CN107705110A

CN107705110A - Fund pool based on multiple target pays sequence optimisation method and device

Info

Publication number: CN107705110A
Application number: CN201710908475.4A
Authority: CN
Inventors: 陈宇; 陈鹏; 熊伟; 汪宁; 芦帅; 谢伟良
Original assignee: Hangzhou Pingpeng Intelligent Technology Co Ltd
Current assignee: Hangzhou Ping Pong Intelligent Technology Co ltd
Priority date: 2017-09-29
Filing date: 2017-09-29
Publication date: 2018-02-16
Anticipated expiration: 2037-09-29
Also published as: CN107705110B

Abstract

The present invention proposes that a kind of fund pool based on multiple target pays sequence optimisation method, system and computer-readable recording medium, equipment, this method and included：The data to be paid of fund pool will be passed in and out in a period and to account data by obtaining；Constraints is limited to when being less than or equal to the payment the latest per fund with actual delivery time of the fund pool per fund, and so that in whole payment process, total amount of the fund is minimum and the total payoff time is soon most optimization aim for paying for first in fund pool, establishes object function；On the premise of constraints is met, one fund pool of generation pays the initial solution of sequence, and the initial solution for being paid sequence to fund pool based on object function is optimized, and obtains the optimal solution set that fund pool pays sequence；Pay for amount first according to default optimal solution is chosen from optimal solution set；The optimal solution that sequence is paid according to fund pool is paid.Solve in existing Third-party payment business, domestic trade company requires the quick contradiction paid for first to account and Third-party payment platform fund between pressure.

Description

Multi-objective-based capital pool payment sequence optimization method and device

Technical Field

The invention relates to the technical field of financial data processing, in particular to a multi-target-based capital pool payment sequence optimization method and system, and a computer-readable storage medium and device.

Background

The third party payment service is an independent mechanism with certain strength and credit guarantee, and adopts a mode of signing a contract with each big bank to provide a network payment mode of a transaction support platform with a bank payment settlement system interface for merchants and consumers. The third party payment service provider cooperates with banks, operators, certification organizations and the like, and provides personalized payment settlement service and marketing value-added service for enterprises and individual users on the basis of the payment settlement function of the banks.

Compared with the third-party payment service, the core element of the third-party cross-border payment service lies in the cross-border payment of the payment, and is a service for carrying out centralized receipt and payment and related settlement and remittance on foreign exchange funds of buyers and sellers in and out the country. Although the buyer and the seller of the third-party cross-border payment service are in different countries or regions, the essence of the transaction mode is not changed, and the third-party cross-border payment mechanism still serves as an intermediary for the transaction of the buyer and the seller, so that credit guarantee and a transaction channel are provided for the smooth completion of the transaction.

In exporting e-commerce and trade, a third party payment mechanism mainly provides remittance payment services. The remittance payment refers to RMB settlement payment service provided by a third party payment enterprise for foreign currency payment income of domestic merchants outside the country, wherein currency exchange and payment processes are completed by an escrow bank. The procedure of collection and payment is as follows:

the domestic merchant initiates a withdrawal application to the third-party payment company, the third-party payment company withdraws the corresponding money to the bank account opened abroad, then the withdrawal money is remitted to the bank account opened domestic after compliance examination is carried out by the foreign bank, and finally the third-party payment company carries out foreign currency exchange on foreign currency funds arriving at domestic and carries out RMB settlement for the domestic merchant at appointed time in a cooperative bank.

For domestic merchants, the payment can be paid after the completion of complex auditing and monitoring processes of platform, bank and third party payment through a plurality of settlement links of a plurality of institutions, so that the fund transfer is slow due to the complex processes, each link bears certain cost, and the enterprise operating cost is increased; meanwhile, the long and low-efficiency fund clearing and settlement process causes great pressure on cross-border merchants, brings certain exchange rate risk, seriously influences the use efficiency of the fund, prevents the merchants from expanding the operation scale and increases the operation risk of the cross-border merchants.

In order to solve the problems, the third-party payment company provides various advance payment schemes for high-quality merchants so as to shorten the time of payment due to remittance and improve cash flow of the merchants. However, excessive capital investment may cause a huge cash flow pressure on the third party payment platform, increasing the operation cost of the third party payment company.

Therefore, the demand of the domestic merchant for the account time and the amount of the third party paying funds are in a contradictory relationship. However, in the prior art, no technique is disclosed for optimizing the payment sequence to resolve the conflict.

