form any limitation of the invention.

Fig. 1 is the client's coordinate position and the time window information that solve the method for band weak rock mass transport truck routing problem of the present invention.

Fig. 2 is the optimum Distribution path scheme solving the method for band weak rock mass transport truck routing problem of the present invention.

Fig. 3 is the optimum dispensing trajectory diagram solving the method for band weak rock mass transport truck routing problem of the present invention.

Embodiment

Describe the specific embodiment of the present invention in detail below in conjunction with accompanying drawing, it illustrates principle of the present invention as the part of this instructions by embodiment, and other aspects of the present invention, feature and advantage thereof will become very clear by this detailed description.In the accompanying drawing of institute's reference, in different figure, same or analogous parts use identical drawing reference numeral to represent.For making the present invention easier to understand, specific embodiments of the invention will be set forth further below.

In order to verify superiority of the present invention, below by the present invention by carrying out instance analysis to the dispensing of certain 12 client, technical scheme of the present invention is clearly and completely described.

As Figure 1-3, a kind of method solving band weak rock mass transport truck routing problem of the present invention comprises the following steps:

S1, obtain band weak rock mass VRP problem basic parameter based on Real-time Traffic Information: the position of logistics distribution center, the geographic position of client, client cargo demand, the time windows constraints of client and the information (dead weight, maximum operating range etc.) of vehicle.It is unique that this method is applicable to home-delivery center, and the means of distribution that distribution vehicle type is unique, solving target is in urban dynamic traffic Information Network, find one group of optimum logistics distribution path serving all clients, reaches distribution cost minimum.

Particularly, logistics distribution center coordinate is (0,0), and as shown in Figure 1, parking lot has load-carrying to be the oversize vehicle of 10 tons to the time windows constraints of the coordinate of client, client cargo demand and client, and each single maximum range that goes out to drive a vehicle is 200km.Vehicle the earliest the time of departure be 8:00 in the morning, the composite factors such as Real-time Traffic Information are considered in delivery process, the dispensing obtained between 3 any 2 dispensing points of different time sections expends time in, and 3 time periods are [8:00-9:00] [9:00-11:00] [11:00-12:00].

S2, in problem, client has weak rock mass requirement: the time window of deliver goods is [et _i, lt _i], distribution vehicle is et in the service time the earliest of dispensing point i _i, arriving distribution time is the latest lt _i, late or early to all producing rejection penalty.Waiting cost per hour is c ₁, deferred charges per hour is c ₂.Vehicle Speed is determined by different time sections, 1 dispatching cycle is divided into some discrete segment periods, vehicle is change in the travel speed of each period, the inner travel speed in each interval is certain, current time travel speed can only be known by dispatching center, future time travel speed and present speed same treatment.

S3, be defined as follows variable:

S4, set up the mathematical model based on the band weak rock mass transport truck routing problem of Real-time Traffic Information according to following constraint condition.

(1) 1 parking lot, N number of client, customer demand is determined;

(2) vehicle of the same race, undercapacity;

(3) each car has maximum dispensing distance restraint and load-carrying constraint;

(4) closed type vehicle route;

(5) time window of client is known, and weak rock mass limits, and the time window of each client requires known;

(6) target is solved: in urban dynamic traffic Information Network, find one group of optimum logistics distribution path serving all clients, reach distribution cost minimum.

Set up Related Mathematical Models according to constraint condition in described step S4, comprise further:

S4-1, mathematicization constraint condition:

$\underset{n=1}{\overset{N}{\Σ}}{d}_{ij}^k\≤D_{max},i,j=1,2,...,N;k=1,2,...,m---\left(7\right)$

et _i≤T _i≤lt _ii＝1,2,...,N；T ₀＝0 (8)

$\begin{array}{c}\underset{i=1}{\overset{N}{\Σ}}{g}_i\underset{j=0}{\overset{N}{\Σ}}{x}_{jik}\left(t\right)\≤G\\ \mathrm{\α}\≤t\≤\mathrm{\β}\end{array}---\left(9\right)$

$\underset{i=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{jk},i,j=1,2,...,N;k=1,2,...,m---\left(10\right)$

$\underset{j=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{jk},i,j=1,2,...,N;k=1,2,...,m---\left(11\right)$

$\underset{i=0}{\overset{N}{\Σ}}\underset{j=0}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)\≤N,k=1,2,...,m---\left(12\right)$

x _ijk(t)＝1 i，j＝1，2，…，N；k＝1，2，…，m (13)

y _ik＝1 i，j＝1，2，…，N；k＝1，2，…，m (14)

