CN102054211A - Police deployment method and system - Google Patents
Police deployment method and system Download PDFInfo
- Publication number
- CN102054211A CN102054211A CN2009102363969A CN200910236396A CN102054211A CN 102054211 A CN102054211 A CN 102054211A CN 2009102363969 A CN2009102363969 A CN 2009102363969A CN 200910236396 A CN200910236396 A CN 200910236396A CN 102054211 A CN102054211 A CN 102054211A
- Authority
- CN
- China
- Prior art keywords
- population
- individuals
- solution set
- police
- deployment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000011156 evaluation Methods 0.000 claims abstract description 41
- 238000010276 construction Methods 0.000 claims abstract description 15
- 230000002068 genetic effect Effects 0.000 claims abstract description 13
- 241001622623 Coeliadinae Species 0.000 claims description 14
- 230000001960 triggered effect Effects 0.000 claims description 4
- 238000005457 optimization Methods 0.000 description 18
- 108090000623 proteins and genes Proteins 0.000 description 15
- 230000008569 process Effects 0.000 description 10
- 230000035772 mutation Effects 0.000 description 7
- 206010039203 Road traffic accident Diseases 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 206010012374 Depressed mood Diseases 0.000 description 1
- 101150076211 TH gene Proteins 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a police deployment method and a police deployment system. The method comprises the following steps of: establishing two populations corresponding to traffic police information and position information respectively; initializing the two populations according to input original data, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting the number of evolution generations as 1; constructing the current generation solution set of police deployment according to the current generation population individuals and corresponding population representative individuals; performing performance evaluation on the current generation solution set; judging whether the number of the current evolution generations meets the preset evolution termination condition or not; if so, generating a non-inferior solution set of the police deployment according to a performance evaluation result; and otherwise, selecting population individuals and corresponding population representative individuals according to the performance evaluation result, performing genetic operation on the population individuals to generate a next generation population, adding 1 to the number of the current evolution generations, and returning to the solution set construction step. The method and the system can be used for improving deployment efficiency and realize multi-objective optimal deployment.
Description
Technical Field
The invention relates to the field of traffic, in particular to a police force deployment method and a police force deployment system for the field of traffic.
Background
The conflict between the rapid increase of the motorized traffic demand and the delay of the construction of traffic infrastructure is opened by reform, so that a series of traffic problems such as traffic jam, traffic accident and the like are brought while people enjoy the automobile civilization. The system can be used as a public security traffic police department for managing traffic, can grasp various massive information in time, has the capacity of processing traffic services and emergencies, and plays an important role in solving the traffic problems. Therefore, public transport is developed, a safe, convenient, comfortable, fast and economic trip mode is provided for vast residents, and sufficient traffic police force is indispensable.
The existing traffic police resource deployment is mainly realized by depending on expert experience, taking the traffic police deployment in a certain city as an example, a city public security traffic police department can regularly deploy police according to the statistical road traffic condition in the district: and a large number of traffic police resources are configured on key road sections such as traffic jam and traffic accident multi-occurrence points, and a small number of traffic police resources are configured on other road sections.
Clearly, such police deployments suffer from the following drawbacks:
the deployment is carried out by depending on expert experience, and the method has subjectivity, and the defects of excessive police strength of some road sections and insufficient police strength of some road sections can be caused, so that the deployment police strength is large;
secondly, because of the subjectivity of deployment, the deployed police force often cannot meet the uncontrollable actual work requirements such as sudden traffic incidents and the like, so that the problem of low deployment performance exists;
thirdly, the deployment efficiency is low, and a large amount of time is needed from the statistics of road traffic conditions in the district to the implementation of police deployment.
In summary, one of the technical problems that needs to be urgently solved by those skilled in the art is: how to provide a police deployment scheme to improve deployment efficiency and realize optimal deployment with minimum deployment police strength and maximum deployment performance.
Disclosure of Invention
The invention aims to provide a police force deployment method and a police force deployment system, which are used for improving deployment efficiency and realizing multi-objective optimized deployment with minimum deployment police force and maximum deployment performance.
In order to solve the above problems, the invention discloses a police force deployment method, comprising:
establishing two populations respectively corresponding to traffic police information and position information;
initializing the two populations according to input original data containing traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
constructing a current generation solution set of police force deployment according to the population individuals of the current generation and the corresponding population representative individuals;
performing performance evaluation on the current generation solution set;
judging whether the current evolution algebra meets preset evolution termination conditions, if so, generating a non-inferior solution set of police deployment according to a performance evaluation result;
otherwise, selecting population individuals and corresponding population representative individuals according to the performance evaluation result, performing genetic operation on the individuals of each population respectively to generate a next generation population, adding 1 to the current evolution algebra, and returning to the solution set construction step.