Disclosure of Invention

The invention aims to solve the technical problem of providing a multi-target-based capital pool payment sequence optimization method, a multi-target-based capital pool payment sequence optimization system, a computer-readable storage medium and a multi-target-based capital pool payment sequence optimization device, and solving the contradiction between quick account arrival required by domestic merchants and capital payment pressure of a third-party payment platform in the conventional third-party payment service.

In order to solve the problems, the invention provides a multi-objective-based funding pool payment sequence optimization method, which comprises the following steps:

s1: acquiring data to be paid and account arrival data which enter and exit a fund pool within a period of time, wherein the data to be paid at least comprises the latest payment time limit of each fund to be paid;

s2: establishing an optimized objective function of a capital pool payment sequence by taking the latest payment time limit of each capital which is less than or equal to the actual payment time of each capital of the capital pool as a constraint condition and taking the minimum gross payment capital and the fastest total payment time of the capital pool as optimization objectives in the whole payment process;

s3: generating an initial solution of a fund pool payment sequence on the premise of meeting the constraint condition, and optimizing the initial solution of the fund pool payment sequence on the basis of the objective function to obtain an optimal solution set of the fund pool payment sequence; the fund pool payment sequence comprises a payment time sequence in the time period and a corresponding payment amount sequence;

s4: selecting an optimal solution from the optimal solution set according to a preset payment limit;

s5: and paying according to the optimal solution of the fund pool paying sequence.

According to one embodiment of the invention, the objective function of the fund pool payout sequence optimization is:

wherein,is a payment sequence; t is_leadFor a time period D after the time k_TIn the whole payment process, the sum of the payment time of all funds to be paid in advance; correspondingly, P is the total fund amount required for the payment from the fund pool to the outside in the whole payment process.

According to one embodiment of the invention, a time period D after time k_TIn the whole payment process, the sum T of the payment time advanced by all funds to be paid_leadExpressed by the following formula (2):

wherein, t_lead(l) The payment time advanced for each outward payment of funds,

D_k(l) The latest payment time limit for the first to-be-paid funds,actual payment time for the first fund to be paid; m is the total number of funds to be paid.

According to one embodiment of the invention, a time period D after time k_TThe total fund amount P required to be paid in the fund pool in the whole payment process is expressed by the following formula (4):

p (i) the amount of funds required to be paid per fund to be paid, and m is the total number of funds to be paid.

According to one embodiment of the invention, the step of calculating p (i) comprises:

for a period D after the time k_TInternal and fund flow payment sequenceIn the actual payment time seriesAnd account time sequence T ═ T_k(1)，T_k(2)，T_k(3)，...，T_k(n-1)，T_k(n)]Rearranging the data according to ascending time sequence to obtain a combined time sequence

ComputingWhen in useIs less thanWhen p (i) is 0;

wherein a is the actual payment timeBefore, the last capital payment needing to be paid for is paid in the actual payment time sequenceThe order of (1); j is the last capital payment to be paidIn between, the first pending funds to be credited in the crediting time series T_kThe order of (1); z is the last capital payment to be paidIn between, the last pending account fund is in the account time sequence T_kThe order of (1); o is_k(l) A payment amount for the first fund to be paid; i is_k(f) And f, the amount of the account to be paid is the f.

According to an embodiment of the present invention, the step S3 includes the steps of:

s31: the number of iterations g is initially set, and a time period D after the time k is generated_TTime series of initial paymentsThe corresponding payment sequence is

S32: time-series of the initial paymentRearranging the time sequence and the account time sequence according to the ascending time sequence to obtain a brand new combined time sequence

S33: based onCalculating a sequence of paymentsNext, a period D after the time k_TIn accordance withThe sum T of the payment time advanced by all the funds to be paid in the whole process of payment_lead0Reciprocal of (a), and total amount of funds P required to be paid in the fund pool₀；

S34: will be provided withAs an initial optimal solution for the payout sequence, put into the optimal solution set S of the payout sequence of the fund pool_bestIn and (2) mixingAssigning the current solution S;

s35: updating the iteration number g, for each element of the current solution SPerforming mutation operation to obtain new productPayment sequence ofPayment sequenceEach element ofFrom each element of the current solution SObtaining after mutation;

s36: actual payment time obtained by variationComposed payment time seriesRearranging the time sequence and the account time sequence according to the ascending time sequence to obtain a brand new combined time sequence

S37: based onCalculating pay time series obtained at varianceNext, a period D after the time k_TIn accordance withThe sum T of the payment time advanced by all the funds to be paid in the whole process of payment_lead1Reciprocal of (a), and total amount of funds P required to be paid in the fund pool₁；

S38: will pay forTime seriesAnd the optimal solution set S before optimization_bestIn the method, Pareto domination comparison is carried out under two dimensions of the sum of the gross fund amount of the payment and the advanced payment time, the updating of an optimal solution set is carried out, and the payment sequence is carried outAssigning the current solution S;

s39: judging whether the iteration times g reach a preset value, if not, returning to the step S35, and continuing to perform iteration optimization; otherwise, the optimal solution set is output, and step S4 is executed.

