A parking space allocation method based on an auction mechanism

Abstract

The application discloses a parking space allocation method based on an auction mechanism, comprising the following steps: S1, acquiring a parking space supplier's parking interval, hourly rental price and total rental price, and generating a parking supply matrix; S2, acquiring a parking space demander's first-choice parking interval, second-choice parking interval and hourly use price, and generating a first-choice parking demand matrix and a second-choice parking demand matrix according to the first-choice parking interval and the second-choice parking interval; S3, generating a bid hourly rental price, a bid total rental price and a bid hourly use price; and S4, performing shared parking space allocation. The application provides a new idea for optimizing parking space sharing and improving the willingness of citizens to participate in the shared market, maximizes the number of parking demanders and guarantees the benefits of parking space suppliers and parking platforms.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent parking, and particularly relates to a parking space allocation method based on an auction mechanism. BACKGROUND

[0002] The increasing number of motor vehicles makes the urban parking problem more and more serious. Due to the time difference of parking demand of parking lots of different formats, the space-time distribution of parking supply and demand is dislocated, resulting in low utilization rate of urban parking resources as a whole.

[0003] In order to realize the effective utilization of parking resources, under the promotion of Internet technology and sharing economy, parking resource management measures based on parking space sharing and allocation gradually come into the public view. Most of the existing researches on parking space sharing start from the perspectives of single parking lot allocation, multi-parking lot allocation, static reservation allocation and dynamic reservation allocation, and discuss the optimal user perception utility, the maximum platform income or the overall parking load balance as the target, and put forward a series of effective allocation schemes. However, the research scene is that the government or parking platform purchases parking right from residents at a uniform price, and then shares it to parking demanders at a fixed price. This sharing mode cannot fully consider the price cognition of different groups to shared parking, and has certain limitations. In addition, how to encourage parking supply and demand to implement sharing and how to make a fair and reasonable income distribution to the stakeholders of shared parking are still lack of research. SUMMARY

[0004] In order to solve the above problems, the application provides a parking space allocation method based on an auction mechanism.

[0005] The technical scheme of the application is as follows: A parking space allocation method based on an auction mechanism comprises the following steps:

[0006] S1: obtaining the parking space sharing interval, the hourly rental price and the total rental price of the parking space supplier, and generating a parking space supply matrix according to the parking space sharing interval;

[0007] S2: obtaining the first preferred parking space use interval, the second preferred parking space use interval and the hourly use price of the parking space demander, and generating a first preferred parking space demand matrix and a second preferred parking space demand matrix according to the first preferred parking space use interval and the second preferred parking space use interval;

[0008] S3: generating a competitive hourly rental price and a competitive total rental price according to the hourly rental price and the total rental price, and generating a competitive hourly use price according to the hourly use price;

[0009] S4: performing shared parking space allocation according to the parking space supply matrix, the first preferred parking space demand matrix, the second preferred parking space demand matrix, the competitive hourly rental price, the competitive hourly use price and the competitive total rental price.

[0010] The application has the beneficial effects of providing a new idea for optimizing parking space sharing and improving the willingness of citizens to participate in the sharing market. By designing a parking space sharing bilateral auction mechanism, on the one hand, parking space suppliers with different travel purposes, different education levels and different occupations can set parking space rental prices according to their own conditions to maximize the number of parking space supplies and improve the problem of insufficient parking spaces in the area; on the other hand, parking space demanders can set parking space use prices according to their own importance of obtaining shared parking spaces this time, and obtain parking space use rights through bidding with other parking space users, thereby maximizing the number of parking space demanders and ensuring the benefits of parking space suppliers and parking platforms.

[0011] Further, in step S1, generating the parking space supply matrix comprises the following sub-steps:

[0012] S11: obtaining a parking space sharing period T, a parking space supply start time T i s and a parking space supply end time T i e ;

[0013] S12: dividing the parking space sharing period T into K time windows, and generating a parking space sharing interval [T i s , T i e ] according to the parking space supply start time T i s and the parking space supply end time T i e ;

[0014] S13: generating a parking space supply matrix S i s according to the parking space sharing interval [T i e , T M×K .

[0015] The beneficial effects of the above further scheme are that in the application, the shareable time of the parking space supplier is divided into several continuous and equal time windows to facilitate model calculation and management of the operator. When the set time window is small enough, the allocation can be more flexible.

[0016] Further, in step S13, the calculation formula of the parking space supply matrix S M×K is:

[0017] S M×K = [s ik ]

[0018]

[0019] Among them, s ik T represents a 0-1 variable. i s T represents the start time of berth supply. i e Let Z represent the end time of berth supply, Z represent the set of integers, t represent the time interval of the division, i = 1, 2, ..., M, k = 1, 2, ..., K, M represent the number of suppliers participating in the auction, and K represent the number of time windows.

