Summary of the invention
The shortcomings that it is a primary object of the present invention to overcome the prior art and deficiency provide a kind of based on different crowd demandParking position distribution method；The problem of for background technique, the invention proposes segmentation service crowd and according to clothesThe parking demand of business crowd redistributes existing parking resource, under the premise of guaranteeing Priority Service specific crowd,Certain parking stall is suitably opened to public vehicles.
The purpose of the present invention is realized by the following technical solution:
A kind of parking position distribution method based on different crowd demand, comprising the following steps:
S1, one is chosen according to the information of acquisition to the personnel's progress information collection for taking parking behavior in survey regionOr multiple characteristics divide crowd, so that it is determined that service group classification number m；
S2, parking characteristics investigation is carried out to different service group classifications respectively, and obtains parking characteristics index；
S3, building linear programming model, measure and supply according to demand magnitude relation, choose adaptable objective function, and tieConstraint condition is closed, the planning Berth number C for obtaining each category services crowd is solvedi；
S4, to planning Berth number CiIt is ranked up, using maximum planning Berth number as current service crowd i；
The all purposes that S5, statistics current service crowd i can be reached, and to destination frequency of occurrences FijIt is arrangedSequence, using the destination of the maximum frequency of occurrences as current destination j；
S6, the Berth number N that current service crowd i should be configured in current destination j is calculatedij, Nij=Ci·Fij, then countCalculate current service crowd i parking stall and current destination j distance D to be allocatedijk, and be ranked up to the distance obtained is calculated,The shorter preceding N of selected distanceijA berth allocation gives current service crowd；
S7, whether traverse with judging all purposes and finish, finished if not traversing, the frequency of occurrences of selected and sorted secondIt as the new maximum frequency of occurrences, and determines new current destination, goes to step S6；If traversal finishes, into next step；
S8, judge whether all service groups traverse and finish, finished if not traversing, the planning pool of selected and sorted secondDigit determines new current service crowd as new planning Berth number, goes to step S5；If traversal finishes, knot is exportedFruit.
Further, the service group classification number m is positive integer, and i=1,2 ..., m.
Further, the information includes gender, age, occupation, monthly income.
Further, the parking characteristics index includes to obtain existing parking stall sum Z, and the parking of the i-th class personnel needsThe amount of asking Ni, parking position turnover rate αi, berth wantage LiWith the idle amount M in berthi。
Further, the parking position turnover rate αiAre as follows:
αi=Si/Ti,
Wherein, SiFor the practical parking capacity for investigating the i-th class of phase personnel, and TiFor the practical parking position number of the i-th class crowdAmount.
Further, the berth wantage LiAre as follows:
Li=αi·Ni-Ci,
Wherein, aiFor parking position turnover rate, ai=Si/Ti；SiFor the practical parking capacity of poll cycle the i-th class personnel, TiFor the practical parking position quantity of the i-th class crowd, NiFor the parking demand of the i-th class personnel；
The idle amount M in the berthiAre as follows:
Mi=max { Ci-Ni, 0 },
Wherein, CiFor the planning berth number of the i-th class crowd.
Further, the step S3 specifically:
In step S3,
The objective function of the linear programming model are as follows:
Wherein constraint condition are as follows:
Wherein, m is service group classification sum, and m >=1 and m are positive integer；I is the i-th class crowd, 1≤i≤m；CiIt is i-thThe planning berth number of class crowd, Ci≥0；Z is existing parking position sum.
Further, the objective function of the linear programming model is chosen according to relation between supply and demand, specifically: work as demandAmount is greater than or equal to supply amount, then objective function are as follows:
When demand is less than supply amount, then objective function are as follows:
Compared with the prior art, the invention has the following advantages and beneficial effects:
The present invention goes to arrange parking stall according to the service characteristic and parking demand of crowd, can be more fully using existingThe parking resource parking environment current with improvement, the public vehicles for facilitating the equilibrium of supply and demand and limiting outside service range enter, and areParking area planning design and parking demand management, especially berth resource allocation provide a kind of new thinking, have practicalMeaning.