According to an embodiment of the present invention, the step S4 includes:

s41: according to the total fund amount P required to be paid in the fund pool and the sum T of the payment time advanced by all funds to be paid_leadEstablishing a coordinate system according to the relation between the inverses;

s42: set the optimal solution S_bestDrawing Pareto front edges through two-dimensional planes by all solutions in the solution;

s43: drawing a straight line x ═ gamma A perpendicular to coordinate axes, cutting off the Pareto front edge, and obtaining the total fund amount closest to the intersection pointThe sequence of (a) is used as an optimal solution, wherein gamma is a preset capital payment margin coefficient, and A is an expected payment limit.

According to an embodiment of the invention, in the step S31, an initial payment time sequenceFor random generation, the generation formula of each payment time is as follows:

wherein Random (0, 1) is [0, 1]]Random real numbers in between; d_k(l) The latest payment time limit for the first to be paid funds.

According to an embodiment of the invention, in the step S35, the payment sequenceEach element ofByThe mutation step comprises:

let R2 Random (0, 1) -1, i.e. R is a Random real number between [ -1, +1], calculated according to the following formula:

wherein Random (0, 1) is [0, 1]]Random real numbers in between; d_k(1) The latest payment time limit for the first to be paid funds.

According to an embodiment of the present invention, the step S38 includes:

suppose that the optimal solution set S at this time_bestThe Pareto frontier formed by o optimal solutions is totally contained in the solution, and the single optimal solution is recorded asq∈[1，...o]Will pay time seriesAnd the optimal solution set S before optimization_bestThe total amount of funds paid in the course of the payment and the advance paymentPerforming Pareto domination comparison under two dimensions of the sum of time, updating the optimal solution set by adopting the following principle, and updating the optimal solution set by adopting the following principle:

for a q from 1 to o,

increasing the iteration number g and paying the sequenceThe current solution S is given.

According to one embodiment of the invention, the data to be paid at least comprises the number of funds to be paid, the latest payment time limit of each fund to be paid and the payment amount of each fund to be paid; the account data at least comprises the number of the funds to be accounted, the account time of each fund to be accounted and the account amount of each fund to be accounted.

The invention also provides a multi-objective-based capital pool payment sequence optimization system, which comprises:

the data acquisition module is used for acquiring data to be paid and account arrival data which enter and exit the fund pool within a period of time, wherein the data to be paid at least comprises the latest payment time limit of each fund to be paid;

the target function establishing module is used for executing a target function which takes the latest payment time limit of each fund as a constraint condition, wherein the actual payment time of each fund in the fund pool is less than or equal to the latest payment time limit of each fund, and in the whole payment process, the target function which is optimized by the fund pool payment sequence is established by taking the minimum gross payment fund amount and the fastest total payment time in the fund pool as optimization targets;

the target optimization module generates an initial solution of the fund pool payment sequence on the premise of meeting the constraint condition, and optimizes the initial solution of the fund pool payment sequence on the basis of the target function to obtain an optimal solution set of the fund pool payment sequence; the fund pool payment sequence comprises a payment time sequence in the time period and a corresponding payment amount sequence;

the optimal solution selection module is used for selecting an optimal solution from the optimal solution set according to a preset payment limit;

and the payment module executes payment according to the optimal solution of the fund pool payment sequence.

The present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a multi-objective based funding pool payment sequence optimization method as described in any of the preceding embodiments.

The present invention also provides a computer apparatus comprising a memory and a processor and a computer program stored on the memory and being invoked by the processor, when executing the computer program, implementing a method of multi-objective based funding pool payment sequence optimization as claimed in any one of the preceding embodiments.

After the technical scheme is adopted, compared with the prior art, the invention has the following beneficial effects:

according to the data to be paid and the account arrival data in a certain time period, a payment sequence and payment amount backing optimization method is adopted, so that on the premise that constraint conditions are met, when a third-party payment platform finishes each external payment according to the optimized payment sequence, the total fund payment amount is kept in a preset amount, and the payment time is as fast as possible, and the beneficial effects of providing high-quality quick account arrival service for merchant customers and reducing the cost and pressure of the payment backing amount are achieved;

the method selects a multi-objective optimization algorithm to carry out iterative optimization on the capital payment sequence, does not need cross operation among chromosomes, can carry out high-efficiency global search on a solution space only through mutation operation, greatly improves the speed and efficiency of solving the optimization problem, and meanwhile, can effectively process the optimization problem with upper and lower limit constraints based on a bit variation strategy realized by real number coding.