[0020] Furthermore, in step S2, generating the preferred berth demand matrix and the secondary berth demand matrix includes the following sub-steps:

[0021] S21: Start time of use of preferred parking spaces for those in need of parking. Secondary berth usage start time Preferred berth usage end time and the termination time of secondary berth use

[0022] S22: Start time of use based on preferred berth and the end time of use of the preferred berth Generate preferred berth usage area Based on the start time of use of the secondary berth and the termination time of secondary berth use Generate secondary berth usage area

[0023] S23: Use of preferred berth area and secondary berth usage area Standardization processes are performed to obtain the preferred berth usage area. and standardized secondary berth usage area

[0024] S24: Preferred berth usage area according to standardization and standardized secondary berth usage area Generate berth preferred demand matrix D N×K Berth secondary demand matrix D' N×K .

[0025] The beneficial effects of the above-mentioned further solutions are as follows: In this invention, the standardized processing of the time period for users to use parking spaces expands the parking time for users to a certain extent. That is, the auction platform reserves some flexible time for users, which helps to reduce the behavior of users parking beyond the time limit. In addition, the setting of a first-choice and a second-choice dual interval for users allows the platform to flexibly adjust the reservation time for users, which improves the utilization rate of parking spaces and the success rate of users in bidding.

[0026] Further, in step S23, the calculation formula of the normalized first berth preference usage interval is as follows:

[0027]

[0028]

[0029] wherein, represents the lower limit of the normalized first berth preference usage interval, represents the upper limit of the normalized first berth preference usage interval, represents the first berth preference usage start time, represents the first berth preference usage end time;

[0030] The calculation formula of the normalized second berth preference usage interval is as follows:

[0031]

[0032]

[0033] wherein, represents the lower limit of the normalized second berth preference usage interval, represents the upper limit of the normalized second berth preference usage interval, represents the second berth preference usage start time, represents the second berth preference usage end time;

[0034] The calculation formula of the berth first preference demand matrix D N×K in step S24 is as follows:

[0035] D N×K = [d jk ]

[0036] wherein, d jk represents a 0-1 variable;

[0037] The calculation formula of the berth second preference demand matrix D’ N×K is as follows:

[0038] D’ N×K = [d’ jk ]

[0039]

[0040]

[0041] wherein, d jkdenotes a 0-1 variable, k = 1, 2, …, K, K denotes the number of time windows, and Z denotes an integer set.

[0042] Further, in step S3, the bidding lease hour price is calculated according to the following formula:

[0043]

[0044] wherein, denotes the lease hour price, denotes the number of times of default of the supplier, and η denotes a penalty factor;

[0045] In step S3, the bidding total lease price is calculated according to the following formula:

[0046]

[0047] wherein, denotes the total lease price;

[0048] In step S3, the bidding use hour price is calculated according to the following formula:

[0049]

[0050] wherein, denotes the use hour price, denotes the number of times of default of the demander.

[0051] The above further scheme has the beneficial effect that, in the present application, the auction platform considers a default penalty mechanism, and the number of times of default of the supplier and the demander is recorded in the shared parking credit. If the number of times of default is large, the platform will reduce the auction priority, which helps to regulate the shared behavior of users.

[0052] Further, step S4 includes the following sub-steps:

[0053] S41: generating a parking space allocation matrix σ M×N of the first-choice demand and a parking space allocation matrix σ' M×N of the second-choice demand, respectively;

[0054] S42: constructing a unit price bid allocation mode and a total price bid allocation mode according to the parking space allocation matrix σ M×N of the first-choice demand, the parking space allocation matrix σ' M×N of the second-choice demand, a parking space supply matrix, a parking space first-choice demand matrix, a parking space second-choice demand matrix, the bidding lease hour price, the bidding use hour price, and the bidding total lease price, and determining a disturbance factor of the bid allocation mode;

[0055] S43: According to the disturbance factor of the bidding allocation mode, the revenue of the unit price bidding allocation mode and the revenue of the total price bidding allocation mode are calculated respectively, and the shared berth allocation is completed,

[0056] The beneficial effects of the further scheme are: in the application, the auction platform sets two bidding strategies of unit price bidding and total price bidding, considers shared time heterogeneity, breaks the limitation of traditional "one-to-one" auction, and realizes "one-to-many" auction supply-demand matching in the self-pricing scenario of berth sharing.

