CN101853332A - Multi-facility fair site selecting method and system - Google Patents

Abstract

The invention relates to a multi-facility fair site selecting method. The method comprises the following steps of: calculating a shortest distance from all demand points to facility points and constructing a distance matrix; calculating a weighted average distance from each selectable facility point to all the demand points according to the distance matrix; constructing a matrix of a weighted distance variance between all the facility points and the demand points; constructing a fair site selecting target function according to a site selecting principle of selecting the facility point with the smallest weighted distance variance; solving the target function and finishing multi-facility fair site selection. The method and a system simplify the solving difficulty of the fair problem of multi-facility site selection, select by adopting an ant colony algorithm, more approach a practical value and are suitable for solving the site selecting problem of public service facilities in all hierarchies.

Description

Multi-facility fair site selecting method and system thereof

Technical field

The present invention relates to public service facility addressing technical field, relate in particular to a kind of multi-facility fair site selecting method and system thereof.

Background technology

Usually after vital emergent event takes place in the somewhere, mainly start emergency preplan by local government, set up floor manager portion, the emergency disposal work of unified command scheduling accident comprises that urgent transfers all available forces tackle with the decision service of dealing with contingencies with resource, composition expert group to provide, issue provisional order in good time and reinforce etc. to transfer relevant strength.Owing to lack the coordination system of emergent strain aspect between Government departments, public sector lacks the consciousness of prevention accident, causes the addressing of existing emergency resources configuration and emergency service to have unreasonable part.The research of present facility site selecting method both domestic and external mainly lays particular emphasis on setting up on the plane and executes many facilities siting analysis on location problem and the discrete networks, only rely on and set up non-linear objective function, by methods such as flord, ask for Gini coefficient minimum or weighted distance variance result hour, the flord algorithm is specially from the weight matrix of the distance of representing two summits, summit of each insertion, the known shortest path of more any point-to-point transmission and issuable path distance when inserting the summit as intermediate vertex, it is new for weight matrix to obtain to get smaller value then.These researchs lay particular emphasis on algorithm design, do not meet requirement of actual application: at first often will face many facilities location problem in the reality; Secondly, should consider the restriction of nature, social condition; At last, result of calculation may the value of departing from objectives.

China is subjected to disaster to influence one of the most serious country in the world, and special geographical environment has determined the disaster of China to have characteristics such as kind is many, generation frequency height, disaster is concentrated, loss is serious.Since 21 century, China's modernization construction enters the new stage, new situation, new problem emerge in an endless stream, because the public safety weak foundation, disaster and people are that the various major accident disasters that cause, great public health event and social security events often take place.Only 2008, China just suffered the attack of snow disaster, hand-foot-and-mouth disease, 3 extensive accidents of earthquake, causes 11,752 hundred million yuan of direct economic losses.These vital emergent events both serious threat human life security, the economic development and the social stability that have endangered China again greatly.

Since country in 1998 has set up national emergent material stock system,, set up national emergent material stock storehouse in 10 cities through ten years development.The national emergency stock goods and materials of China have for the storage unit: 10 provinces such as Tianjin, Liaoning, Heilungkiang, Anhui, Henan, Hubei, Hunan, Guangxi, Sichuan, Shaanxi (district, city) Bureau of Civil Affairs (office).National emergent goods and materials fixed point deposit in Tianjin, Shenyang, Harbin, Hefei, Zhengzhou, Wuhan, Changsha, Chengdu, Nanning, 10 cities, Xi'an.Simultaneously, part province, city, county have also set up emergent material stock at the corresponding levels storehouse by variety of way.But on the whole, emergent material stocks at different levels storehouse ubiquity storage area is little at present, construction criteria is low, the problem of basic equipment wretched insufficiency, can not satisfy present disaster relief work needs well, more can't adapt to new period party and state's new demand that preparedness, disaster relief work are proposed.For improving emergent material stock system, must be newly-built, reconstruct and extend a collection of standard compliant emergent material stock storehouse, " the comprehensive mitigation Eleventh Five-Year Plan of country " and " national disaster relief emergency preplan " all make explicit provisions to this, and have begun to start.According to www.chinanews.com on May 11st, 2009, national emergent material stock storehouse, the whole nation will be increased to 24 by 10." the disaster relief supplies warehouse construction criteria " of house town and country construction portion, State Development and Reform Commission's approval issue (building mark 121-2009) is in execution on October 1 in 2009.