Embodiment:
A kind of parking position distribution method based on crowd demand, as shown in Figure 1, specifically:
The linear programming model that the present invention establishes is provided first, it is assumed that has m class crowd, variable-definition is as follows:
The existing parking stall sum of Z-
CiThe planning parking position quantity of-the i-th class crowd, i=1,2 ..., m
NiThe parking demand of-the i-th class crowd, i=1,2 ..., m
αiThe parking position turnover rate of-the i-th class crowd, αi=Si/Ti, SiFor the practical parking for investigating the i-th class of phase crowdAmount, and TiFor the practical parking position quantity of the i-th class crowd, i=1,2 ..., m
LiThe wantage in the-the i-th class berth, the i.e. difference of demand and supply, Li=αi·Ni-Ci, i=1,2 ..., m
MiThe idle quantity in the-the i-th class berth, Mi=max { Ci-Ni, 0 }, i=1,2 ..., m
It is as follows model can be obtained:
It is that the relation between supply and demand obtained according to investigation is chosen about the objective function of model.When the need that investigation obtainsThe amount of asking is greater than or equal to supply amount, then the demand of certain a kind of crowd or a few class crowds cannot get or just met, soObjective function usesBerth wantage should be allowed as small as possible, allow more people that can stop as far as possible.When investigation obtainsDemand be less than supply amount, then explanation drug on the market, extra berth is not utilized at this time, cause the waste of resource,Therefore it selectsIt as objective function, makes idle berth as few as possible, improves the utilization rate of resource.For constraining itemPartAnd Ci>=0, the planning parking position number C of the i-th class crowdi, should be the integer more than or equal to 0, and summation should be less thanExisting parking stall sum Z.In addition, the Optimized model belongs to linear programming model, because the order of constraint matrix is less than or equal to aboutThe number of Shu Bianliang, so the model has solution.
Based on above discussion and Fig. 1, a kind of parking position distribution method based on different crowd demand include withLower step:
Step 1
According to the classification number m of the destination determination service crowd drafted, which is any positive integer, is voluntarily set by userIt is fixed.Such as in campus, crowd can be divided into teaching and administrative staff, student and guest, that is, determine m=3；It such as within the hospital, can be withCrowd is divided into medical worker, emergency case, general patient and the family members that visit a patient, that is, determines m=4.
Step 2
Parking characteristics investigation is carried out to every a kind of crowd and obtains serial index.Including obtaining existing parking stall sum Z,The parking demand N of i-th class personneli, parking position turnover rate αi, berth wantage LiWith the idle amount M in berthi.Wherein, i is positiveInteger, value is between 1 and m.
For existing parking stall sum Z, indicate parking lot it is existing can receiving maximum capacity, unit be it is a, can be fromScene directly acquires the data.
For the planning berth number C of the i-th class crowdi, it is nonnegative integer, it is unknown quantity to be solved that unit, which is a,.CiWith the practical parking position quantity T of the i-th class crowdiThere is difference substantially, which is the parking demand by various peopleIt calculates, value size can distinguish the importance of different crowd, provide foundation for the distribution of parking stall below.
For the parking demand N of the i-th class crowdi, the main parking demand for reflecting different crowd, unit be it is a, can be withUsing common parking incidence model, Model On Relationship Analysis, motor vehicle OD predicted method and the volume of traffic-parking demand model etc.Solved, but no matter using where method, it is preferred that emphasis is different object crowds can be distinguished before analysis.
For the parking position turnover rate α of the i-th class crowdi, it illustrate during observation berth be repeated stopped it is flatEqual vehicle number reflects the producing level of parking facility, and its calculation formula is αi=Si/Ti.Wherein, SiTo investigate the i-th class of phase personnelPractical parking capacity, and TiFor the actual parking position quantity of the i-th class crowd, the two unit is a, and can be obtained from sceneAccess evidence.
For the berth wantage L of the i-th class crowdi, unit be it is a, its calculation formula is Li=αi·Ni-Ci。
It leaves unused for the berth of the i-th class crowd and measures Mi, unit be it is a, its calculation formula is Mi=max { Ci-Ni, 0 }.
Step 3
Solve the planning Berth number C of every a kind of crowdi.The data of steps 1 and 2 are substituted into formula (1), by manually calculating orThe parking position number C to match with every a kind of crowd can be obtained in special Mathematical software (such as Matlab and Lingo)i。
Step 4
By from big to small to planning Berth number CiIt is ranked up, with CiMaximum value is current value and determines current service peopleGroup i.
Step 5
With counting all purposes that current crowd i can be reached j, by from big to small to destination frequency of occurrences FijIt is arrangedSequence, with FijMaximum value is current value and determines current destination j.