[0057] Further, in step S41, the berth allocation matrix σ M×N of the first preferred demand is calculated according to the following formula:

[0058] σ M×N =[δ ij ]

[0059] Wherein, δ ij represents a 0-1 variable;

[0060] In step S41, the berth allocation matrix σ' M×N of the second preferred demand is calculated according to the following formula:

[0061] σ' M×N =[δ' ij ]

[0062] Wherein, δ' ij represents a 0-1 variable;

[0063] In step S42, the constraint conditions of the unit price bidding allocation mode include the first constraint condition, the second constraint condition and the third constraint condition; the constraint conditions of the total price bidding allocation mode include the first constraint condition, the second constraint condition, the third constraint condition and the fourth constraint condition;

[0064] The expression of the first constraint condition is δ ij ∈{0,1}, δ' ij ∈{0,1};

[0065] The expression of the second constraint condition is Wherein, M represents the number of participants in the auction, and N represents the number of participants in the auction;

[0066] The expression of the third constraint condition is Wherein, K represents the number of time windows, s ik represents a 0-1 variable, d jk represents a 0-1 variable, and d' jk represents a 0-1 variable;

[0067] The expression of the fourth constraint condition is

[0068] In step S42, the calculation formula of the disturbance factor Δ of the bid allocation mode is:

[0069] Δ = Δ1- Δ2

[0070]

[0071]

[0072]

[0073] wherein Δ1 represents that the parking space provider with low bid and long sharing time is preferentially matched with the long parking demander, Δ2 represents that the first parking interval of the demander is preferentially allocated when both the first and second parking intervals of the demander meet the requirements, ω represents a minimum coefficient, represents that the provider with low bid and long sharing time is preferentially matched, represents the bidding lease hour price;

[0074] In step S43, the calculation formula of the bid allocation benefit maxW all1 of the unit price bid allocation mode is:

[0075]

[0076] wherein represents the bidding use hour price, t represents the time interval, and T represents the parking space sharing period;

[0077] In step S43, the calculation formula of the bid allocation benefit maxW all2 of the total price bid allocation mode is:

[0078]

[0079]

[0080] wherein y i represents a 0-1 variable, represents the bidding lease total price. BRIEF DESCRIPTION OF DRAWINGS

[0081] Figure 1 is a flowchart of the parking space allocation method;

[0082] Figure 2 is a provider parking space allocation diagram of the unit price bid mode;

[0083] Figure 3 is a provider parking space allocation diagram of the total price bid mode. DETAILED DESCRIPTION

[0084] Embodiments of the present application will be further described below with reference to the drawings.

[0085] As shown in Figure 1 , the present application provides a parking space allocation method based on an auction mechanism, comprising the following steps:

[0086] S1: obtaining the parking space sharing interval, hourly rental price and total rental price of the parking space provider, and generating a parking space supply matrix according to the parking space sharing interval;

[0087] S2: obtaining the preferred parking space usage interval, secondary parking space usage interval and hourly usage price of the parking space demander, and generating a preferred parking space demand matrix and a secondary parking space demand matrix according to the preferred parking space usage interval and the secondary parking space usage interval;

[0088] S3: generating a bid hourly rental price and a bid total rental price according to the hourly rental price and the total rental price, and generating a bid hourly usage price according to the hourly usage price;

[0089] S4: performing shared parking space allocation according to the parking space supply matrix, the preferred parking space demand matrix, the secondary parking space demand matrix, the bid hourly rental price, the bid hourly usage price and the bid total rental price.

[0090] In the embodiments of the present application, in step S1, generating the parking space supply matrix comprises the following sub-steps:

[0091] S11: obtaining the parking space sharing time period T, the parking space supply start time T i s and the parking space supply end time T i e ;

[0092] S12: dividing the parking space sharing time period T into K time windows, and generating a parking space sharing interval [T i s , T i e ] according to the parking space supply start time T i s and the parking space supply end time T i e ; the submitted start and end times must fall on the time window nodes, so as to generate the parking space sharing interval.

[0093] S13: generating a parking space supply matrix S M×K according to the parking space sharing interval [T i s , T i e ] and the K time windows.

[0094] In the embodiment of the present application, in step S13, the berth supply matrix S M×K The calculation formula is:

[0095] S M×K = [s ik ]

[0096]

[0097] Wherein, s ik is a 0-1 variable, T i s represents the start time of berth supply, T i e represents the end time of berth supply, Z represents an integer set, t represents a time interval, i = 1, 2, …, M, k = 1, 2, …, K, M represents the number of suppliers participating in the auction, and K represents the number of time windows. When s ik = 0, it indicates that the berth provided by the i th berth supplier is in an available state at the k th time window. Similarly, s ik = 1 indicates that the berth is not available.