$t_{ij}=\frac{d_{ij}}{v_{ij}\left(t\right)},i,j=1,2,...,N---\left(15\right)$

T _always=t _0i+ t _ij+ ...+t _k0i, j, k=1,2 ..., N (16)

Formula (7) is the constraint of vehicle operating range, and wherein n is client's number that vehicle k serves, and is N to the maximum.Formula (8) is vehicle arrival client i time windows constraints, T _ifor the time of vehicle point of arrival i.Formula (9) represents that the goods weight that each car transports can not exceed vehicle load quantitative limitation.Formula (10) and (11) represent the relation between Two Variables.Formula (12) represents that client's sum of each car of guarantee is less than or equal to total client's number.Formula (13) directly drives to client j for its service after representing vehicle service client i.Formula (14) represents that each client can only be served by 1 car and each client can be served.Formula (15) t _ijrepresent the running time from i to j.Formula (16) represents the T.T. in subpath driving process.

S4-2, set up band weak rock mass logistics transportation scheduling mathematic model based on Real-time Traffic Information, as shown in (17) formula:

$\begin{array}{c}\min z=c_x*\underset{t=\mathrm{\α}}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}{x}_{ijk}\left(t\right)+c_r*\underset{t=t_k}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}\[r_{ijk}\left(t\right)-t\]x_{ijk}\left(k\right)+c_1*\underset{i=1}{\overset{N}{\Σ}}\max \{({\mathrm{et}}_i-T_i),0\}\\ +c_2*\underset{i=1}{\overset{N}{\Σ}}\max \{(T_i-{\mathrm{lt}}_i),0\}+c_w*\underset{k=1}{\overset{m}{\Σ}}(t_k-t_0)\end{array}---\left(17\right)$

Objective function Z is total transport cost objective function, and α, β are initial, the end time of dispatching cycle, comprise five parts altogether: Section 1 is the payment for initiation use of vehicle, c _xfor vehicle launch unit costs, c _x=100 yuan/; Section 2 is run cost, c _rfor vehicle travels unit costs, c _r=0.5 yuan/minute; Section 3 and Section 4 are time window rejection penalty, wherein c ₁and c ₂represent respectively and dispatch buses earlier and the late penalty coefficient arriving dispensing place, c ₁=10 yuan/hour, c ₂=100 yuan/hour, Section 5 is driver's expense, c _wfor driver's unit cost, c _w=50 yuan/hour, T _irepresent the time arriving client i, solving target is make total distribution cost of vehicle minimum, represent that vehicle k has travelled the distance of client i to j.

The adaptive updates of S5, design ant group algorithm pheromones improves optimizing ability, the noise immunity of algorithm, and then avoids algorithm to be absorbed in local optimum, is embodied in:

S5-1, improve overall update strategy, introduce elitism strategy.Pheromone update is carried out in the path being greater than the average overall metric function value of current iteration to overall metric function value after each algorithm circulation, and does not carry out Pheromone update to the path being less than the average overall metric of current iteration.Overall situation update rule is only for optimum solution path, and rule is as follows:

In above formula, Q represents pheromones intensity, and its value is constant, L _gbrepresent the current global optimum's path found.

The bound of S5-2, configuration information element and pheromones increment is to reduce the big-difference excessively of pheromones on path.By suitable adjustment, by the pheromones intersity limitation on each paths built in interval [τ _min, τ _max] in, when each lastest imformation element, pheromones increment is limited in interval [Δ τ simultaneously _min, Δ τ _max].According to above-mentioned improvement and optimization, total expression formula of pheromones adaptive updates is:

In (4) formula, pheromones increment

The chaos self-adaptative adjustment of S6, design ant group algorithm calculating parameter, improves algorithm to the optimizing ability of this mathematical model and counting yield.When the optimizing of ant group algorithm every generation iteration starts, α gets initial value is Arbitrary Digit between 1 to 5, and the initial value that ρ gets is the Arbitrary Digit between 0.1-0.5, then carries out chaos self-adaptative adjustment by as shown in the formula to heuristic factor α and pheromones volatility coefficient ρ:

S7, utilize self-adaptation Chaos Ant Colony Optimization to seek the optimum Distribution path of the band weak rock mass transport truck routing problem based on Real-time Traffic Information, concrete steps are as follows:

S7-1, initialization self-adaptation Chaos Ant Colony Optimization parameter: maximum iteration time NC _max, ant number m, ρ (0), α (0), Q, τ, Δ τ, iteration count NC=0 etc., when specifically implementing, m=20, greatest iteration step number is N _c=200 times, Information Meter intensity Q=100, factor of influence α=2, β=2, random number q ₀=0.85.