Summary of the invention

(1) technical matters that will solve

The technical problem to be solved in the present invention is: simplify the difficulty of finding the solution of many facilities location problem, a kind of public service multi-facility fair site selecting method and system thereof that is applicable on each grade level is provided.

(2) technical scheme

For achieving the above object, the invention provides a kind of multi-facility fair site selecting method, the method comprising the steps of:

S1. calculate the bee-line of each demand point, make up distance matrix to facility point;

S2. according to described distance matrix, calculate the weighted average distance of each selectable facility point to each demand point;

S3. make up the matrix of the weighted distance variance of each facility point and demand point;

S4. according to the addressing principle of choosing the facility point of weighted distance variance minimum, make up the fair site selecting objective function;

S5. find the solution described objective function, finish multi-facility fair site selecting.

Wherein, the distance among the step S1 is Euclidean distance, running distance, running time or network distance.

Wherein, step S2 further comprises:

S2.1 obtains the size of population a of each demand point _i

S2.2 tries to achieve the weighted average distance of each selectable facility point to each demand point:

${\stackrel{\‾}{d}}_i=\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i{d}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein, d _IjBe the bee-line of each demand point to facility point, m is the number of demand point.

Wherein, step S3 further comprises:

S3.1 calculates each demand point and facility point Weighted distance a between any two _id _Ij

S3.2 is with described Weighted distance a _id _IjDeduct described weighted average distance

Obtain the deviation of each demand point to facility point;

S3.3 carries out square each described deviation, promptly

S3.4 according to the deviation of step S3.3 gained square, make up the matrix of the weighted distance variance of facility point and each demand point.

Wherein, step S4 further comprises step:

S4.1 determines decision variable X _IjAnd Y _Ij, covered by facility point j as if demand point i, then X _IjGet 1, otherwise get zero, if selected Facilities Construction, the then Y of carrying out of selectable facility point j _IjGet 1, otherwise get 0;

S4.2 sets up described objective function according to the addressing principle of choosing the facility point of weighted distance variance minimum:

$\min z=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{(a_i{d}_{\mathrm{ij}}-{\stackrel{\‾}{d}}_j)}^2{X}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein, X _Ij≤ Y _j,

N is the number of facility point, and p is the number of the facility point of selection.

Wherein, the method for finding the solution described objective function among the step S5 is an ant group algorithm.

The present invention also provides a kind of multi-facility fair site selecting system, and this system comprises: distance matrix makes up module, is used for the bee-line of computation requirement point to each facility point, makes up distance matrix; The weighted average distance computing module is used for the distance matrix according to described distance matrix structure module construction, calculates the weighted average distance of each selectable facility point to each demand point; Weighted distance variance matrix computations module is used to make up the matrix of the weighted distance variance of each facility point and demand point; Objective function makes up module, is used for making up the fair site selecting objective function according to the addressing principle of choosing the facility point of weighted distance variance minimum; Find the solution module, be used to find the solution described objective function, finish multi-facility fair site selecting.

(3) beneficial effect

Method of the present invention and system simplification thereof many facilities addressing equity problem find the solution difficulty, should adopt ant group algorithm to choose, more the closing to reality value is applicable to the solution of the public service facility location problem on each grade level.

Description of drawings

Fig. 1 is the multi-facility fair site selecting method process flow diagram according to one embodiment of the present invention.

Embodiment

The multi-facility fair site selecting method that the present invention proposes is described in detail as follows in conjunction with the accompanying drawings and embodiments.