[0098] In step S1, the bidders do not know the bidding information of each other, and the bidders are only allowed to bid once. If the bid fails, the bidder loses the qualification of renting the berth.

[0099] In the embodiment of the present application, in step S2, generating the berth first-choice demand matrix and the berth second-choice demand matrix comprises the following sub-steps:

[0100] S21: obtaining the first-choice berth use start time the second-choice berth use start time the first-choice berth use end time and the second-choice berth use end time To increase the winning rate of the berth demander, the present application allows the demander to submit two time windows;

[0101] S22: generating the first-choice berth use interval according to the first-choice berth use start time and the first-choice berth use end time generating the second-choice berth use interval according to the second-choice berth use start time and the second-choice berth use end time

[0102] S23: performing standardization processing on the first-choice berth use interval and the second-choice berth use interval to obtain the standardized first-choice berth use interval and the standardized secondary preferred parking space usage interval

[0103] Since it is difficult for the demander to predict the interval of using the shared parking lot to completely fall within the specified time window, the demander usage interval submitted in S21 needs to be standardized;

[0104] S24: generating the standardized primary preferred parking space usage interval and the standardized secondary preferred parking space usage interval generating the primary preferred demand matrix D N×K and the secondary preferred demand matrix D' N×K .

[0105] In the embodiment of the present application, in step S23, the standardized primary preferred parking space usage interval is calculated according to the following formula:

[0106]

[0107]

[0108] wherein, represents the lower limit of the standardized primary preferred parking space usage interval, represents the upper limit of the standardized primary preferred parking space usage interval, represents the starting time of the primary preferred parking space usage, represents the ending time of the primary preferred parking space usage;

[0109] the standardized secondary preferred parking space usage interval is calculated according to the following formula:

[0110]

[0111]

[0112] wherein, represents the lower limit of the standardized secondary preferred parking space usage interval, represents the upper limit of the standardized secondary preferred parking space usage interval, represents the starting time of the secondary preferred parking space usage, represents the ending time of the secondary preferred parking space usage;

[0113] In step S24, the primary preferred demand matrix D N×K is calculated according to the following formula:

[0114] D N×K = [d jk ]

[0115] wherein, d jk represents a 0-1 variable;

[0116] Parking berth sub-selection demand matrix D' N×K The calculation formula is:

[0117] D' N×K =[d' jk ]

[0118]

[0119]

[0120] Wherein, d jk ' represents a 0-1 variable, k = 1, 2,..., K, K represents the number of time windows, Z represents an integer set. When d jk , d jk ' = 1, it indicates that the jth demander has parking space demand in the kth period, d jk , d jk ' = 0 indicates no parking demand.

[0121] In step S2, the bidders do not know the bidding information of each other, and the demanders are only allowed to bid once. If the bid fails, the demander loses the qualification of using the parking space.

[0122] In the embodiment of the application, in step S3, the auction platform converts the leasing price submitted by the supplier to obtain the bidding leasing price after comprehensively considering the factors such as the default of the supplier sharing parking. If the supplier i has a default behavior, the platform will increase the leasing hour price submitted by the supplier based on the historical default times of the supplier, and the calculation formula of the bidding leasing hour price is as follows:

[0123]

[0124] Wherein, represents the leasing price per hour, represents the default times of the supplier, and η represents the punishment factor;

[0125] In step S3, the calculation formula of the total bidding leasing price is as follows:

[0126]

[0127] Wherein, represents the total leasing price;

[0128] In step S3, the auction platform converts the use price submitted by the demander to obtain the bidding use price after comprehensively considering the factors such as the payment of the demander and the default of the shared parking. If the demander j has a default behavior, the platform will regard the bid as lower than the bid, and increase the use hour price Based on its historical number of defaults Reduce by a certain amount, and use the hourly price for bidding. The calculation formula is:

[0129]

[0130] in, This indicates the hourly usage price. This indicates the number of times the demander defaulted.

[0131] In this embodiment of the invention, step S4 includes the following sub-steps:

[0132] S41: Generate the berth allocation matrix σ for the preferred demand. M×N Berth allocation matrix σ' for secondary demand M×N ;

[0133] S42: Berth allocation matrix σ based on preferred demand M×N σ', berth allocation matrix for secondary demand M×N The system constructs a unit price bidding allocation method and a total price bidding allocation method, based on the following: berth supply matrix, berth preferred demand matrix, berth secondary demand matrix, hourly price for competitive leasing, hourly price for competitive use, and total price for competitive leasing; and determines the disturbance factor for the bidding allocation method.