S7-2, ant are from the position of home-delivery center.

S7-3, to ant k from 1 to m, repeat S7-4 and S7-6 step, until ant k has traveled through all dispensing points.

S7-4, wherein allow mP _mant carries out Chaos Search by ant group algorithm self-adaptation is chaotization, remaining m (1-P _m) ant dispensing point target of selecting next step to allow.

$p_{ij}^m\left(k\right)=\left\{\begin{array}{c}{\[{\mathrm{\τ}}_{ij}\left(k\right)\]}^{\mathrm{\α}}/\underset{s\∈{\mathrm{allowed}}_i}{\Σ}{\[{\mathrm{\τ}}_{is}\left(k\right)\]}^{\mathrm{\α}}(j\∈{\mathrm{allowed}}_i)\\ 0\end{array}\right.---\left(18\right)$

If S7-5 provides and delivers, target j meets the constraint condition of model, then a kth ant is moved to dispensing point j; Dispensing target j is joined in the goal set taboo list set of having accessed.

S7-6, judge whether to be absorbed in local optimum, carry out the chaos self-adaptative adjustment of parameter by formula (5) and (6), and carry out Pheromone update according to formula (3) and (4) and restricted information element increment bound.

S7-7, obtain current all efficient solutions and find the optimum solution of minimum deflection.

If S7-8 is in the continuous T time iteration of setting, the optimal path that algorithm obtains obviously does not become excellent, then carry out the overall situation according to formula (3) to the pheromones on optimum solution path and upgrade and disturbance.Otherwise carry out the overall situation to the pheromones on optimum solution path to upgrade.

If S7-9 does not reach maximum iterations, then forward step S7-2 to; Otherwise export and obtain optimum solution, termination algorithm, show that distribution route is as shown in Figure 3.

The method solving band weak rock mass transport truck routing problem of the present invention, for the band weak rock mass transport truck routing problem based on Real-time Traffic Information, adopts time window penalty mechanism, sets up its mathematical model; Use self-adaptation Chaos Ant Colony Optimization to solve this model, improved the optimizing ability of algorithm by the adaptive updates of algorithm information element and the chaos self-adaptative adjustment of algorithm parameter.Simulation analysis shows, the present invention has the overall situation and local optimal searching ability preferably, has higher efficiency and stability when solving the transport truck routing problem based on the band weak rock mass of Real-time Traffic Information.Logistics distribution in the present invention and real productive life is more identical, using self-adaptation Chaos Ant Colony Optimization to solve this problem, because of having better Optimizing Search ability, effectively avoiding search procedure to be absorbed in local optimum, improves diversity and the global optimizing ability of solution.The present invention sets up a kind of mathematical model that more can show logistics transportation practical problems, and propose a kind of self-adaptation Chaos Ant Colony Optimization problem is solved, adaptive updates by algorithm information element: improve overall update strategy, introduce elitism strategy, suitably improve the positive feedback of the pheromones of high-quality ant release; The bound of configuration information element and pheromones increment, to reduce the big-difference excessively of pheromones on path, utilizes self-adaptation Chaos Ant Colony Optimization to solve classical vehicle route optimization problem.

Finally to should be noted that; above embodiment is only in order to illustrate technical scheme of the present invention but not limiting the scope of the invention; although be explained in detail the present invention with reference to preferred embodiment; those of ordinary skill in the art is to be understood that; can modify to technical scheme of the present invention or equivalent replacement, and not depart from essence and the scope of technical solution of the present invention.

