Method for generating a multi-factor financial customer service route based on security mechanisms

Abstract

This invention belongs to the field of route planning and discloses a multi-factor financial customer service route generation method based on a security mechanism. The method includes: S1, obtaining the locations of all financial customers requiring door-to-door service and storing each location as a node in a node set V; S2, obtaining an edge set E based on the node set V; S3, calculating the improved weight of each edge in the edge set E; S4, constructing an objective function based on the improved weights; and S5, optimizing the objective function using an improved ant colony algorithm to obtain the service route. This invention can effectively improve the practicality of the planned service routes and significantly increase the probability of completing door-to-door service tasks for a predetermined number of financial customers.

Description

Technical Field

[0001] This invention relates to the field of route planning, and more particularly to a method for generating multi-factor financial customer service routes based on security mechanisms. Background Technology

[0002] Before providing on-site services to financial clients, route planning is usually required. In existing technologies, the shortest path generation algorithm is generally used to plan the on-site service route before the on-site service. However, the planning usually only considers the distance between two financial clients that are adjacent in the on-site service sequence. The service route is obtained by minimizing the total distance of the routes to all financial clients, thereby minimizing the total time spent on on-site services and enabling on-site services to be provided to as many financial clients as possible in the same amount of time.

[0003] Obviously, this optimization method does not take into account dynamic factors such as weather and traffic conditions during door-to-door service. Therefore, the total time of the optimized service route may differ significantly from the actual total time spent visiting financial customers according to the route. This could result in a lower probability of completing the planned number of door-to-door service tasks on time when following the planned route. Summary of the Invention

[0004] The purpose of this invention is to disclose a method for generating multi-factor financial customer service routes based on security mechanisms, thereby solving the technical problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] This invention provides a method for generating multi-factor financial customer service routes based on security mechanisms, including:

[0007] S1: Obtain the locations of all financial customers awaiting door-to-door service, and store each location as a node in the node set V;

[0008] S2, obtain the edge set E based on the node set V;

[0009] S3, calculate the improved weight of each edge in the edge set E, including:

[0010] For edge (i,j), the formula for calculating its improved weight is:

[0011]

[0012] w ij Let d be the improved weight of edge (i,j). ij Let (i,j) be the length of the edge, where i and j represent nodes.ij and trf ij These represent the weather factor and traffic congestion factor on the side (i,j) of the day the service is provided; v is the preset speed.

[0013] S4, Construct the objective function based on the improved weights;

[0014] S5 uses an improved ant colony algorithm to optimize the objective function and obtain the service route.

[0015] Preferably, obtaining the edge set E based on the node set V includes:

[0016] Connect any two nodes in the node set V pairwise to obtain NV. 2 -NV edges, where NV represents the total number of nodes in the node set V;

[0017] NV 2 -NV edges are stored in edge set E.

[0018] Preferably, the length of edge (i,j) is the distance between nodes i and j.

[0019] Preferably, the process for determining the weather factors on the day of the on-site service is as follows:

[0020] Obtain the weather type for node i and node j on the day of the on-site service;

[0021] Determine the weather coefficients corresponding to nodes i and j based on the weather types of nodes i and j.

[0022] Calculate the weather factor for the edge (i,j) on the day of the door-to-door service based on the weather coefficients of node i and node j.

[0023] Preferably, the weather types include sunny, snowy, rainy, and sandstorm;

[0024] The weather coefficients for sunny days, snowy days, rainy days, and storms are 1, 1.2, 1.5, and 2, respectively.

[0025] Preferably, the weather factor for the day of the on-site service is calculated based on the weather coefficients of nodes i and j, including:

[0026] Calculate we using the following formula ij :

[0027] we ij =λ×cef i +(1-λ)×cef j

[0028] λ is the weather calculation weight, cef i and CEF jThese are the weather coefficients corresponding to nodes i and j, respectively.

[0029] Preferably, the process for determining the traffic congestion factors on the day of the door-to-door service is as follows:

[0030] Obtain historical traffic data;

[0031] Based on historical traffic data, traffic conditions are predicted for the road corresponding to edge (i,j) to obtain the traffic condition type of the road corresponding to edge (i,j);

[0032] Traffic congestion factors are determined based on traffic condition type.

[0033] Preferably, the traffic conditions include smooth traffic, light congestion, moderate congestion, and heavy congestion.