[0134] S43: Based on the disturbance factor of the bidding allocation method, calculate the revenue of the unit price bidding allocation method and the revenue of the total price bidding allocation method respectively, and complete the allocation of shared berths.

[0135] In this embodiment of the invention, in step S41, the allocation matrix is ​​a decision matrix, representing the matching relationship between the supply and demand parties participating in the auction. If supplier i and demander j are successfully matched, then δ ij The value is 1 if the berth requirement is met, and 0 otherwise. Since the platform allows berth demanders to submit both preferred and secondary berth options, the berth allocation matrix σ for preferred berth demand is... M×N The calculation formula is:

[0136] σ M×N =[δ ij ]

[0137] Where, δ ij Represents 0-1 variables;

[0138] In step S41, the berth allocation matrix σ' of the secondary demand is... M×N The calculation formula is:

[0139] σ' M×N =[δ' ij ]

[0140] Where, δ'ij Represents 0-1 variables;

[0141] In step S42, the auction platform aims to maximize social welfare by allocating shared parking spaces using two modes: unit price bidding and total price bidding. Social welfare is represented as the difference between the sum of bids from all winning buyers (parking space demanders) and the sum of bids from all winning sellers (parking space suppliers). Considering the differences in suppliers' bidding habits, the platform designs two modes: unit price bidding and total price bidding. Under the unit price bidding mode, the supplier's revenue is the product of its hourly rental price and the parking duration of all matching demanders. Under the total price bidding mode, regardless of how many demanders the supplier matches, its revenue remains constant at its total rental price. The constraints for the unit price bidding allocation method include the first, second, and third constraints; the constraints for the total price bidding allocation method include the first, second, third, and fourth constraints.

[0142] The first constraint stipulates that the allocation matrix of the model is a 0-1 matrix; the second constraint constrains the "one-to-many" relationship of supply and demand matching, indicating that demanders do not move their vehicles and that the preferred parking time and the secondary parking time cannot be allocated repeatedly; the third constraint ensures that there is no parking time conflict between demanders of successfully allocated parking spaces.

[0143] The expression for the first constraint is δ ij ∈{0,1},δ' ij ∈{0,1};

[0144] The expression for the second constraint is: Where M represents the number of suppliers participating in the auction, and N represents the number of demanders participating in the auction;

[0145] The expression for the third constraint is: Where K represents the number of time windows, s ik d represents a 0-1 variable. jk d' represents a 0-1 variable. jk Represents 0-1 variables;

[0146] The expression for the fourth constraint is:

[0147] In step S42, to avoid the supply and demand allocation method having a non-unique optimal solution, a disturbance term is introduced. The formula for calculating the disturbance factor Δ of the bidding allocation method is as follows:

[0148] Δ=Δ1-Δ2

[0149]

[0150]

[0151]

[0152] Δ1 represents that the parking space provider with low bid and long sharing time is matched with the long parking demander first, Δ2 represents that the first parking interval of the demander is allocated first when both the first and second parking intervals meet the requirement, ω represents a minimum coefficient, which can be selected flexibly according to the supply and demand scale, Δ1 represents that the parking space provider with low bid and long sharing time is matched with the long parking demander first, represents the bidding lease hour price;

[0153] In step S43, the bid allocation income maxW all1 of the unit price bid allocation mode is calculated according to the following formula:

[0154]

[0155] Δ1 represents that the parking space provider with low bid and long sharing time is matched with the long parking demander first, Δ2 represents that the first parking interval of the demander is allocated first when both the first and second parking intervals meet the requirement, ω represents a minimum coefficient, which can be selected flexibly according to the supply and demand scale, represents the bidding use hour price, t represents the time interval, and T represents the parking space sharing period;

[0156] In step S43, the bid allocation income maxW all2 of the total price bid allocation mode is calculated according to the following formula:

[0157]

[0158]

[0159] wherein y i represents a 0-1 variable, represents the bidding lease total price. y i is used to determine whether the provider i wins the bid. If the bid is won, y i = 1, otherwise y i = 0.

[0160] In the present application, the income allocation can also be performed by step S5, which specifically comprises:

[0161] Step S5 comprises the following sub-steps:

[0162] S51: calculating the parking lot operator income W p according to the income of the unit price bid allocation mode and the total price bid allocation mode;

[0163] S52: calculating the total provider benefit W s and the total demander benefit W d according to the parking lot operator income W p ;

[0164] S53: Based on the total supplier benefit W s Total benefits to consumers (W) d Calculate supplier utility separately and demander utility

[0165] S54: Based on supplier utility and demander utility Calculate the supplier's real income P i s and the actual payment made by the user Complete the distribution of profits.