Preferably, the step of evaluating the performance of the current generation solution set includes:
calculating a fitness function value of the current generation solution set;
and according to the fitness function value, performing non-dominated sorting on the current generation solution set by adopting a non-dominated sorting algorithm and a density evaluation algorithm.
Preferably, the fitness function is the sum of the minimum distances between every two traffic policemen;
the distance between every two traffic policemen is the product of the distance between two positions corresponding to the two traffic policemen and the road weight between the two positions.
Preferably, the method further comprises:
and merging the current generation population into the next generation population, and performing solution set construction operation by using the merged current generation solution set.
Preferably, the initialization step includes:
and randomly generating L initial generation population individuals by encoding according to the original data, and randomly selecting m individuals from the L initial generation population individuals as initial generation population representative individuals, wherein L, m is a natural number, and m is less than L.
Preferably, the step of constructing the current generation solution set comprises:
and respectively carrying out position matching on all individuals of one population and representative individuals of another population, and combining the position matching results to obtain a current generation solution set.
The embodiment of the invention also discloses a police force deployment system, which comprises:
the group establishing module is used for establishing two groups, and the two groups respectively correspond to the traffic police information and the position information;
the population initialization module is used for initializing the two populations according to input original data containing traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
the solution set construction module is used for constructing a current generation solution set of police force deployment according to the population individuals of the current generation and the corresponding population representative individuals;
the evaluation module is used for evaluating the performance of the current generation solution set;
the judging module is used for judging whether the current evolution algebra meets a preset evolution termination condition, if so, the non-inferior solution set generating module is triggered, and otherwise, the evolution module is triggered;
the non-inferior solution set generation module is used for generating a non-inferior solution set of police deployment according to the performance evaluation result;
and the evolution module is used for selecting population individuals and corresponding population representative individuals according to the performance evaluation result, performing genetic operation on the individuals of each population respectively to generate a next generation population, adding 1 to the current evolution algebra, and triggering the solution set construction module.
Preferably, the evaluation module includes:
a calculating unit, configured to calculate a fitness function value of the current generation solution set;
and the sorting unit is used for carrying out non-dominant sorting on the current generation solution set by adopting a non-dominant sorting algorithm and a density evaluation algorithm according to the fitness function value.
Preferably, the system further comprises:
and the merging module is used for merging the current generation population into the next generation population and triggering the solution set construction module after merging is finished.
Preferably, the population initialization module includes:
the individual generation unit is used for randomly generating L initial generation population individuals through coding according to the original data;
and the representative individual selecting unit is used for randomly selecting m initial generation population representative individuals from the L initial generation population individuals, wherein L, m is a natural number, and m is less than L.
Compared with the prior art, the invention has the following advantages:
the police deployment needs to meet two optimization targets, and the first is that the smaller the number of traffic polices needing to be deployed, the better the traffic polices needing to be deployed on the premise of meeting the deployment performance; secondly, the larger the deployment performance is, the better the deployment performance is, the invention refers to a coevolution algorithm, maps the multi-objective optimization problem into an ecological system consisting of two groups, and can enable the whole system to continuously evolve due to the interaction and the co-evolution of various groups in the ecological system, so that the aim of optimization is achieved by the evolution of the ecological system, and therefore, the invention can automatically deploy the traffic polices and the single traffic polices at the positions corresponding to one another one by one, thereby realizing the multi-objective optimization aim of the minimum number of the traffic polices and the maximum deployment performance based on a non-inferior solution set;
secondly, the sum of the minimum distances between every two traffic polices reflecting the deployment performance is used as the basis for evaluating the current generation solution set, and different weights can be given to the roads between two positions corresponding to any two traffic polices according to actual road traffic conditions such as traffic jam, traffic accidents and the like, so that the important degree of the different roads on the requirements of the traffic polices can be measured, and the multi-objective optimization of the number of the minimum traffic polices and the maximum deployment performance is facilitated;
moreover, the invention inherits the biological characteristics of the evolutionary algorithm, has the capabilities of self-organization and self-learning, has high execution efficiency in the actual use process, and is suitable for police deployment in the traffic field.