[0034] Preferably, the traffic congestion factor is determined based on the type of traffic condition, including:

[0035] The traffic congestion factors for smooth traffic, light congestion, moderate congestion, and heavy congestion are 1, 1.3, 1.8, and 2.5, respectively.

[0036] Preferably, the objective function is constructed based on the improved weights, including:

[0037] The objective function is as follows:

[0038]

[0039] miF is the objective function, RT is the set of edges that make up the service path, and w k For the improved weight of edge k, ord k Let k be the access order value of edge k in the service route; nRT represents the total number of edges in RT.

[0040] Beneficial effects:

[0041] Compared to existing technologies, this invention improves the edge weighting by considering not only distance but also traffic and weather conditions along the corresponding roads when calculating the weights. This allows the edge weights to take into account more real-world factors affecting travel speed, making the calculated service routes more practically applicable and avoiding excessive discrepancies between the actual total time taken to visit the destination and the theoretically calculated total time. Therefore, this invention effectively improves the practicality of the planned service routes and significantly increases the probability of completing on-site service tasks for a predetermined number of financial clients. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the multi-factor financial customer service route generation method based on security mechanisms of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0045] like Figure 1 As shown in one embodiment, the present invention provides a multi-factor financial customer service route generation method based on a security mechanism, including:

[0046] S1: Obtain the locations of all financial customers awaiting door-to-door service, and store each location as a node in the node set V.

[0047] In one implementation, the location of the financial customer requiring on-site service can be obtained from the database of the financial customer management system. The financial customer requiring on-site service can be one who needs on-site service within one day.

[0048] The database uses asymmetric encryption to encrypt and store financial customers' personal information (such as location and contact information), which effectively enhances the security of their personal information. Only encrypted data can be retrieved from the database.

[0049] Encryption is performed using the public key of an asymmetric encryption algorithm during storage.

[0050] After obtaining the encrypted location, the location of the financial client is decrypted using a private key obtained through an asymmetric encryption algorithm.

[0051] In the present invention, the asymmetric encryption algorithm may be the RSA algorithm. The private key may be managed and used by personnel with administrative permissions.

[0052] Furthermore, the encryption process of the RSA algorithm is as follows:

[0053] The first step is to generate keys:

[0054] Randomly generate two large prime numbers p1 and q1 (p1 ≠ q1); p1 and q1 are two distinct large prime numbers (usually more than 1024 bits);

[0055] Calculate the modulus n1: n1 = p1 × q1.

[0056] Calculate

[0057] denotes the number of integers less than n1 and relatively prime to n1;

[0058] Calculate the public key exponent e: Select an integer e that satisfies

[0059] Calculate the private key exponent d: Calculate using the extended Euclidean algorithm.

[0060] Public key: (e, n1), private key: (d, n1).

[0061] Destroy p1, q1,

[0062] The second step is to perform encryption:

[0063] Input: plaintext M (in integer form, satisfying 0 ≤ M < n1, convert the text to be encrypted into an integer, such as through ASCII or a padding scheme), public key: (e, n1), output: ciphertext C.

[0064] Encryption formula:

[0065] C ≡ M e mod n1.

[0066] Furthermore, the decryption process of the RSA algorithm is as follows:

[0067] Input: ciphertext C, private key: (d, n1).

[0068] Output: plaintext M.

[0069] Decryption formula:

[0070] M ≡ C d (mod n1).

[0071] In one implementation, after obtaining multiple locations, each location is numbered and treated as a node, thereby obtaining a node set V.

[0072] S2, obtain the edge set E based on the node set V.

[0073] After obtaining the set of nodes, it is necessary to construct the connection relationships between the nodes. Since the connection relationships between the nodes are not known at the beginning, this invention obtains all possible connection relationships between nodes by traversing, which makes it easier to calculate the weight of the edges between the nodes in advance. This allows the stored weights to be called directly during subsequent optimization, thereby improving the efficiency of optimization.

[0074] In one implementation, obtaining the edge set E based on the node set V includes:

[0075] Connect any two nodes in the node set V pairwise to obtain NV. 2 -NV edges, where NV represents the total number of nodes in the node set V;

[0076] NV 2 -NV edges are stored in edge set E.

[0077] The edge set E stores all possible edges. Therefore, selecting multiple edges from the edge set E can form a service route.