Drawings

FIG. 1 is a schematic flow chart of a multi-objective based funding pool payout sequence optimization method according to an embodiment of the present invention;

FIG. 2 is a flowchart illustrating a process of target optimization according to an embodiment of the present invention;

fig. 3 is a schematic diagram of optimal solution selection according to an embodiment 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 detail below.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather construed as limited to the embodiments set forth herein.

Referring to FIG. 1, in one embodiment, a multi-objective based funding pool payout sequence optimization method includes the steps of:

s2: establishing an optimized objective function of a capital pool payment sequence by taking the latest payment time limit of each capital which is less than or equal to the actual payment time of each capital of the capital pool as a constraint condition, and taking the minimum gross payment capital and the fastest total payment time of the capital pool as optimization objectives in the whole payment process;

The fund pool payment optimization method according to the embodiment of the present invention is described in detail below, but should not be construed as being limited thereto.

In step S1, data to be paid and account data to be entered and exited into the fund pool within a time period are acquired, and the data to be paid at least includes the latest payment time limit of each fund. The specific length of the time period is not limited, but at least comprises a plurality of data to be paid and account-arriving data, and the data is used as input for optimization. The time period is obviously a time period of about to occur fund flow input and output to and from the fund pool, the payment condition of the fund pool in the time period is optimized this time, another time period can be replaced by sub-optimization, and the two time periods can be overlapped or not overlapped, and are not limited specifically.

Preferably, the data to be paid at least comprises the number of funds to be paid, the latest payment time limit of each fund to be paid and the payment amount of each fund to be paid; the account data at least comprises the number of the funds to be accounted, the account time of each fund to be accounted and the account amount of each fund to be accounted. Of course, the data to be paid and the billing data may also include other data information, without being limited thereto.

Specifically, assume that at time k, there are m cash withdrawal requests to be paid for a fund pool, and the latest payment period per withdrawal is D_k(1) The amount is O_k(l)，l∈[1,…m]Which isIn D_k(1) In ascending chronological order, i.e. D_k(1)≤D_k(2)≤D_k(3)≤…≤D_k(m-1)≤D_k(m) corresponding to O_k(l) And D_k(1) Sorted correspondingly.

A period D after the time k_TIn the method, n funds are required to be paid out, and the time for paying out is T_k(j) The amount of account is I_k(j),j∈[1，...n]Wherein T is_kIn ascending order, i.e. T_k(1)≤T_k(2)≤T_k(3)≤…≤T_k(n-1)≤T_k(n); i.e. for the third party payment system, a period of time D from time k into the future_TIn addition, the following two input and output fund flows are arranged in ascending time sequence and are respectively represented by table (1) and table (2):

TABLE (1) Fund flow S to be output_{Go out}

Table (2) to-be-input fund flow S_Into

In the embodiment of the invention, the fund flow S to be output_{Go out}And the fund flow S to be input_IntoAre known and established.

Then, step S2 is executed to establish a mathematical model of the payment sequence and the payment amount problem: and establishing an objective function for optimizing the payment sequence of the capital pool by taking the latest payment time limit of each capital which is less than or equal to the actual payment time of each capital of the capital pool as a constraint condition, and taking the minimum gross underlying payment capital and the fastest total payment time in the capital pool as optimization objectives in the whole payment process.

The invention needs to solve to obtain the current optimal scheduling time k toAfter a period of time D_TIn a payment sequence that minimizes the total amount of funds paid for the advance payment and maximizes the sum of the advance payment times throughout the payment processIn one embodiment, the goal of maximizing the sum of the upfront payment times is expressed in a form of minimizing the reciprocal thereof (i.e. when the reciprocal is minimum, the sum of the upfront payment times is maximum), and the above problem can be described by a unified minimization goal, but it should not be limited thereto, and of course, other expressions can be used instead.

In one embodiment, the objective function for the funding pool payout sequence optimization is:

Preferably, a time period D after the time k_TThe sum T of the payment time advanced by all the funds to be paid_leadExpressed by the following formula (2):

wherein, t_lead(l) The payment time advanced for each outward payment of funds,D_k(l) Is the first penThe latest payment time limit for the funds to be paid out,actual payment time for the first fund to be paid; m is the total number of funds to be paid.