Drawings
Fig. 1 is a flow chart of an embodiment 1 of a police force deployment method of the present invention;
fig. 2 is a flow chart of an embodiment 2 of a police force deployment method of the present invention;
fig. 3 is a flow chart of an embodiment 3 of a police force deployment method of the present invention;
fig. 4 is a block diagram of an embodiment of a police deployment system of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The traffic police force deployment method has two optimization targets for the police force deployment problem, wherein the first is that the number of the traffic polices to be deployed is expected to be as small as possible on the premise of meeting the deployment performance; the second is deployment performance, which mainly reflects the time for traffic police to arrive at the scene of the accident.
Therefore, the police force deployment problem of traffic policemen is a typical multi-objective optimization problem, and a deployment scheme which finally meets the good compromise between the minimum number of traffic policemen and the deployment performance is not a solution but a non-inferior solution set.
One of the core ideas of the invention is that the traffic polices and single traffic polices are automatically deployed at the positions corresponding to each other by using a coevolution mechanism in the nature for reference, so as to realize the multi-objective optimization purpose of the minimum traffic polices number and the maximum deployment performance based on a non-inferior solution set.
Referring to fig. 1, a flowchart of an embodiment 1 of a police force deployment method of the present invention is shown, which may specifically include:
step 101, establishing two populations respectively corresponding to traffic police information and position information;
in the coevolution algorithm, the fewer the decomposed population, the simpler the computation of the coevolution algorithm, and the fewer the search space and time. Therefore, the embodiment decomposes the police force deployment problem into two types of populations, for example, the population 1 is used to determine the traffic police information to be deployed in a certain area; the population 2 is used to determine location information within the area.
102, initializing the two populations according to input original data comprising traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
in a specific implementation, the initialization step may be: and randomly generating L initial generation population individuals by encoding according to the original data, and randomly selecting m individuals from the L initial generation population individuals as initial generation population representative individuals, wherein L, m is a natural number, and m is less than L.
In practice, the encoding information of the population may include encoding mode, encoding length and gene position information. For example, population 1 and population 2 may use different encoding schemes. The encoding mode of the population 1 can be binary encoding; the code length is the maximum number of traffic policemen available; each gene position in the binary code represents the state of the corresponding traffic police, and the number of the traffic police needing to be deployed in a certain area can be determined according to the value of the gene position. Specifically, when the genetic locus value is 0, the state of the traffic police is unselected, and the number of the traffic police is not increased; when the gene position value is 1, the state of the traffic police is selected, and the number of the traffic police is increased by 1. The encoding mode of the population 2 can be real number encoding; the encoding length can be determined according to the maximum number of traffic polices, for example, if the maximum number of traffic polices is n, where n is a natural number, the encoding length of the population 2 is 2 n; each locus in the real number code represents the position information of a certain deployment point in the region, for example, the first n locus in 2n represents the longitude information of the deployment point, the last n locus represents the latitude information of the deployment point, and the combination of the kth locus and the (n + k) th locus represents the position information of the certain deployment point in the region, wherein, k is more than or equal to 1 and less than or equal to n.
Since the present invention is configured with only one traffic police at one deployment site, the position information of the population 2 corresponds to the traffic police information of the population 1 one-to-one, for example, the kth gene site of the population 1 corresponds to the combination of the kth gene site and the n + k gene site of the population 2 one-to-one. Therefore, the encoding information of the population 2 varies depending on the population 1. For example, when the kth gene locus of the population 1 takes a value of 0, the kth gene locus and the (n + k) th gene locus of the population 2 are also correspondingly invalid.
In practice, the position information of the n most dense deployment points in a certain area can be determined according to the available maximum number n of traffic polices, and then when the number of the traffic polices corresponding to the population 1 is reduced, the number of the deployment points is reduced through the corresponding gene positions of the invalid population 2, and the position information of the current deployment point is thinned. For example, when police force is deployed for a certain road section, initially, the position information of the most dense 10 deployment points can be calculated according to the maximum available traffic police number of 10, and when the current traffic police number is reduced to 5, the interval of the corresponding 5 deployment points is changed to 2 times of the original interval.
As an example, the operation of randomly generating L initial generation population individuals for population 1 may be to randomly generate integers of [0, 1] interval to fill all gene position-linked gene chains of the L individuals one by one, and since the generated random numbers are uniformly distributed, the initial generation population traverses the whole solution space and can sufficiently reflect the behavior of the solution of the optimization problem.