[0078] S3, calculate the improved weight of each edge in the edge set E, including:

[0079] For edge (i,j), the formula for calculating its improved weight is:

[0080]

[0081] w ij Let d be the improved weight of edge (i,j). ij Let (i,j) be the length of the edge, where i and j represent nodes. ij and trf ij These represent the weather factor and traffic congestion factor at point (i,j) on the day of the on-site service, respectively; v is the preset speed.

[0082] Existing technologies generally only consider the length of the edge when calculating the weight of the edge. Therefore, the obtained service route is only the shortest route in terms of geographical distance, but there is a high probability that it is not the route with the shortest time.

[0083] Therefore, this invention improves the weights of the edges to obtain improved weights, thereby making the final service routes more realistically valuable for reference.

[0084] In one implementation, the preset speed can be 60 km / h.

[0085] In one implementation, the length of edge (i,j) is the distance between nodes i and j, which can be obtained by querying the shortest driving distance between nodes i and j using map query software.

[0086] In one implementation, the process for determining the weather factors for the day of the on-site service is as follows:

[0087] Obtain the weather type for node i and node j on the day of the on-site service;

[0088] Determine the weather coefficients corresponding to nodes i and j based on the weather types of nodes i and j.

[0089] Calculate the weather factor for the edge (i,j) on the day of the door-to-door service based on the weather coefficients of node i and node j.

[0090] On the day of the on-site service, the weather type for each location can be obtained through the weather forecast. For example, the service route can be planned the day before the service, which makes the weather forecast more valuable.

[0091] Preferably, the weather types include sunny, snowy, rainy, and sandstorm;

[0092] The weather coefficients for sunny days, snowy days, rainy days, and storms are 1, 1.2, 1.5, and 2, respectively.

[0093] By setting different weather coefficients for different weather types, the weather coefficient can be increased for weather types that have a greater impact on vehicle speed.

[0094] Preferably, the weather factor for the day of the on-site service is calculated based on the weather coefficients of nodes i and j, including:

[0095] Calculate we using the following formula ij :

[0096] we ij =λ×cef i +(1-λ)×cef j

[0097] λ is the weather calculation weight, cef i and CEF j These are the weather coefficients corresponding to nodes i and j, respectively.

[0098] Since the weather types differ between different regions, by weighting the weather coefficients between two nodes, the calculated weather factors can more accurately reflect the weather conditions of edge (i, j).

[0099] In one implementation, the weather calculation weight can be 0.5.

[0100] In one implementation, the process for determining the traffic congestion factors on the day of the door-to-door service is as follows:

[0101] Obtain historical traffic data;

[0102] Based on historical traffic data, traffic conditions are predicted for the road corresponding to edge (i,j) to obtain the traffic condition type of the road corresponding to edge (i,j);

[0103] Traffic congestion factors are determined based on traffic condition type.

[0104] In one implementation, historical traffic data includes date type (weekday / weekend / holiday) and average vehicle speed (e.g., average speed is obtained by averaging measurements taken every hour during a set time period from 7:00 to 20:00).

[0105] Average vehicle speed can be obtained from publicly available data from traffic management departments.

[0106] Historical traffic data is compiled on a daily basis.

[0107] In one implementation, traffic conditions are predicted for the road corresponding to edge (i,j) based on historical traffic data, and the traffic condition type of the road corresponding to edge (i,j) includes:

[0108] Get the date type for the day the service is provided;

[0109] Get the reference vehicle speed for this date type within the past month;

[0110] The type of traffic condition on the road is determined based on the reference vehicle speed.

[0111] In one implementation, obtaining the reference vehicle speed for that date type within the most recent month includes:

[0112] For a date type p, store the average vehicle speed of date type p over the most recent month into the set UAp;

[0113] Use the following formula to obtain the reference vehicle speed spd corresponding to date type p. p :

[0114]

[0115] nuap is the total number of average vehicle speeds in UAP, and b is an element in UAP.

[0116] In one implementation, determining the road traffic condition type based on a reference vehicle speed includes:

[0117] Traffic conditions are categorized as smooth, lightly congested, moderately congested, and heavily congested.

[0118] If the reference speed is less than 10 kilometers per hour, the traffic situation is classified as severe congestion.

[0119] If the reference speed is greater than or equal to 10 km / h and less than 30 km / h, the traffic situation is classified as moderate congestion.

[0120] If the reference speed is greater than or equal to 30 km / h and less than 60 km / h, the traffic situation is classified as mild congestion.