Preferably, the total fund amount P required to be paid in the fund pool is expressed by the following formula (4):

Further, the step of calculating p (i) comprises:

first, for a period D after time k_TInternal and fund flow payment sequenceIn the actual payment time seriesAnd account time sequence T ═ T_k(1)，T_k(2)，T_k(3)，...，T_k(n-1)，T_k(n)]Rearranging the data according to ascending time sequence to obtain a combined time sequenceOf course, since the actual payment time series and the account arrival time series both have time attributes, they are arranged in ascending order according to the time, and of course, the two series have respective ordering.

Hypothesis combining time seriesCan be expressed by the following formula:

the above formula can also be expressed by other arrangement orders, and the expression of the p (i) mathematical general formula is not influenced.

In thatIn the sequence, forFor the external payment at the moment:

for inFor the external payment at the moment:

for payment pending itemIn a word:

from the above formula, it can be inferred thatEach of which isFinding the time corresponding to the last fund payment required to be paid before the last time, and comparing the payment time with the payment timeIf the total amount of the account item is less than the total amount of the payment item in the period of time, the payment needs to be carried out, and the payment carrying amount is the difference between the total amount of the payment item to be carried out and the total amount of the payment item to be carried out; otherwise, no padding is required.

Then, deducing according to the formula to obtain a calculation formula of p (i):

when in useIs less thanWhen p (i) is 0.

Wherein a is the actual payment timeBefore, the last capital payment needing to be paid for is paid in the actual payment time sequenceThe order of (1); j is the last capital payment to be paidIn between, the first pending funds to be credited in the crediting time series T_kThe order of (1); z is the last capital payment to be paidIn between, the last pending account fund is in the account time sequence T_kThe order of (1); o is_k(l) A payment amount for the first fund to be paid; i is_k(f) And f, the amount of the account to be paid is the f.And the actual payment time when the ith to-be-paid fund is paid.

As can be appreciated, the first and second,the following constraints should be satisfied:

and enabling the objective function to meet the constraint condition that the actual payment time of each fund in the fund pool is less than or equal to the latest payment time limit of each fund, namely the withdrawal requirement of the customer must be met within the commitment time, wherein the constraint is hard.

Executing step S3, generating an initial solution of the funding pool payment sequence on the premise that the constraint condition is satisfied, and optimizing the initial solution of the funding pool payment sequence based on the objective function to obtain an optimal solution set of the funding pool payment sequence; the fund pool payment sequence comprises a payment time sequence in the time period and a corresponding payment amount sequence. The initial payment sequence satisfying the constraint condition can be generated randomly, and iterative optimization is carried out on the initial payment sequence based on an objective function by using a multi-objective optimization algorithm.

Further, referring to fig. 2, the step S3 includes the following steps S31-S39:

s31: the initial iteration number g may be set to be 0, and a payment time series within a period of time after the time k is generated as the initial payment time seriesThe corresponding payment sequence is(also including a sequence of payment amounts, which may be considered to be associated with a sequence of payment times); wherein each payment time isThe time satisfies the constraint of equation (7).

Preferably, the initial payment time seriesFor random generation, the generation formula of each payment time is as follows:

wherein Random (0, 1) is [0, 1]]Random real numbers in between; d_k(1) The latest payment time limit for the first to be paid funds. As is known from the principle of equation (7),the requirement of the constraint in equation (7) can be satisfied.

S32: the initial payment time series obtained based on the formula (8)Pay time for each timeAnd the time T of account arrival_k(j) Of the formula T ═ T_k(1)，T_k(2)，T_k(3)，...，T_k(n-1)，T_k(n)]The account-arriving time sequence is rearranged according to the ascending time sequence to obtain a brand-new combined time sequence

In thatTo each oneThe time corresponding to the last fund payment needing to be paid before is in sequenceThe order in (1) is determined, that is, a in formula (6) is determined, and the final capital payment for payment is followed toThe charging time corresponding to the first charging is T_kOrdering j in the sequence, after the last payment of the capital to be paidIn the meantime, the charging time corresponding to the last charging is T_kThe ordering z in the sequence is determined.

S33: based onCalculating a sequence of paymentsNext, a period D after the time k_TIn accordance with the payment sequenceThe sum T of the payment time advanced by all the funds to be paid in the whole process of payment_lead0Reciprocal of (a), and total amount of funds P required to be paid in the fund pool₀。

Based onAccording to the formulas (1), (2), (3), (4) and (6), calculating in the payment sequenceNext, a period D from time k to the next_TThe sum T of the payment time advanced by all the funds to be paid_lead0And the total fund amount P of the required payment in the fund pool₀：

S34: will be provided withAs an initial optimal solution for the payout sequence, put into the optimal solution set S of the payout sequence of the fund pool_bestIn and (2) mixingThe current solution S is given.