103, constructing a current generation solution set of police force deployment according to the population individuals of the current evolution algebra and the corresponding population representative individuals;
in the coevolution algorithm, a single population cannot represent a complete solution of the problem to be optimized, and the complete solution of the problem to be optimized can be generated through mutual cooperation of various population individuals. In the embodiment of the invention, the number of the current traffic polices needing to be deployed can be obtained by counting the number of the gene positions with the value of '1' in each individual 1 of the population, so that the information of all the gene positions of the population 2 is determined according to the number of the current traffic polices. For example, if a certain gene position takes a value of "1" and the relative position of the population 1 individual is k, the specific position information of a certain deployment point is obtained by combining the kth position and the (n + k) th position of the population 2 individual. Therefore, the individual of the comprehensive population 1 and the population 2 can cooperate to generate the traffic police needing to be deployed and the specific deployment position of each traffic police, and the combination of the two is used as a group of complete solutions of the current generation solution set.
To reduce the amount of computation, in a preferred embodiment of the present invention, a position matching method is used to obtain the current generation solution set. Specifically, all individuals of one population may be respectively position-matched with representative individuals of another population, and the position-matching results are combined to obtain a current generation solution set.
For example, for the case of L population individuals and m population representative individuals, L individuals of the population 1 and m representative individuals of the population 2 are subjected to position matching, and L × m complete solutions can be obtained; similarly, the L individuals of the population 2 cooperate with the m representative individuals of the population 1 to obtain L × m complete solutions. Thus, the number of complete solutions in the current generation solution set is 2 × L × m.
Of course, the above process of constructing the current generation solution set by using the position matching method is only an example, and a person skilled in the art may use other construction methods according to actual situations, for example, a traversal combination method, etc., and the present invention does not limit the specific construction method.
104, evaluating the performance of the current generation solution set;
the method comprises the step of evaluating the performance of a current generation solution set obtained by a collaborative optimization algorithm.
Step 105, judging whether the current evolution algebra meets preset evolution termination conditions, if so, executing step 106; otherwise, go to step 107;
if the evolution termination condition is met, ending the evolution process and outputting an evolution result; otherwise, returning to re-evolution of each population. The preset evolution termination condition may be that the evolution process reaches a certain number of generations, for example, the preset evolution number of generations is 100, and the evolution is terminated after 100 evolutions.
106, generating a non-inferior solution set of police deployment according to the performance evaluation result;
with respect to the single-objective optimization problem, the result of multi-objective optimization is not an optimal solution, but a set of solutions that conform to the optimal concept. Therefore, the finally generated police deployed non-inferior solution set of the invention can comprise a plurality of non-inferior solutions, wherein each non-inferior solution corresponds to a police deployed scheme of the number of traffic policemen and the corresponding deployment position of a single police. In actual work, an expert can select a solution scheme according to needs. Since step 102 encodes the original data including the traffic police information and the location information, after obtaining the non-inferior solution set, the non-inferior solution set needs to be decoded into the target data as the police deployment scheme.
For example, the final non-inferior solution set contains three deployment scenarios of 25 police, 27 police and 30 police, while the currently available police for dispatch is 35, and if the expert finds 6 people with low mood, the 27 police scenario can be selected.
Step 107, selecting population individuals and corresponding population representative individuals according to the performance evaluation result;
for example, the population individuals may be selected by selecting L complete solutions from 2 × L × m complete solutions according to the performance evaluation result, and decomposing the L complete solutions into two populations of individuals.
And 108, respectively carrying out genetic operation on the individuals of each population to generate a next generation population, adding 1 to the current evolution algebra, and returning to the step 103.
In the step, the genetic operation of the population is respectively and independently carried out on the population 1 and the population 2 on the basis of the current generation population, and the next generation population is generated. The genetic algorithm mainly comprises the following three genetic operations: selection, crossover, and mutation. In order to ensure the diversity of the population and the effectiveness of the algorithm, the crossover and mutation operations both comprise several specific implementation methods, and the specific implementation methods can be randomly selected by the algorithm in the actual operation process.
1) Selection operation
In practice, the individual selection operation can be performed using roulette. In order to prevent the optimal individuals from being overlooked during the selection operation, an optimal individual preservation method is adopted, in which the individuals with the highest performance evaluation in step 106 are directly copied to the next generation without crossing and mutation operations.