[0121] If the reference speed is greater than or equal to 60 kilometers per hour, the traffic condition is considered smooth.

[0122] In one implementation, traffic condition types include smooth traffic, light congestion, moderate congestion, and heavy congestion.

[0123] In one implementation, determining a traffic congestion factor based on traffic condition type includes:

[0124] The traffic congestion factors for smooth traffic, light congestion, moderate congestion, and heavy congestion are 1, 1.3, 1.8, and 2.5, respectively.

[0125] S4, construct the objective function based on the improved weights.

[0126] Preferably, the objective function is constructed based on the improved weights, including:

[0127] The objective function is as follows:

[0128]

[0129] miF is the objective function, RT is the set of edges that make up the service path, and w k For the improved weight of edge k, ord k Let k be the access order value of edge k in the service route; nRT represents the total number of edges in RT.

[0130] The access order value for service routes is determined as follows:

[0131] The edges in the service route are numbered sequentially, with the edge appearing earlier in the service route having a smaller number. These numbers are then used as the access order values.

[0132] For example, for the service route v2-v5-v1-v7, the edge between nodes v2 and v5 is numbered 1. The edge between nodes v5 and v1 is numbered 2.

[0133] In addition to using improved weights, the objective function of this invention also introduces access order values ​​during its construction. This allows the objective function to optimize the uncertain data of weather factors in a more targeted manner, making the calculated service area routes more valuable for reference.

[0134] In the objective function, the larger the access order value of edge k, the smaller its impact on the calculation result of the objective function. This can reduce the impact of the uncertainty of weather forecast on the generation result of service route, because the longer the time since the weather forecast was generated, the worse the accuracy of the forecast.

[0135] S5 uses an improved ant colony algorithm to optimize the objective function and obtain the service route.

[0136] In one implementation, an improved ant colony algorithm is used to optimize the objective function to obtain service routes, including:

[0137] An improved ant colony algorithm is obtained by improving the pheromone evaporation coefficient.

[0138] The objective function is optimized using an improved ant colony algorithm to obtain service routes.

[0139] Traditional ant colony algorithms typically set the pheromone evaporation coefficient to a fixed value. This setting can easily lead to the pheromone evaporation coefficient becoming too large or too small during the iteration process.

[0140] If the value is too small, the pheromone will evaporate slowly, and the residual pheromone concentration in the early discovered paths (even if they are not optimal) will be too high. This will cause subsequent ants to concentrate too much in these paths, weakening their overall exploration capabilities.

[0141] For example, when the pheromone evaporation coefficient is 0.1, the algorithm requires more iterations to find the optimal solution and is prone to getting trapped in local optima. Residual pheromones inhibit the exploration of new paths, and the algorithm needs more time to break free from historical path dependence.

[0142] If the value is too high, the pheromone will evaporate quickly, resulting in insufficient accumulation of pheromones on high-quality paths. This leads to a lack of guidance in ant selection and an increase in blind searching.

[0143] Therefore, this invention improves the pheromone evaporation coefficient of the traditional ant colony algorithm.

[0144] In one implementation, the improved formula for calculating the pheromone volatility coefficient is as follows:

[0145]

[0146] ρ min and ρ max These represent the minimum and maximum values ​​of the set pheromone evaporation coefficient, respectively; t is the current iteration number, T is the total iteration number, η is the iteration acceleration factor, δ is the weight factor, Ω is the sensitivity coefficient, and gap is the iteration effect factor.

[0147] Adjusting the pheromone evaporation coefficient solely based on the number of iterations has significant limitations: it ignores the actual optimization progress of the algorithm (such as how close the current solution is to the optimal solution). By dynamically adjusting ρ by incorporating both the iteration progress and the solution quality gap, the algorithm's exploratory capability (global search) and exploitation capability (convergence speed) can be more accurately balanced.

[0148] The pheromone evaporation coefficient of this invention is a dynamically changing value. By comprehensively considering the cumulative number of iterations and the changes in the optimal solution, the pheromone evaporation coefficient increases as the cumulative number of iterations increases and the gap between the current optimal solution and the global optimal solution decreases, thus accelerating convergence. Conversely, the pheromone evaporation coefficient decreases as the cumulative number of iterations decreases and the gap between the current optimal solution and the global optimal solution increases, thereby improving the search capability of the ant colony algorithm and reducing the probability of getting trapped in local optima. Because of this invention...