Can be combined withAnd storing the initial optimal solution into an external archive, and after subsequent iteration optimization, performing Pareto comparison with an optimization result.

S35: the number of iterations g is updated, e.g. g may be increased by 1 for each element of the current solution SCarrying out variation operation to obtain a brand new payment sequencePayment sequenceEach element ofFrom each element of the current solution SObtained after mutation.

Preferably, in step S35, the payment sequenceEach element ofByThe mutation step comprises:

wherein Random (0, 1) is [0, 1]]Random real numbers in between; d_k(l) The latest payment time limit for the first to be paid funds. Obtained by the above formulaIt is still guaranteed that the requirements of the constraint in equation (7) are met.

Actual payment time obtained by variationComposed time payment sequenceAnd the time T of account arrival_k(j) The composed sequence T_k(1)，T_k(2)，T_k(3)，...，T_k(n-1)，T_k(n) arranging in ascending order to obtain a new time series

Can be derived inTo each oneThe time corresponding to the last fund payment needing to be paid before is in sequenceIn the order, after the last payment of the funds to be paidThe first time between them is entered at T_kThe order in the sequence, and the last payment of funds to be paid uponLast account is entered at T_kThe order in the sequence.

S37: based onCalculating pay time series obtained at varianceNext, a period D after the time k_TIn accordance withThe sum T of the payment time advanced by all the funds to be paid in the whole process of payment_lead1Reciprocal of (a), and total amount of funds P required to be paid in the fund pool₁。

Based onCalculating the actual payment time sequence obtained in the variation according to the formulas (1), (2), (3), (4) and (6)Next, a period D from time k to the next_TSum T of payment time advanced by all funds to be paid_lead1And the total fund amount P of the required payment in the fund pool₁。

S38: time series of paymentsAnd the optimal solution set S before optimization_bestIn the method, Pareto domination comparison is carried out under two dimensions of the sum of the gross fund amount of the payment and the advanced payment time, the updating of an optimal solution set is carried out, and the payment sequence is carried outThe current solution S is given.

Preferably, step S38 includes:

suppose that the optimal solution set S at this time_bestThe Pareto frontier formed by o optimal solutions is totally contained in the solution, and the single optimal solution is recorded asq∈[1，...o]Will pay time seriesAnd the optimal solution set S before optimization_bestIn the method, Pareto domination comparison is carried out on each single optimal solution under two dimensionalities of the sum of the gross fund amount of the payment and the payment time in advance, the optimal solution set is updated by adopting the following principle, and the optimal solution set is updated by adopting the following principle:

for a q from 1 to o,

increasing the iteration number g and paying the sequenceThe current solution S is given. The updating of the external archive is completed.

The iteration times can be preset according to the calculation time of the computer and the requirements of the computer, if the iteration times reach, the iteration optimization is ended, otherwise, the current solution S is optimized, and the iteration optimization is continued.

And then, executing step S4, selecting an optimal solution from the optimal solution set according to a preset payment limit.

Preferably, step S4 includes:

s41: according to the total fund amount P required to be paid in the fund pool and the advance payment of all funds to be paidSum of time of payment T_leadEstablishing a coordinate system according to the relation between the inverses;

Specifically, assuming that the expected fund payment amount is A, as shown in FIG. 3, the sum T of the payment time advanced by all the funds to be paid is taken as the x-axis of the total fund amount P required to be paid in the fund pool_leadEstablishing a coordinate system by taking the reciprocal of the optimal solution set S as the y axis_bestAll solutions in (a) draw Pareto fronts through two-dimensional planes.

Because of the need to control capital risk, a capital underpaying margin coefficient is usually set so that the underpaying amount is controlled within a certain range. Assuming that gamma is a preset capital payment margin coefficient, generally taking 0.8, drawing a straight line x which is gamma A and perpendicular to a coordinate axis, and cutting off a Pareto front edge, the straight line is closest to the intersection point and is closest to the intersection pointThe optimal solution of the scheme is the optimal solution of the scheme.

Then, step S5 is executed to pay according to the optimal solution of the fund pool payment sequence. The optimal solution payment sequence obtained in step S4 may be issued to a payment system to perform payment.

And the fund pool payment system carries out external payment according to the payment time sequence in the optimal solution payment sequence and the corresponding payment amount sequence, so that the payment aging is greatly accelerated while the total fund payment amount is controlled within the preset amount range in the whole process of carrying out payment according to the payment sequence.