Step 6
Calculate the parking stall number N that current crowd i should be equipped in current destination jijAfterwards, it calculates separately all unappropriatedParking stall and destination j distance Dijk(number that k indicates each parking stall), and by being ranked up from small to large, selected distance is shorterPreceding NijGive service group i in a parking stall.Specific calculating process is as follows:
Firstly, the Berth number N that current service crowd i should be configured in current destination jij=Ci·Fij；
Then, for service group i, whole unallocated parking stalls are calculated at a distance from current destination j, with specific reference to offerOr the plane drawing of design, the geometric center of parking stall and destination to be allocated is marked first, then geometric center connecting the two,The length of obtained line segment is the parking stall at a distance from current destination j, which can be realized by computer or artificial measuring and calculating,It after all unallocated parking stalls are obtained at a distance from current destination j, is ranked up, is determined from small to large according to its numerical values recitedDijk(wherein, apart from shortest k=1, distance time short k=2, the k=3 short apart from third, and so on)；
Finally, according to ranking results above-mentioned, the shorter preceding N of selected distanceijA unappropriated berth allocation is to current clothesBusiness crowd i.
Step 7
Judge whether all j traverse to finish, if so, 8 are gone to step, if it is not, then selected and sorted is only second to current FijIt is newFijIt as current value and determines new destination j, then goes to step 6.
Step 8
Judge whether all i traverse to finish, if so, output is as a result, entire iterative process terminates, if it is not, the then row of selectionSequence is only second to current CiNew CiIt as current value and determines new service group i, then goes to step 5.
Step 4 now provides specific example containing complicated loop structure to step 8 in order to be better described.
Assuming that existing 2 class crowd, i.e. i=1,2, obtained CiRespectively C1=10, C2=5；1st class crowd has 3 differencesDestination, i.e. j=1,2,3, the frequency F after statisticsijRespectively F11=0.3, F12=0.2, F13=0.5；2nd class crowd has 2A destination, wherein have a destination identical as the destination 2 of the 1st class, and another is different from destination above-mentioned, becauseThis, is for the 2nd class crowd, j=2,4, the frequency after then counting is respectively F22=0.4, F24=0.6；It is above-mentioned for visual representationData, available table 1.
According to step 4, C is determinediCurrent value is 10, and current service crowd is the 1st class crowd.In general, distribution obtainsCiIt is bigger, illustrate that such crowd is more important, i.e. such crowd object for should be Priority Service, therefore, the 1st class crowd should be prior to2nd class crowd is serviced.So next parking stall distribution is first since the 1st class crowd, the 1st class crowd is assignedFollowed by the 2nd class crowd is allocated.
According to step 5, the destination that current crowd (the 1st class crowd) can reach is counted.In general, an individual is answeredThere are one or more intention destinations, by the frequency of the available different destinations of statistics Different Individual, but in order to preferablyParking stall is distributed, is then frequency by Frequancy digit conversion.Frequency is higher, shows that such crowd is bigger to the demand of the destination, such asF in upper example13Maximum, illustrating that crowd 1 arrives at the destination 3 demand can be higher than to destination 1 and destination 2, so distributionShould meet the needs of crowd 1 arrives the destination when specific parking stall first.
According to step 6, for 1 class crowd, the parking stall number that destination 3 should be equipped with is N13=10 × 0.5=5, thenAccording to the parking stall in specific parking lot distribution and destination locations (such as Fig. 2 (a)), calculate between each parking stall and destination away fromFrom that is, parking stall geometric center point is at a distance from the geometric center point of destination, such as Fig. 2 (b).Because there are 15 parking stalls, institute in parking lotWith, k=1,2 ..., 15, and iteration coding in front and back will not repeat.From Fig. 2 (b) it is found that D136<D131<D137<D132<D138<D133<D139<D134<D1310<D135<D1314<D1315<D1312<D1311<D1313, therefore, 6,1,7,2,8 this 5 parking stalls is selected to distribute to the 1stClass crowd, such as Fig. 2 (c).
According to step 7, there is no traversals to finish (also surplus destination 1 and destination 2) by j, because of F11It is only second to F13, soIt selects destination 1 as next destination, executes step 6 again, the parking stall number that can calculate destination 1 should be equipped with isN13=10 × 0.3=3, and D1111<D1112<D1113<D1114<D119<D1115<D1110<D113<D114<D115(such as Fig. 2 (d)), becauseThis, selects 11,12,13 this 3 parking stalls to distribute to the 1st class crowd, such as Fig. 2 (e).Because there are no traversals to finish, selectionDestination 2 is current destination and executes step 6 again, finally can distribute to the 1st class crowd for 14,15.
According to step 8, there are no traversals to finish by i, therefore the 2nd class crowd is selected to carry out distribution as above, and i traversal finishesAfterwards, shown in the result of final output such as Fig. 2 (f), so far, entire assigning process terminates.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodimentLimitation, other any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present invention,It should be equivalent substitute mode, be included within the scope of the present invention.