Multi-facility fair site selecting method of the present invention is to be support with emergent material stock at different levels storehouse, countries and regions, analyze the vital emergent event lower area to the demand of facility and the coverage in the emergent material stock storehouse of each level, calculate the distance between each point, selected facility point.This site selecting method comprises the setting of demand point, facility point and covering radius.Wherein: demand point is the element that needs emergent goods and materials and service in the zone, can the representative, community or villages and small towns; Facility point is the layout in the emergent material stock storehouse of each level, among the present invention with they abstract be " point " element; Covering radius is meant the weighted average distance of facility point j to all demand point i, and its setting is subjected to the influence of the actual range between demand point number, demand point population, facility point and the demand point, and wherein, m is the number of demand point, and n is the number of facility point.As shown in Figure 1, the multi-facility fair site selecting method according to one embodiment of the present invention comprises step:

S1. calculate or investigate and obtain the bee-line d of each demand point to facility point _Ij, make up distance matrix;

S2. according to the distance matrix of gained, calculate the weighted average distance of each selectable facility point to each demand point;

S3. make up the matrix of the weighted distance variance of each facility point and demand point;

S4. according to the addressing principle of choosing the facility point of weighted distance variance minimum, make up the fair site selecting objective function;

S5. find the solution objective function, finish multi-facility fair site selecting.

Wherein, step S1 further comprises:

S1.1 determines the range determination standard of service facility point and demand for services point, and the distance here is an Euclidean distance, also can be running distance, running time, perhaps network distance etc., and these distances are in contact area actual time, are easier to investigate obtain;

S1.2 calculates or investigation obtains the bee-line d of each demand point to each facility point _Ij

S1.3 makes up the distance matrix of service facility point and demand for services point.

Step S2 further comprises:

S2.1 obtains the size of population a on each particular point in time of each demand point _i

S2.2 tries to achieve the weighted average distance of each selectable facility point to each demand point:

${\stackrel{\‾}{d}}_i=\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i{d}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein, m is the number of demand point.

Step S3 further comprises:

S3.1 calculates each demand point and facility point Weighted distance a between any two _id _Ij

S3.2 Weighted distance a _id _IjDeduct weighted average distance

Obtain the deviation of each demand point to facility point;

S3.3 carries out square each deviation, promptly

S3.4 according to the deviation of step S3.3 gained square, make up the matrix of the weighted distance variance of facility point and each demand point.

Step S4 further comprises step:

S4.1 determines decision variable X _IjAnd Y _Ij, X _IjBe the 0-1 variable, covered by facility point j as if demand point i, then X _IjGet 1, otherwise get zero, Y _IjFor also being 0-1 Face Changing, if selected Facilities Construction, the then Y of carrying out of selectable facility point j _IjGet 1, otherwise get 0;

S4.2 is according to the addressing principle of choosing the facility point of weighted distance variance minimum, that is: the multi-facility fair site selecting problem of weighted distance variance minimum, be equivalent to and determine p facility point, make it to serve demand point, and make the weighted distance variance minimum of each demand point to facility point, set up objective function:

$\min z=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{(a_i{d}_{\mathrm{ij}}-{\stackrel{\‾}{d}}_j)}^2{X}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein, n is the number of facility point, additional afterwards necessary constraint condition, and its basic step is:

(1) makes and selected to be at 10 and to cover demand points that X is arranged _Ij≤ Y _j,

(2) efficiency of service of reinforcement facility point guarantees that each demand point can have a facility point that service is provided at most, eliminates the repetition covering problem between each facility point, that is:

$\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{X}_{\mathrm{ij}}\≤1,\∀i;$

(3) specify the facility of selecting to count and be p, promptly

$\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{Y}_j=p;$

(4) restriction decision variable X and Y, X _Ij=0,1, Y _Ij=0,1.