[0166] In step S51, the parking lot operator's revenue W p The calculation formula is:

[0167] W p =θ·W all

[0168] Where θ represents the commission rate taken by the parking auction platform, and W all This indicates the revenue from the two bidding allocation methods;

[0169] In step S52, the supplier's total benefit W s The calculation formula is:

[0170]

[0171] Where I represents the supplier rejection rate and J represents the demand rejection rate;

[0172] In step S52, the total benefit W for the demander d The calculation formula is:

[0173]

[0174] In step S53, the supplier's utility The calculation formula is:

[0175]

[0176] in, It represents the ratio of the reciprocal of the supplier's bid to the sum of the reciprocals of all winning bids from suppliers. This represents the hourly rental price, and M represents the number of suppliers participating in the auction.

[0177] In step S53, the utility of the demander The calculation formula is:

[0178]

[0179] in, represents the ratio of the demander's bid to the sum of all winning bids of demanders, represents the demander's expected expenditure, and N represents the number of demanders participating in the auction;

[0180] In step S54, the actual income P i s of the supplier is calculated according to the following formula:

[0181]

[0182] wherein, represents the expected income of the supplier;

[0183] In step S54, the actual payment C of the demander is calculated according to the following formula:

[0184]

[0185] The technical solutions of the present application are further described below in conjunction with examples:

[0186] S11, divide the parking space sharing period T = [7:00-19:00] into multiple continuous and equal length time intervals t = 30 min, forming K = 24 time windows; for example, the supplier i = 1 submits the parking space supply starting time T i s = 7:00, the termination time T i e = 18:00, and generates the parking space sharing interval [7:00-18:00].

[0187] S12, according to the parking space sharing interval submitted by the supplier, the parking space supply matrix of all suppliers in this embodiment is

[0188] S13, the supplier confirms the hourly rental price and the total rental price Table 1 shows the hourly rental price and the total rental price of all suppliers in this embodiment. For ease of calculation, the total rental price of the supplier is simplified as the product of the hourly rental price and the rental hours.

[0189] Table 1

[0190]

[0191] S21. The auction platform allows users to submit two time windows. Table 2 shows the start and end times of the preferred parking space and the start and end times of the secondary parking space for all users in this embodiment. For simplicity, the "+1:00:00" column indicates that the user's secondary parking time is postponed by 1 hour from the preferred time.

[0192] Table 2

[0193]

[0194]

[0195] S22. Standardize the user usage range submitted in S21. Similarly, the standardized secondary usage range can be obtained. The results after processing are shown in Table 3.

[0196] Table 3

[0197]

[0198]

[0199] S23. Based on the standardized preferred / secondary berth usage demand obtained in step S22, the preferred berth demand matrix in this embodiment is: The secondary demand matrix is ​​as follows:

[0200] S24. The user confirms the hourly usage price based on their desired price. and total price Table 4 shows the hourly usage price and total usage price for all users in this embodiment. For ease of calculation, the total usage price for users is simplified to the product of the hourly usage price and the number of hours used.

[0201] Table 4

[0202]

[0203]

[0204] S31, according to the formula Calculate the hourly rental price of suppliers bidding. Where the η punishment factor is 0.2, This indicates the number of times the supplier defaults; similarly, the total price of the competitively bid lease can be obtained. As shown in Table 5.

[0205] Table 5

[0206]

[0207]

[0208] S32, according to the formula Calculate the hourly price for demanders bidding. Where the η punishment factor is 0.2, The table shows the number of times the customer defaulted, as shown in Table 6.

[0209] Table 6

[0210]

[0211]

[0212] S41. Define the decision variable berth allocation matrix. Since this invention allows users to submit two berth usage periods, the decision variable σ for the preferred berth usage period is defined. M×N =binvar(M,N,'full'), where σ is the decision variable for the second-choice berth's usage period. M×N = binvar(M,N,'full'). Where binvar indicates that the decision matrix is ​​a 0-1 variable, M and N represent the dimensions of the decision matrix, and 'full' indicates that the decision matrix is ​​an asymmetric matrix.

[0213] S421. Set the coefficient ω to 0.0001 to ensure that the disturbance term does not affect the maximum social welfare. Constrain the disturbance term Δ according to the following formula.