[0149] In one implementation, the minimum and maximum values ​​of the pheromone evaporation coefficient are set to 0.1 and 0.7, respectively.

[0150] In one implementation, the iteration acceleration factor is 1.5. This allows ρ to grow faster in later stages, thus accelerating convergence.

[0151] In one implementation, the weighting factor is 0.4.

[0152] In one implementation, the sensitivity coefficient is set to 5.

[0153] In one implementation, the formula for calculating the iteration effect factor is:

[0154]

[0155] f current f is the objective function value of the optimal solution obtained in the most recent iteration; global The objective function value is the known global optimal solution.

[0156] Specifically, the known global optimal solution is the optimal solution obtained in all the iterations that have been performed.

[0157] The optimal solution is the one that minimizes the objective function value.

[0158] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-factor financial customer service route generation method based on security mechanisms, characterized in that, include: S1: Obtain the locations of all financial customers awaiting door-to-door service, and store each location as a node in the node set V; S2, obtain the edge set E based on the node set V; S3, calculate the improved weight of each edge in the edge set E, including: For edge (i,j), the formula for calculating its improved weight is: an improved weight for edge (i, j), a length of edge (i, j), i and j represent nodes, and are a weather factor and a traffic congestion factor of edge (i, j) on the day of the upper door service, respectively; v is a preset speed; The process for determining the weather factors on the day of the on-site service is as follows: Obtain the weather type for node i and node j on the day of the on-site service; Determine the weather coefficients corresponding to nodes i and j based on the weather types of nodes i and j. Based on the weather coefficients of nodes i and j, calculate the weather factor of the edge (i,j) on the day of the door-to-door service. Set different values ​​for the weather coefficients for different weather types, so that the weather type that has a greater impact on vehicle speed has a larger corresponding weather coefficient. S4. Construct the objective function based on the improved weights, where the objective function is as follows: miF is the objective function, RT is the set of edges that compose the service route, is the improved weight of edge k, is the access order value of edge k in the service route; nRT represents the total number of edges in RT; S5, the objective function is optimized using the improved ant colony algorithm to obtain the service route. The pheromone evaporation coefficient of the ant colony algorithm is improved to obtain the improved ant colony algorithm. The improved formula for calculating the pheromone volatility coefficient is as follows: and are the minimum and maximum values of the pheromone evaporation coefficient set respectively; t is the current iteration number, T is the total iteration number, is the iteration acceleration factor, is the weight factor, is the sensitivity coefficient, is the iteration effect factor.

2. The method for generating a multi-factor financial customer service route based on a security mechanism according to claim 1, characterized in that, Obtain the edge set E based on the node set V, including: Connect any two nodes in the node set V pairwise to obtain An edge, where NV represents the total number of nodes in the node set V; Will Each edge is stored in the edge set E.

3. The security mechanism based multi-factor financial customer service routing generation method as claimed in claim 1, wherein, The length of edge (i,j) is the distance between nodes i and j.

4. The security mechanism based multi-factor financial customer service routing generation method as claimed in claim 1, wherein, Weather types include sunny, snowy, rainy, and dust storm; The weather coefficients for sunny days, snowy days, rainy days, and storms are 1, 1.2, 1.5, and 2, respectively.

5. The security mechanism based multi-factor financial customer service routing generation method as claimed in claim 1, wherein, Based on the weather coefficients of nodes i and j, the weather factors for the day of the on-site service (i,j) are calculated, including: The following formula is used to calculate : Calculate the weights for weather conditions. and These are the weather coefficients corresponding to nodes i and j, respectively.

6. The security mechanism based multi-factor financial customer service routing generation method as claimed in claim 1, wherein, The process for determining traffic congestion factors on the day of the on-site service is as follows: Obtain historical traffic data; Based on historical traffic data, traffic conditions are predicted for the road corresponding to edge (i,j) to obtain the traffic condition type of the road corresponding to edge (i,j); Traffic congestion factors are determined based on traffic condition type.

7. The security mechanism based multi-factor financial customer service routing generation method as claimed in claim 6, wherein, Traffic conditions are categorized as smooth, lightly congested, moderately congested, and heavily congested.

8. The security mechanism based multi-factor financial customer service routing generation method of claim 7, wherein, Traffic congestion factors are determined based on traffic condition type, including: The traffic congestion factors for smooth traffic, light congestion, moderate congestion, and heavy congestion are 1, 1.3, 1.8, and 2.5, respectively.

Images