2) Crossover operation
Suppose that
And
two population individuals crossed for the r generation, l is the length of the individual, and in a specific implementation, the following three crossing operations can be randomly adopted:
discrete crossing: randomly selecting individual position delta epsilon [1, l ], then the corresponding individual of the r +1 th generation is:
arithmetic intersection: λ ∈ [0, 1] is a uniformly distributed random number, then the corresponding individuals of the r +1 th generation are:
and (3) heuristic crossing: λ ∈ [0, 1] is a uniformly distributed random number, then the corresponding individuals of the r +1 th generation are:
wherein <math><mrow><msubsup><mi>H</mi><mi>s</mi><mrow><mi>r</mi><mo>+</mo><mn>1</mn></mrow></msubsup><mo>∈</mo><mo>[</mo><msup><mi>H</mi><mi>min</mi></msup><mo>,</mo><msup><mi>H</mi><mi>max</mi></msup><mo>]</mo><mo>,</mo></mrow></math> <math><mrow><msubsup><mi>H</mi><mi>t</mi><mrow><mi>r</mi><mo>+</mo><mn>1</mn></mrow></msubsup><mo>∈</mo><mo>[</mo><msup><mi>H</mi><mi>min</mi></msup><mo>,</mo><msup><mi>H</mi><mi>max</mi></msup><mo>]</mo></mrow></math>
3) Mutation operation
Suppose that
For the r-th generation, i is the length of the individual, and the following two mutation operations can be randomly adopted:
single point mutation: randomly selecting individual position delta epsilon [1, l ∈ ]]And random number
Wherein,
and
for the maximum and minimum number of individual δ positions, the corresponding individuals of the r +1 th generation are:
multipoint mutation: randomly selecting a plurality of individual positions δk∈[1，l]k∈[1，l]And a plurality of random numbers
Wherein,
and
a plurality of deltas corresponding to an individualkMaximum and minimum number of positions, corresponding individuals of the r +1 th generation are:
the invention maps the multi-objective optimization problem into an ecological system consisting of two groups by using a co-evolution algorithm for reference, and the evolution of the ecological system is jointly carried out due to the interaction and the co-evolution of various groups in the ecological system, so that the whole system can be continuously evolved, and the aim of optimization is achieved by the evolution of the ecological system, therefore, the invention can carry out automatic deployment of traffic polices and corresponding positions, thereby realizing the multi-objective optimization aim of the minimum number of the traffic polices and the maximum deployment performance.
Referring to fig. 2, a flowchart of an embodiment 2 of the police force deployment method of the present invention is shown, which may specifically include:
202, initializing the two populations according to input original data comprising traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
with respect to the single-objective optimization problem, the result of multi-objective optimization is not an optimal solution, but a set of solutions that conform to the optimal concept. Due to the fact that different evolution strategies are adopted in the evolution process, performance evaluation of the algorithm is very difficult, and a quantitative evaluation method of the system is lacked up to now.
In this embodiment, from the perspective of deployment performance that needs to be satisfied, a fitness function of a current generation solution set is evaluated according to a deployment performance definition, and a non-dominated sorting algorithm and a density evaluation algorithm are adopted to perform non-dominated sorting on the current generation solution set according to the fitness function value.
In practice, since deployment performance is mainly reflected as the time of arrival of the traffic police at the accident scene, deployment performance can be measured by the distance between one traffic police and the nearest other traffic police.
In the embodiment of the present invention, it is preferable that the sum of the minimum distances between two traffic polices is used to measure the deployment performance, that is, the definition of the fitness function may be the sum of the minimum distances between two traffic polices; the distance between two traffic policemen may be defined as a product of a distance between two locations corresponding to the two traffic policemen and a road weight between the two locations.
For example, the distance between the ith traffic police and the jth traffic police is defined as d (i,j) wherein d (i, j) is the product of the distance between two positions where the two traffic policemen are deployed and the road weight between the two positions, which can be calculated by a GIS (Geographic Information System), and the corresponding shortest distance d for the ith policemin(i) Comprises the following steps: dmin(i) Min (d (i, j)), therefore, the fitness function may be calculated as:
where N is the number of traffic polices deployed by the current generation solution set.
it is always desirable to choose a better solution during the algorithm's operation. For example, if there are two schemes, each using 25 polices, the first scheme has a deployment performance of 30, and the second scheme has a deployment performance of 40, then it is desirable to have the first deployment scheme in front of the algorithm run. The significance of the non-dominant sorting algorithm and the intensity evaluation algorithm for the non-dominant sorting of the current generation solution set is that a better complete solution is sorted in front, the sorting result is used as the basis for selecting the current generation solution set in the subsequent steps, and the current generation solution set is selected from front to back.
The non-dominant ranking algorithm is an algorithm for ranking the entire population P according to a non-dominant level, by which the population P is divided into a plurality of sets FiEach FiHave the same non-dominant level. The specific algorithm is as follows:
inputting: the whole population P; and (3) outputting: set of individuals F with the same level of non-dominancei
1) For each population individual xpTwo parameters are calculated for P: n ispAnd SpWherein n ispIndicates dominance xpNumber of individuals of (S)pIs represented by xpA set of dominated individual constituents;
2) all n arepThe 0 individuals constitute a first non-dominant level set F of individuals1Setting i to be 1;
3) access FiEach of the individuals x inpSet S of dominated individual componentspFor any individual xq∈SpN is a reaction ofqMinus 1, if nqIs reduced to 0, the corresponding x isqAnother set Q is drawn;
4) let i equal i +1, FiIf F, ═ QiIf not, returning to 3), otherwise, terminating the operation.