The technical solution is further explained below with reference to a specific embodiment:

if 8, 15 and k days are the time, the total payment amount is 3000 ten thousand RMB, and the maximum total payment amount is 2400 ten thousand RMB for controlling the capital risk. The details of the money to be paid and the money to be billed in the half-month period from 16 to 30 days are as follows (3):

watch (3)

Randomly generating an initial payment sequence according to a formula (8), wherein Random (0, 1) is 0.8, and the generated initial payment time sequence, the amount of payment per pad, and the early arrival time per account correspond to the sequence to be paid and the sequence to be billed as shown in the following table (4):

watch (4)

The generated initial payment time meets the constraint condition of the formula (7), the sum of the total advanced payment time of external payment is 16.4 days, the sum of the advanced payment sum is 1700 thousands of RMB under the initial payment sequence in the period of 15 days to 30 days, and the coordinate point (1700, 1/16.4) is represented by the point E on figure 3.

Will be provided withI.e., the quantities shown in Table (5) below, as the initial optimal solutions, are stored into the optimal solution set S_bestIn and (2) mixingThe current solution S is given.

Watch (5)

Performing iterative optimization on the initial payment time sequence according to a formula (9), and obtaining the optimal payment time sequence, the payment amount per pad, the advanced payment time, the to-be-paid sequence and the to-be-billed sequence, which correspond to the following table (6):

watch (6)

As can be seen from the above table, in the optimized payout sequence obtained by optimizing the period of 15 days to 30 days, the sum of the total advance-to-account times of the outward payout is 29.52 days, the sum of the advance payment amounts is 2400 ten thousand, and the coordinate point (2400, 1/29.52) is indicated by point F in fig. 3.

Performing Pareto dominant comparison on the solution obtained by optimization and the optimal solution of the initial optimal solution set, wherein the solution obtained by optimization and the optimal solution of the initial optimal solution set are not dominant, and therefore, the solution obtained by optimization, namely the quantity in the following table (7) is used

Watch (7)

Storing the solution into an optimal solution set S_bestAnd assigning the payment sequence to the current solution S.

And (3) continuously carrying out iterative optimization on the optimization sequence according to a formula (9), wherein the obtained optimized payment time sequence, each payment amount and the advanced account-reaching time correspond to the sequence to be paid and the sequence to be checked in the following table:

watch (8)

From the above table, in the optimized payment sequence from 15 days to 30 days, the sum of the total advanced payment time of the external payment is 52.136 days, the sum of the advance payment sum is 3700 ten thousand of rmb, and the optimized solution and the optimal solution set S are obtained_bestThe non-inferior solutions in (1) do not mutually dominate, so the payment sequence obtained by further optimization is stored into the optimal solution set S_bestAnd the corresponding payment sequence is assigned to the current solution S, the coordinate point (3700, 1/52.136) is represented by point G on fig. 3.

In this embodiment, the maximum underpayable total amount is 2400 ten thousand RMB, and the underpayable total amount is 3700 ten thousand RMB and larger than the maximum underpayable total amount in the optimized sequence in Table (8), and the underpayable total amount in the payment sequence in Table (6)The sum is 2400 ten thousand, and is closest to the maximum chargeable total sum, so in this embodiment, the optimal solution S is_bestThe sequence is represented by table (9) for the following payment sequence:

watch (9)

And the payment system carries out outward payment according to the optimized payment sequence, so that the total payment amount of the pad payment is 2400 ten thousand, and the sum of the time for reaching the account in advance is 29.52 days.

the objective function establishing module is used for establishing an objective function for optimizing a fund pool payment sequence by taking the latest payment time limit of each fund as a constraint condition, and taking the minimum gross payment fund amount and the fastest total payment time in the fund pool as optimization targets in the whole payment process;

For the specific contents of the system, the computer-readable storage medium and the computer device for optimizing the sequence of funding based on multiple objectives according to the present invention, reference may be made to the specific description of the method for optimizing the sequence of funding based on multiple objectives according to the foregoing embodiments, and further description is omitted here.

Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the scope of the claims, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention.

Claims

1. A multi-objective based funding pool payment sequence optimization method is characterized by comprising the following steps:

2. The multi-objective based funding pool pay sequence optimization method of claim 1, wherein an objective function of the funding pool pay sequence optimization is:

<mrow> <mi>O</mi> <mi>b</mi> <mi>j</mi> <mi>f</mi> <mrow> <mo>(</mo> <mover> <mi>S</mi> <mo>&RightArrow;</mo> </mover> <mo>)</mo> </mrow> <mo>=</mo> <mi>min</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>/</mo> <msub> <mi>T</mi> <mrow> <mi>l</mi> <mi>e</mi> <mi>a</mi> <mi>d</mi> </mrow> </msub> <mo>,</mo> <mi>P</mi> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>

wherein,is a payment sequence; t is_leadFor a time period D after the time k_TIn the whole payment process, the sum of the payment time of all funds to be paid in advance; correspondingly, P is paid for the wholeIn the process, the capital pool pays the total capital amount required by the payment.