The method of finding the solution described objective function among the step S5 is an ant group algorithm, finds the solution to comprise step:

S5.1nc ← 0 (the iteration step number or the searching times of nc finger counting method), each parameter initialization;

S5.2 is provided with the initial pool of corresponding each variable of each ant, and each ant is calculated the minimum value of corresponding variable combination, calculates the difference of variable combination, calculate transition probability and whether carry out combined exchange, if exchange then will be made up i and be substituted with j, increase the pheromones of each variable of j;

S5.3 calculates the target function value of each ant, writes down currently preferably to separate;

S5.4 presses renewal equation and revises track intensity, and renewal equation is:

${\mathrm{\τ}}_i\left(\mathrm{next}\right)=(1-\mathrm{\ρ}){\mathrm{\τ}}_i\left(\mathrm{old}\right)+\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\τ}}_i^k;$

S5.5 puts $\▿{\mathrm{\τ}}_{\mathrm{ij}}\←0;\mathrm{nc}\←\mathrm{nc}+1;$

S5.6 is if nc less than predetermined iterations and there is not the degeneration behavior, then goes to step S5.2;

S5.7 obtains and currently preferably separates.

Multi-facility fair site selecting system according to one embodiment of the present invention comprises: distance matrix makes up module, is used for the bee-line of computation requirement point to each facility point, makes up distance matrix; The weighted average distance computing module is used for the distance matrix according to described distance matrix structure module construction, calculates the weighted average distance of each selectable facility point to each demand point; Weighted distance variance matrix computations module is used to make up the matrix of the weighted distance variance of each facility point and demand point; Objective function makes up module, is used for making up the fair site selecting objective function according to the addressing principle of choosing the facility point of weighted distance variance minimum; Find the solution module, be used to find the solution described objective function, finish multi-facility fair site selecting.

Stocking the facility location problem with the emergent goods and materials of China national level is example, and implementation process is considered the related request of " People's Republic of China's national master plan for responding to public emergencies ", " national accident reply method "." People's Republic of China's national master plan for responding to public emergencies " according to factors such as its character, the order of severity, controllability and coverages, generally is divided into level Four to all kinds of Emergent Public Events: I level (great especially), II level (great), III level (bigger) and IV level (generally).

Method of the present invention mainly is applicable to the facility location problem under reply I level (great especially), II level (great) emergency circumstances, also promptly after accident takes place, under the central government instructs, provincial government's (perhaps uniting between provincial government) is responsible for, and need call the situation of national emergent material stock resource.Because therefore huge economy, culture, technology and the scheduling of resource ability of each provincial seat of government, be without loss of generality, represent each facility point and demand point in the inventive method model with each provincial seat of government of China.

Consider to select in the China mainland scope 24 provincial government units to set up national emergent material stock storehouse, with the situation under reply I level (great especially), II level (great) accident, suppose that each facility point (emergent material stock storehouse) can independently satisfy the relief demand of the demand point in the own covering radius, does not need the assistance of other facility point.Under the above-mentioned condition, determine that China's national emergency resources warehouse addressing equity criterion is the criterion of weighted distance variance minimum, can satisfy the requirement of China's national emergency resources warehouse location problem according to objective function that the inventive method is built fully.Distance is represented with Euclidean distance between each provincial government of China, and data owner is wanted and can be wanted the distance measurement function of map net website to record according to me.As shown in Figure 1, the multi-facility fair site selecting method according to one embodiment of the present invention mainly comprises step:

S1. record 31 provincial government unit's distances between any two of China mainland, promptly demand point makes up distance matrix to the bee-line of facility point;

S2. according to the gained distance matrix, calculate the weighted average distance of each selectable facility point to each demand point, basic step is as follows:

The S2.1 investigation obtains the size of population a on each particular point in time of each provincial government unit of China mainland _i

S2.2 tries to achieve the weighted average distance of selectable facility point (the provincial government unit in the emergent material stock storehouse of possible selected building national level) to each demand point:

${\stackrel{\‾}{d}}_i=\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i{d}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

S3. make up the matrix of the weighted distance variance of each facility point and demand point, basic step is as follows:

S3.1 calculates the Weighted distance a between any two of each provincial government unit (being each facility point and demand point) _id _Ij

S3.2 with each demand point to the Weighted distance a between the facility point _id _IjDeduct the weighted average distance of each selectable facility point to each demand point

Obtain the deviation of each demand point to facility point; S3.3 carries out square each demand point to the deviation of facility point, promptly

S3.4 makes up the matrix of the weighted distance variance of each facility point and demand point according to the deviation of step S3.3 gained square.