Claims (3)

1. solve a method for band weak rock mass transport truck routing problem, it is characterized in that, comprise the following steps:

S1, obtain band weak rock mass VRP problem basic parameter based on Real-time Traffic Information: the position of logistics distribution center, the geographic position of client, client cargo demand, the time windows constraints of client and the information of vehicle;

S2, in problem, client has weak rock mass requirement: the time window of deliver goods is [et _i, lt _i], distribution vehicle is et in the service time the earliest of dispensing point i _i, arriving distribution time is the latest lt _i, late or early to all producing rejection penalty; Waiting cost per hour is c ₁, deferred charges per hour is c ₂; Vehicle Speed is determined by different time sections, 1 dispatching cycle is divided into some discrete segment periods, vehicle is change in the travel speed of each period, the inner travel speed in each interval is certain, current time travel speed can only be known by dispatching center, future time travel speed and present speed same treatment;

S3, be defined as follows variable:

S4, set up the mathematical model based on the band weak rock mass transport truck routing problem of Real-time Traffic Information according to following constraint condition:

(1) 1 parking lot, N number of client, customer demand is determined;

(2) vehicle of the same race, undercapacity;

(3) each car has maximum dispensing distance restraint and load-carrying constraint;

(4) closed type vehicle route;

(5) time window of client is known, and weak rock mass limits, and the time window of each client requires known;

(6) target is solved: in urban dynamic traffic Information Network, find one group of optimum logistics distribution path serving all clients, reach distribution cost minimum;

The adaptive updates of S5, design ant group algorithm pheromones improves optimizing ability, the noise immunity of algorithm, and then avoids algorithm to be absorbed in local optimum, is embodied in:

S5-1, improve overall update strategy, introduce elitism strategy, Pheromone update is carried out in the path being greater than the average overall metric function value of current iteration to overall metric function value after each algorithm circulation, and Pheromone update is not carried out to the path being less than the average overall metric of current iteration, overall situation update rule is only for optimum solution path, and rule is as follows:

In above formula, Q represents pheromones intensity, and its value is constant, L _gbrepresent the current global optimum's path found;

The bound of S5-2, configuration information element and pheromones increment crosses big-difference, by suitable adjustment, by the pheromones intersity limitation on each paths built in interval [τ with what reduces pheromones on path _min, τ _max] in, when each lastest imformation element, pheromones increment is limited in interval [Δ τ simultaneously _min, Δ τ _max], according to above-mentioned improvement and optimization, total expression formula of pheromones adaptive updates is:

In (4) formula, pheromones increment

The chaos self-adaptative adjustment of S6, design ant group algorithm calculating parameter, improves algorithm to the optimizing ability of this mathematical model and counting yield; When the optimizing of ant group algorithm every generation iteration starts, α gets initial value is Arbitrary Digit between 1 to 5, and the initial value that ρ gets is the Arbitrary Digit between 0.1-0.5, then carries out chaos self-adaptative adjustment by as shown in the formula to heuristic factor α and pheromones volatility coefficient ρ:

S7, utilize self-adaptation Chaos Ant Colony Optimization to seek the optimum Distribution path of the band weak rock mass transport truck routing problem based on Real-time Traffic Information, concrete steps are as follows:

S7-1, initialization self-adaptation Chaos Ant Colony Optimization parameter: maximum iteration time NC _max, ant number m, ρ ₍₀₎, α ₍₀₎, Q, τ, Δ τ, iteration count NC=0;

S7-2, ant are from the position of home-delivery center;

S7-3, to ant k from 1 to m, repeat S7-4 and S7-6 step, until ant k has traveled through all dispensing points.

2. the method solving band weak rock mass transport truck routing problem according to claim 1, is characterized in that, sets up Related Mathematical Models in described step S4 according to constraint condition, comprises further:

S4-1, mathematicization constraint condition:

$\underset{n=1}{\overset{N}{\Σ}}{d}_{ij}^k\≤D_{max},i,j=1,2,\mathrm{...},N;k=1,2,\mathrm{...},m---\left(7\right)$

et _i≤T _i≤lt _ii＝1,2,...,N；T ₀＝0 (8)

$\begin{array}{c}\underset{i=1}{\overset{N}{\Σ}}{g}_i\underset{j=0}{\overset{N}{\Σ}}{x}_{jik}\left(t\right)\≤G\\ \mathrm{\α}\≤t\≤\mathrm{\β}\end{array}---\left(9\right)$

$\begin{array}{cc}\underset{i=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{jk}& i,j=1,2,\mathrm{...},N;k=1,2,\mathrm{...},m\end{array}---\left(10\right)$

$\begin{array}{cc}\underset{j=1}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)=y_{ik}& i,j=1,2,\mathrm{...},N;k=1,2,\mathrm{...},m\end{array}---\left(11\right)$