The density evaluation algorithm is a measure taken on keeping the diversity of the population, and the density of the population at the designated individual is estimated by calculating the density distance of a fitness function between two adjacent individuals of the designated individual. The specific algorithm is as follows:
inputting: output F of the fast non-dominated sorting algorithmi；FiThe number of individuals L in (a); number of fitness functions N
And (3) outputting: fiThe intensity evaluation value x of each of the individualsq_distance
1) For each individual xq∈FiSet x thereofq_distance＝0；
2) Presetting a fitness function pointer m as 1;
3) sorting all the individuals in the Fi according to the mth fitness function value;
4) to ensure that individuals at the boundary are always selected, x is set1_distance＝xL_distance＝∞；
5) Preset FiThe middle number pointer q is 2;
6)xq_distance＝xq_distance+(xq+1_m-xq-1_m) Wherein x isq+1_mRepresenting an individual xq+1The mth fitness function value of (1);
7) q is q +1, if q < L, then return to 6), otherwise, execute 8);
8) and m is m +1, if m is less than or equal to N, returning to 3), otherwise, terminating the operation.
for example, set F at its first non-dominant level1And S police force deployment schemes corresponding to the number of S traffic policemen and the deployment positions of the corresponding single policemen are selected, wherein S is a natural number.
208, selecting population individuals and corresponding population representative individuals according to the non-dominated sorting result;
for example, the population individuals may be selected by selecting the top L complete solutions from the 2 × L × m complete solutions according to the non-dominated sorting result, and decomposing the L complete solutions into two populations of individuals; the selection process of the population representative individuals can be that a set F at a first non-dominant level is obtained according to the fitness function values of the selected first L current generation solution sets1Randomly selecting m complete solutions, and taking decomposed individuals corresponding to the m complete solutions as representative individuals of respective populations.
And 209, performing genetic operation on the individuals of each population respectively to generate a next generation population, adding 1 to the current evolution algebra, and returning to 203.
The difference between this embodiment and embodiment 1 is that the sum of the minimum distances between two traffic polices that reflect the deployment performance is used as a basis for evaluating the current generation solution set, and different weights can be given to the roads between two locations corresponding to any two traffic polices according to actual road traffic conditions such as traffic jam and traffic accident, so that the importance degree of the different roads on the requirements of the traffic polices can be measured, and multi-objective optimization of the number of the minimum traffic polices and the maximum deployment performance is facilitated.
Referring to fig. 3, a flowchart of embodiment 3 of the police force deployment method of the present invention is shown, which may specifically include:
301, establishing two populations respectively corresponding to traffic police information and position information;
303, constructing a current generation solution set of police force deployment according to the population individuals of the current evolution algebra and the corresponding population representative individuals;
308, respectively carrying out genetic operation on the individuals of each population to generate a next generation population;
and 309, merging the current generation population into the next generation population, adding 1 to the current evolution generation number, and returning to the step 303.
For example, in the case of L population individuals and m population representing individuals, the present step may add L individuals of the parent to the next generation population, and evolve with the combined population.
In order to prevent the good individuals or fragments in the parents from disappearing along with the evolution process and causing waste, the embodiment adopts an elite retention strategy in the evolution process, so that the good individuals in the parents are retained in the next generation of population, and the population can be favorably and more stably evolved.
Referring to fig. 4, a block diagram of an embodiment of a police force deployment system of the present invention is shown, which may specifically include:
a population establishing module 401, configured to establish two populations, where the two populations respectively correspond to traffic police information and location information;
a population initialization module 402, configured to initialize the two populations according to input original data including traffic police information and location information, determine an initial generation population individual and a corresponding initial generation population representative individual, and set a co-evolution algebra to be 1;
in a specific implementation, the population initialization module may include the following units:
an individual generating unit 421, configured to randomly generate L initial generation population individuals by encoding according to the original data;
a representative individual selecting unit 422, configured to randomly select m representative individuals from the L initial generation population individuals as initial generation population, where L, m is a natural number and m < L.
In practice, the two populations may adopt different encoding modes. For example, populations corresponding to traffic police information employ binary coding; the population corresponding to the position information is encoded using real numbers.