S4. analyze and obtain: on the basis of step S1-3 work, many facilities addressing equity problem of weighted distance variance minimum, be equivalent to and determine a plurality of facility point, make it to serve demand point, and make the weighted distance variance minimum of each demand point, thereby set up the objective function of China's national emergent material stock storehouse fair site selecting problem to facility point;

S5. the constraint condition of additional above-mentioned necessity is finished China's national emergent material stock storehouse multi-facility fair site selecting optimal design;

S6. utilize Borland Delphi 7.0 to write the ant group algorithm program, under PC windows xp environment, move, finish the calculating of finding the solution of multi-facility fair site selecting problem.

Above embodiment only is used to illustrate the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; under the situation that does not break away from the spirit and scope of the present invention; can also make various variations and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.

Claims (7)

1. multi-facility fair site selecting method, the method comprising the steps of:

S1. calculate the bee-line of each demand point, make up distance matrix to facility point;

S2. according to described distance matrix, calculate the weighted average distance of each selectable facility point to each demand point;

S3. make up the matrix of the weighted distance variance of each facility point and demand point;

S4. according to the addressing principle of choosing the facility point of weighted distance variance minimum, make up the fair site selecting objective function;

S5. find the solution described objective function, finish multi-facility fair site selecting.

2. multi-facility fair site selecting method as claimed in claim 1 is characterized in that, the distance among the step S1 is Euclidean distance, running distance, running time or network distance.

3. multi-facility fair site selecting method as claimed in claim 1 is characterized in that step S2 further comprises:

S2.1 obtains the size of population a of each demand point _i

S2.2 tries to achieve the weighted average distance of each selectable facility point to each demand point:

${\stackrel{\‾}{d}}_i=\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i{d}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein, d _IjBe the bee-line of each demand point to facility point, m is the number of demand point.

4. multi-facility fair site selecting method as claimed in claim 3 is characterized in that step S3 further comprises:

S3.1 calculates each demand point and facility point Weighted distance a between any two _id _Ij

S3.2 is with described Weighted distance a _id _IjDeduct described weighted average distance

Obtain the deviation of each demand point to facility point;

S3.3 carries out square each described deviation, promptly

S3.4 according to the deviation of step S3.3 gained square, make up the matrix of the weighted distance variance of facility point and each demand point.

5. multi-facility fair site selecting method as claimed in claim 4 is characterized in that, step S4 further comprises step:

S4.1 determines decision variable X _IjAnd Y _Ij, covered by facility point j as if demand point i, then X _IjGet 1, otherwise get zero, if selected Facilities Construction, the then Y of carrying out of selectable facility point j _IjGet 1, otherwise get 0;

S4.2 sets up described objective function according to the addressing principle of choosing the facility point of weighted distance variance minimum:

$\min z=\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{(a_i{d}_{\mathrm{ij}}-{\stackrel{\‾}{d}}_j)}^2{X}_{\mathrm{ij}}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_i}$

Wherein,

N is the number of facility point, and p is the number of the facility point of selection.

6. multi-facility fair site selecting method as claimed in claim 1 is characterized in that, the method for finding the solution described objective function among the step S5 is an ant group algorithm.

7. a multi-facility fair site selecting system is characterized in that, this system comprises:

Distance matrix makes up module, is used for the bee-line of computation requirement point to each facility point, makes up distance matrix;

The weighted average distance computing module is used for the distance matrix according to described distance matrix structure module construction, calculates the weighted average distance of each selectable facility point to each demand point;

Weighted distance variance matrix computations module is used to make up the matrix of the weighted distance variance of each facility point and demand point;

Objective function makes up module, is used for making up the fair site selecting objective function according to the addressing principle of choosing the facility point of weighted distance variance minimum;

Find the solution module, be used to find the solution described objective function, finish multi-facility fair site selecting.

Images