$\begin{array}{cc}\underset{i=0}{\overset{N}{\Σ}}\underset{j=0}{\overset{N}{\Σ}}{x}_{ijk}\left(t\right)\≤N& k=1,2,\mathrm{...},m\end{array}---\left(12\right)$

x _ijk(t)＝1i，j＝1，2，…，N；k＝1，2，…，m (13)

y _ik＝1i，j＝1，2，…，N；k＝1，2，…，m (14)

$\begin{array}{cc}{t}_{ij}=\frac{d_{ij}}{v_{ij}\left(t\right)}& i,j=1,2,\mathrm{...},N\end{array}---\left(15\right)$

T _always=t _0i+ t _ij+ ...+t _k0i, j, k=1,2 ..., N (16)

Formula (7) is the constraint of vehicle operating range, and wherein n is client's number that vehicle k serves, and is N to the maximum; Formula (8) is vehicle arrival client i time windows constraints, T _ifor the time of vehicle point of arrival i; Formula (9) represents that the goods weight that each car transports can not exceed vehicle load quantitative limitation; Formula (10) and (11) represent the relation between Two Variables; Formula (12) represents that client's sum of each car of guarantee is less than or equal to total client's number; Formula (13) directly drives to client j for its service after representing vehicle service client i; Formula (14) represents that each client can only be served by 1 car and each client can be served; Formula (15) t _ijrepresent the running time from i to j; Formula (16) represents the T.T. in subpath driving process;

S4-2, set up band weak rock mass logistics transportation scheduling mathematic model based on Real-time Traffic Information, as shown in (17) formula:

$\begin{array}{c}\min z=c_x*\underset{t=\mathrm{\α}}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}{x}_{ijk}\left(t\right)+c_r*\underset{t=t_k}{\overset{\mathrm{\β}}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}\underset{k=1}{\overset{m}{\Σ}}\[r_{ijk}\left(t\right)-t\]x_{ijk}\left(t\right)+c_1*\underset{i=1}{\overset{N}{\Σ}}\max \{\left({\mathrm{et}}_i-T_i\right),0\}\\ +c_2*\underset{i=1}{\overset{N}{\Σ}}\max \{\left(T_i-{\mathrm{lt}}_i\right),0\}+c_w*\underset{k=1}{\overset{m}{\Σ}}(t_k-t_0)\end{array}---\left(17\right)$

Objective function Z is total transport cost objective function, and α, β are initial, the end time of dispatching cycle, comprise five parts altogether: Section 1 is the payment for initiation use of vehicle, c _xfor vehicle launch unit costs; Section 2 is run cost, c _rfor vehicle travels unit costs; Section 3 and Section 4 are time window rejection penalty, wherein c ₁and c ₂represent respectively and dispatch buses earlier and the late penalty coefficient arriving dispensing place, Section 5 is driver's expense, c _wfor driver's unit cost, T _irepresent the time arriving client i, solving target is make total distribution cost of vehicle minimum.

3. the method solving band weak rock mass transport truck routing problem according to claim 2, is characterized in that, also comprise after described step S7-3:

S7-4, wherein allow mP _mant carries out Chaos Search by ant group algorithm self-adaptation is chaotization, remaining m (1-P _m) ant dispensing point target of selecting next step to allow;

$p_{ij}^m\left(k\right)=\{\begin{array}{c}{\[{\mathrm{\τ}}_{ij}\left(k\right)\]}^{\mathrm{\α}}/\underset{s\∈{\mathrm{allowed}}_i}{\Σ}{\[{\mathrm{\τ}}_{is}\left(k\right)\]}^{\mathrm{\α}}(j\∈{\mathrm{allowed}}_i)\\ 0\end{array}---\left(18\right)$

If S7-5 provides and delivers, target j meets the constraint condition of model, then a kth ant is moved to dispensing point j; Dispensing target j is joined in the goal set taboo list set of having accessed;

S7-6, judge whether to be absorbed in local optimum, carry out the chaos self-adaptative adjustment of parameter by formula (5) and (6), and carry out Pheromone update according to formula (3) and (4) and restricted information element increment bound;

S7-7, obtain current all efficient solutions and find the optimum solution of minimum deflection;

If S7-8 is in the continuous T time iteration of setting, the optimal path that algorithm obtains obviously does not become excellent, then carry out the overall situation according to formula (3) to the pheromones on optimum solution path to upgrade and disturbance, otherwise overall situation renewal is carried out to the pheromones on optimum solution path;

If S7-9 does not reach maximum iterations, then forward step S7-2 to; Otherwise export and obtain optimum solution, termination algorithm, draw distribution route.