[0214]

[0215] S422. The unit price bidding model can be directly solved using the branch and bound method to determine the supplier's berth allocation, as shown below. Figure 2 As shown. The numbers in parentheses in the Supplier ID column represent the supplier's bidding hourly rental price. The striped areas represent the supplier's non-rental periods for berths, while blank areas represent periods available for rental. The shaded areas indicate the demander allocation; the first number represents the demander's ID, and the first item in parentheses represents the demander's bidding hourly usage price.

[0216] S423. The lump-sum bidding model can be transformed into an M-order linear programming problem to search for a local optimum. The solution yields the following berth allocation for suppliers: Figure 3 As shown. The value in parentheses in the Supplier ID column represents the supplier's total bidding price for leasing. The striped areas indicate non-leasing periods for the supplier's berths, while blank areas indicate periods available for leasing. The shaded areas indicate the allocation of demanders; the first number represents the demander's ID, and the first item in parentheses represents the demander's hourly bidding price.

[0217] S51, Platform commission rate θ = 50%, W all (Unit price) = 210, Wp (unit price) = 105, W all (total price) = 236.72, W p (total price) = 118.36.

[0218] S52, calculate W s (unit price) = 62.05, W d (unit price) = 42.95, wherein, I (unit price) = 0.3, J (unit price) = 0.43333; W s (total price) = 63.13, W d (total price) = 55.23, wherein I (total price) = 0.35, J (total price) = 0.4.

[0219] S53, the supplier utility, demander utility in the allocation unit price, total price bidding mode is shown in Table 7 and Table 8.

[0220] Table 7

[0221]

[0222]

[0223] Table 8

[0224]

[0225]

[0226] S54, according to The supplier actual income, demander actual payment cost in the allocation unit price, total price bidding mode is shown in Table 9 and Table 10.

[0227] Table 9

[0228]

[0229] Table 10

[0230]

[0231]

[0232] The application provides a new idea for optimizing parking space sharing and improving the willingness of citizens to participate in the sharing market. Through the design of a parking space sharing bilateral auction mechanism, on the one hand, parking space suppliers with different travel purposes, different education levels and different occupations can independently set the parking space rental price according to their own conditions, so as to maximize the parking space supply quantity and improve the problem of parking space shortage in the area; on the other hand, parking space demanders can independently set the parking space use price according to their own importance of obtaining the shared parking space, and obtain the parking space use right through bidding with other parking space users, so as to maximize the number of parking space demanders and ensure the benefits of parking space suppliers and parking platforms.

[0233] Those skilled in the art will realize that the embodiments described herein are for the purpose of helping the reader understand the principles of the application and should be understood as not limiting the scope of protection of the application to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations according to the technical inspiration disclosed in the application without departing from the essence of the application, and these modifications and combinations are still within the scope of protection of the application.

Claims

1. A method for allocating parking spaces based on an auction mechanism, characterized in that, The method comprises the following steps: S1: obtaining the parking space sharing interval, hourly rental price and total rental price of a parking space provider, and generating a parking space supply matrix according to the parking space sharing interval; S2: obtaining the first preferred parking space usage interval, the second preferred parking space usage interval and the hourly usage price of a parking space demander, and generating a first preferred parking space demand matrix and a second preferred parking space demand matrix according to the first preferred parking space usage interval and the second preferred parking space usage interval; S3: generating a bid hourly rental price and a bid total rental price according to the hourly rental price and the total rental price, and generating a bid hourly usage price according to the hourly usage price; S4: performing shared parking space allocation according to the parking space supply matrix, the first preferred parking space demand matrix, the second preferred parking space demand matrix, the bid hourly rental price, the bid hourly usage price and the bid total rental price; In the step S2, the generation of the first preferred parking space demand matrix and the second preferred parking space demand matrix comprises the following sub-steps: S21: obtaining a first-preferred parking space use starting time of the parking space demander , a second-preferred parking space use starting time , a first-preferred parking space use ending time , and a second-preferred parking space use ending time ; S22: Start time of use based on preferred berth and the end time of use of the preferred berth Generate preferred berth usage area ; based on the start time of use of the secondary berth and the termination time of secondary berth use Generate secondary berth usage area ; S23: a preferred berth use interval and a non-preferred berth use interval standardized to obtain a standardized preferred berth use interval and a standardized non-preferred berth use interval ; S24: Generating berth preferred demand matrix D and berth secondary demand matrix Ds according to the standardized berth preferred berth usage interval and the standardized berth secondary berth usage interval N×K .​​​ 2. The parking space allocation method based on an auction mechanism according to claim 1, characterized in that, In the step S1, the generation of the parking space supply matrix comprises the following sub-steps: S11: acquiring a parking space sharing time period of the parking space provider T , a parking space providing start time , and a parking space providing end time ; S12: generating a berth sharing time interval T is divided into K time windows, and a berth supply start time and a berth supply end time are generated ; S13: generating berth supply matrix S according to berth sharing interval and K time window M×K .