A solution set constructing module 403, configured to construct a solution set of the current generation of police force deployment according to the population individuals of the current generation and the corresponding population representative individuals;
an evaluation module 404, configured to perform performance evaluation on the current generation solution set;
in a preferred embodiment of the invention, the evaluation module may comprise the following units:
a calculating unit 441, configured to calculate a fitness function value of the current generation solution set;
the sorting unit 442 is configured to perform non-dominated sorting on the current generation solution set by using a non-dominated sorting algorithm and a density evaluation algorithm according to the fitness function value.
A judging module 405, configured to judge whether the current evolution algebra meets a preset evolution termination condition, if yes, trigger a non-inferior solution set generating module, and otherwise trigger the evolution module;
a non-inferior solution set generating module 406, configured to generate a non-inferior solution set of police deployment according to the performance evaluation result;
and the evolution module 407 is configured to select population individuals and corresponding population representative individuals according to the performance evaluation result, perform genetic operations on the individuals of each population respectively to generate a next generation population, add 1 to the current evolution algebra, and trigger the solution set construction module.
In order to prevent the good individuals or fragments in the parents from disappearing along with the evolution process and causing waste, in another preferred embodiment of the present invention, the system may further comprise:
and a merging module 408, configured to merge the current generation population into the next generation population, and trigger the solution set constructing module after merging.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The police force deployment method and the police force deployment system provided by the invention are described in detail, specific examples are applied in the text to explain the principle and the implementation mode of the invention, and the description of the above embodiments is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A police deployment method, comprising:
establishing two populations respectively corresponding to traffic police information and position information;
initializing the two populations according to input original data containing traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
constructing a current generation solution set of police force deployment according to the population individuals of the current generation and the corresponding population representative individuals;
performing performance evaluation on the current generation solution set;
judging whether the current evolution algebra meets preset evolution termination conditions, if so, generating a non-inferior solution set of police deployment according to a performance evaluation result;
otherwise, selecting population individuals and corresponding population representative individuals according to the performance evaluation result, performing genetic operation on the individuals of each population respectively to generate a next generation population, adding 1 to the current evolution algebra, and returning to the solution set construction step.
2. The method of claim 1, wherein the step of performing a performance evaluation on the current generation solution set comprises:
calculating a fitness function value of the current generation solution set;
and according to the fitness function value, performing non-dominated sorting on the current generation solution set by adopting a non-dominated sorting algorithm and a density evaluation algorithm.
3. The method of claim 2, wherein the fitness function is the sum of the minimum distances between two traffic police officers;
the distance between every two traffic policemen is the product of the distance between two positions corresponding to the two traffic policemen and the road weight between the two positions.
4. The method of claim 1, further comprising:
and merging the current generation population into the next generation population, and performing solution set construction operation by using the merged current generation solution set.
5. The method of claim 1, wherein the initializing step comprises:
and randomly generating L initial generation population individuals by encoding according to the original data, and randomly selecting m individuals from the L initial generation population individuals as initial generation population representative individuals, wherein L, m is a natural number, and m is less than L.
6. The method of claim 1, wherein the step of constructing a current generation solution set comprises:
and respectively carrying out position matching on all individuals of one population and representative individuals of another population, and combining the position matching results to obtain a current generation solution set.
7. A police deployment system comprising:
the group establishing module is used for establishing two groups, and the two groups respectively correspond to the traffic police information and the position information;
the population initialization module is used for initializing the two populations according to input original data containing traffic police information and position information, determining initial generation population individuals and corresponding initial generation population representative individuals, and setting a juxtaposition evolution algebra to be 1;
the solution set construction module is used for constructing a current generation solution set of police force deployment according to the population individuals of the current generation and the corresponding population representative individuals;
the evaluation module is used for evaluating the performance of the current generation solution set;
the judging module is used for judging whether the current evolution algebra meets a preset evolution termination condition, if so, the non-inferior solution set generating module is triggered, and otherwise, the evolution module is triggered;
the non-inferior solution set generation module is used for generating a non-inferior solution set of police deployment according to the performance evaluation result;
and the evolution module is used for selecting population individuals and corresponding population representative individuals according to the performance evaluation result, performing genetic operation on the individuals of each population respectively to generate a next generation population, adding 1 to the current evolution algebra, and triggering the solution set construction module.
8. The system of claim 7, wherein the evaluation module comprises:
a calculating unit, configured to calculate a fitness function value of the current generation solution set;
and the sorting unit is used for carrying out non-dominant sorting on the current generation solution set by adopting a non-dominant sorting algorithm and a density evaluation algorithm according to the fitness function value.