3. The multi-objective based funding pool payment sequence optimization method of claim 2, wherein a time period D after time k_TIn the whole payment process, the sum T of the payment time advanced by all funds to be paid_leadExpressed by the following formula (2):

<mrow> <msub> <mi>T</mi> <mrow> <mi>l</mi> <mi>e</mi> <mi>a</mi> <mi>d</mi> </mrow> </msub> <mo>=</mo> <msubsup> <mi>&Sigma;</mi> <mrow> <mi>l</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>m</mi> </msubsup> <msub> <mi>t</mi> <mrow> <mi>l</mi> <mi>e</mi> <mi>a</mi> <mi>d</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>l</mi> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>

wherein, t_lead(l) The payment time advanced for each outward payment of funds,

4. The multi-objective based funding pool payment sequence optimization method of claim 2, wherein a time after time k isSegment D_TThe total fund amount P required to be paid in the fund pool in the whole payment process is expressed by the following formula (4):

<mrow> <mi>P</mi> <mo>=</mo> <msubsup> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>m</mi> </msubsup> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow>

5. The multi-objective based funding pool payment sequence optimization method of claim 4, wherein the step of calculating p (i) comprises:

ComputingWhen in useIs less thanWhen p (i) is 0;

wherein a is the actual payment timeBefore, the last capital payment needing to be paid for is paid in the actual payment time sequenceThe order of (1); j is the last capital payment to be paidIn between, the first pending funds to be credited in the crediting time series T_kThe order of (1); z is the last capital payment to be paidIn between, the last pending account fund is in the account time sequence T_kThe order of (1); o is_k(l) A payment amount for the first fund to be paid; i is_k(f) And the balance amount for the balance to be paid for the f-th fund.

6. The multi-objective based funding pool payment sequence optimization method of any one of claims 1-5, wherein the step S3 comprises the steps of:

s35: updating the iteration number g, for each element of the current solution SCarrying out variation operation to obtain a brand new payment sequencePayment sequenceEach element ofFrom each element of the current solution SObtaining after mutation;

S38: will be provided withPayment time seriesAnd the optimal solution set S before optimization_bestIn the method, Pareto domination comparison is carried out under two dimensions of the sum of the gross fund amount of the payment and the advanced payment time, the updating of an optimal solution set is carried out, and the payment sequence is carried outAssigning the current solution S;

7. The multi-objective based funding pool payment sequence optimization method of claim 6, wherein the step S4 comprises:

8. The multi-objective based funding pool payment sequence optimization method of claim 6, wherein in the step S31, an initial payment time sequenceFor random generation, the generation formula of each payment time is as follows:

9. The multi-objective based funding pool payment sequence optimization method of claim 6, wherein in the step S35, the payment sequenceEach element ofByThe mutation step comprises:

10. The multi-objective based funding pool payment sequence optimization method of claim 6, wherein the step S38 comprises:

suppose that the optimal solution set S at this time_bestThere are 0 optimal solutions to form Pareto frontier, and the single optimal solution is recorded asq∈[1，...0]Will pay time seriesAnd the optimal solution set S before optimization_bestIn the method, Pareto domination comparison is carried out on each single optimal solution under two dimensionalities of the sum of the gross fund amount of the payment and the payment time in advance, the optimal solution set is updated by adopting the following principle, and the optimal solution set is updated by adopting the following principle:

for a q from 1 to 0, the ratio,

11. The multi-objective fund pool payment sequence optimization method according to any one of the claims 1-10, wherein the data to be paid comprises at least the number of funds to be paid, the latest payment time limit for each fund to be paid and the payment amount for each fund to be paid; the account data at least comprises the number of the funds to be accounted, the account time of each fund to be accounted and the account amount of each fund to be accounted.

12. A multi-objective based funding pool payment sequence optimization system, comprising:

13. A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, implements the multi-objective based funding pool payment sequence optimization method of any one of claims 1-10.

14. A computer apparatus comprising a memory and a processor and a computer program stored on the memory and invokable by the processor, when executing the computer program, implementing the multi-objective based funding pool payment sequence optimization method of any of claims 1-10.