3. The parking space allocation method based on an auction mechanism according to claim 2, characterized in that, In the step S13, the berth supply matrix S M×K The calculation formula is: wherein, s ik denotes a 0-1 variable, denotes the start time of berth supply, denotes the end time of berth supply, Z denotes a set of integers, t denotes a partitioned time interval, i = 1, 2,..., M , k = 1, 2,..., K , M denotes the number of participants in the auction, K denotes the number of time windows.

4. The parking space allocation method based on an auction mechanism according to claim 1, characterized in that, In the step S23, the standardized preferred berth use interval The calculation formula is: wherein, denotes a lower limit of the standardized preferred berth usage interval, denotes an upper limit of the standardized preferred berth usage interval, denotes a preferred berth usage start time, denotes a preferred berth usage end time; Standardized secondary berth usage interval The calculation formula is: wherein, denotes the lower limit of the standardized secondary berth usage interval, denotes the upper limit of the standardized secondary berth usage interval, denotes the secondary berth usage start time, denotes the secondary berth usage end time; In the step S24, the berth preferred demand matrix D N×K The calculation formula is: wherein d jk denotes a 0-1 variable; berth sub-selection demand matrix The calculation formula is: wherein denotes a 0-1 variable, k = 1,2,..., K , K denotes the number of time windows, Z denotes a set of integers.

5. The method of claim 1, wherein, In the step S3, the bidding lease hour price The calculation formula is: wherein, represents the hourly rental price, represents the number of supplier defaults, η represents the penalty factor; In the step S3, the total price of the bidding lease The calculation formula is: wherein, represents the total price of the lease; In the step S3, the bidding uses the hourly price The calculation formula is: where, represents the hourly usage price, represents the number of demander violations.

6. The method of claim 1, wherein, The step S4 comprises the following sub-steps: S41: generate a berth allocation matrix σ of the first demand M×N and a berth allocation matrix σ of the second demand ; S42: Assigning the berth according to the berth allocation matrix σ of the first choice demand M×N , the berth allocation matrix σ of the second choice demand , the berth supply matrix, the berth first choice demand matrix, the berth second choice demand matrix, the bidding lease hour price, the bidding use hour price and the bidding lease total price, constructing the unit price bidding allocation mode and the total price bidding allocation mode, and determining the disturbance factor of the bidding allocation mode; S43: calculating the revenue of the unit price bid allocation mode and the revenue of the total price bid allocation mode according to the disturbance factor of the bid allocation mode, and completing the shared parking space allocation.

7. The parking space allocation method based on an auction mechanism according to claim 6, characterized in that, In the step S41, the berth allocation matrix σ of the first demand is allocated M×N The calculation formula is: wherein δ ij denotes a 0-1 variable; In the step S41, the berth allocation matrix of the second selected demand The calculation formula is: wherein denotes a 0-1 variable; In the step S42, the constraint conditions of the unit price bid allocation mode comprise a first constraint condition, a second constraint condition and a third constraint condition; and the constraint conditions of the total price bid allocation mode comprise the first constraint condition, the second constraint condition, the third constraint condition and a fourth constraint condition; The expression of the first constraint condition is ; The expression of the second constraint condition is wherein, M represents the number of suppliers participating in the auction, N represents the number of demanders participating in the auction; The expression of the third constraint condition is wherein, K denotes the number of time windows, s ik denotes a 0-1 variable, d jk denotes a 0-1 variable, denotes a 0-1 variable; The expression of the fourth constraint condition is ; In the step S42, the calculation formula of the disturbance factor Δ of the bid allocation mode is: wherein, indicates that the parking space provider with low bid and long sharing time is preferentially matched with the long parking demander, indicates that the first preferred parking interval of the demander is preferentially allocated when both the first preferred parking interval and the second preferred parking interval of the demander meet the requirements, ω indicates a minimum coefficient, φ i indicates that the provider with low bid and long sharing time is preferentially matched, indicates the bidding lease hour price; In the step S43, the bid allocation revenue max of the single bid allocation mode is allocated W all1 The calculation formula is: wherein, represents the price per hour of bidding usage, t represents the time interval, T represents the parking sharing period; In the step S43, the total price bidding distribution mode bidding distribution benefit max W all2 The calculation formula is: wherein, y i denotes a 0-1 variable, denotes the total price of the competitive lease.

Images