9. The system of claim 7, further comprising:
and the merging module is used for merging the current generation population into the next generation population and triggering the solution set construction module after merging is finished.
10. The system of claim 7, wherein the population initialization module comprises:
the individual generation unit is used for randomly generating L initial generation population individuals through coding according to the original data;
and the representative individual selecting unit is used for randomly selecting m initial generation population representative individuals from the L initial generation population individuals, wherein L, m is a natural number, and m is less than L.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102363969A CN102054211A (en) | 2009-10-28 | 2009-10-28 | Police deployment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102363969A CN102054211A (en) | 2009-10-28 | 2009-10-28 | Police deployment method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102054211A true CN102054211A (en) | 2011-05-11 |
Family
ID=43958507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102363969A Pending CN102054211A (en) | 2009-10-28 | 2009-10-28 | Police deployment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102054211A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104834979A (en) * | 2015-05-25 | 2015-08-12 | 携程计算机技术（上海）有限公司 | Leader group scheduling method and system |
| CN104965974A (en) * | 2015-06-08 | 2015-10-07 | 浙江银江研究院有限公司 | Police resource deployment evaluating and optimizing method based on coverage degree |
| CN106855963A (en) * | 2015-12-09 | 2017-06-16 | 天维尔信息科技股份有限公司 | The data processing method and device of guiding police strength deployment |
-
2009
- 2009-10-28 CN CN2009102363969A patent/CN102054211A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104834979A (en) * | 2015-05-25 | 2015-08-12 | 携程计算机技术（上海）有限公司 | Leader group scheduling method and system |
| CN104834979B (en) * | 2015-05-25 | 2018-07-20 | 上海携程商务有限公司 | Leader row group method and system |
| CN104965974A (en) * | 2015-06-08 | 2015-10-07 | 浙江银江研究院有限公司 | Police resource deployment evaluating and optimizing method based on coverage degree |
| CN104965974B (en) * | 2015-06-08 | 2018-04-27 | 浙江银江研究院有限公司 | A kind of method assessed based on coverage and optimize police strength resource deployment |
| CN106855963A (en) * | 2015-12-09 | 2017-06-16 | 天维尔信息科技股份有限公司 | The data processing method and device of guiding police strength deployment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shanmugasundaram et al. | Genetic algorithm-based road network design for optimising the vehicle travel distance | |
| Bai et al. | A bi-objective model for location planning of electric vehicle charging stations with GPS trajectory data | |
| Hu et al. | A non-dominated sorting genetic algorithm for the location and districting planning of earthquake shelters | |
| Szeto et al. | A simultaneous bus route design and frequency setting problem for Tin Shui Wai, Hong Kong | |
| Jayaram et al. | Efficient sampling and data reduction techniques for probabilistic seismic lifeline risk assessment | |
| CN102044011B (en) | Method and system for scheduling police force resource | |
| CN106951994A (en) | A kind of site selecting method of marine emergency management and rescue website | |
| CN111612252A (en) | Automatic site selection method and device for large-scale emergency facilities and readable storage medium | |
| CN102594909A (en) | Multi-objective community detection method based on spectrum information of common neighbour matrix | |
| CN107145991B (en) | Time-varying random network dynamic path searching method considering road section correlation | |
| Mohareb | Emergency evacuation model: accessibility as a starting point | |
| Kumar et al. | Developing context sensitive planning criteria for transit oriented development (TOD): A fuzzy-group decision approach | |
| CN102789542B (en) | Load distribution method for hydrothermal power system based on multi-target distribution estimation | |
| Shen et al. | Energy-efficient multi-trip routing for municipal solid waste collection by contribution-based adaptive particle swarm optimization | |
| CN102054211A (en) | Police deployment method and system | |
| CN117041992A (en) | Method for realizing urban base station construction based on 5G technology and electronic equipment | |
| CN111260491A (en) | Method and system for discovering network community structure | |
| Sathe et al. | Optimizing location and relocation of response units in guarding critical facilities | |
| Escribano-Macias et al. | Optimal design of Rapid evacuation strategies in constrained urban transport networks | |
| Song et al. | Model analysis of traffic emergency dispatching in intelligent transportation system under cloud computing | |
| Liu et al. | Application specific instance generator and a memetic algorithm for capacitated arc routing problems | |
| CN110929950B (en) | Electric automobile load prediction method and system | |
| Qiu et al. | Urban residential land suitability index system and its comprehensive evaluation—a case study of Wenzhou | |
| Ewert et al. | Efficient per query information extraction from a hamming oracle | |
| WO2023088131A1 (en) | Traffic state prediction method and apparatus, and device, medium and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110511 |