Enterprise cross-department data collaborative optimization method based on distributed architecture

Abstract

The invention discloses an enterprise cross-department data collaborative optimization method based on a distributed architecture, which comprises the following steps: S1, deploying distributed computing nodes in each department of an enterprise, and connecting the distributed computing nodes through a network to form a distributed architecture, S2, establishing a multi-subject game model; s3, executing a multi-stage optimization process through the multi-agent game model; s4, in a local optimization stage, independently optimizing a department strategy of each department based on the enterprise cross-department data updated in real time; s5, in a cross-department cooperation and competition stage, all departments are based on shared data; s6, in the global optimization stage, the distributed system continuously updates data of all departments and feeds back the data to the multi-subject game model in the system; s7, applying a dynamic programming algorithm to dynamically adjust the enterprise overall collaboration strategy based on the real-time data and the historical data; and S8, in the distributed computing nodes, monitoring strategy execution results of all departments. According to the method, the data collaboration capability and decision-making efficiency in an enterprise are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of cross-departmental data collaboration technology, and in particular to an enterprise cross-departmental data collaboration optimization method based on a distributed architecture. Background Art

[0002] In modern enterprise management, the demand for data sharing and collaboration is increasing, especially in the environment of cross-departmental collaboration. With the rapid development of information technology, traditional enterprise data management methods have gradually revealed their limitations. Existing technologies mainly rely on centralized data processing architectures, and each department often manages and uses data independently. Although this has improved the efficiency of data processing to a certain extent, it is difficult to achieve effective collaboration and information sharing between departments.

[0003] In the existing cross-departmental data sharing technology, a common practice is to bring together data from various departments by establishing an information sharing platform. The need for manual intervention in data integration and synchronization can easily lead to information lags and the emergence of data islands. When departments share data, there is a lack of unified standards and processes. Differences in data formats and structures also bring difficulties to the interaction and use of information. In addition, the security and privacy protection issues of information sharing have not been effectively resolved in traditional technologies. Insufficient trust in data sharing between departments has led to a decrease in the willingness to share.

[0004] In addition, traditional data sharing methods are slow to respond to rapidly changing market environments and are unable to achieve real-time data sharing and synchronization. Under changing market demands and complex decision-making environments, companies are unable to quickly adjust their strategies, resulting in low collaboration efficiency, which in turn affects the flexibility and competitiveness of the overall business. At the same time, existing technologies lack effective methods for how to adjust strategies under dynamically changing market conditions, making it difficult to meet companies' needs for rapid decision-making and response.

[0005] In summary, the existing cross-departmental data sharing technology has the defects of centralized architecture, data silos, information lag and insufficient security, which not only hinders the efficient collaboration within the enterprise, but also affects the market competitiveness of the enterprise. Therefore, a new method is urgently needed to optimize the overall operational efficiency and decision-making ability of the enterprise by improving data sharing and collaboration strategies. Summary of the invention

[0006] One purpose of the present invention is to propose a method for enterprise cross-departmental data collaborative optimization based on a distributed architecture, which significantly improves the data collaboration capability and decision-making efficiency within the enterprise.

[0007] According to an embodiment of the present invention, a method for enterprise cross-departmental data collaborative optimization based on a distributed architecture includes the following steps:

[0008] S1. Deploy distributed computing nodes in various departments of the enterprise. The distributed computing nodes are connected through the network to form a distributed architecture to share and synchronize enterprise cross-departmental data in real time;

[0009] S2. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, define a set of data conversion strategies for each department, and set a profit function for each game participant. The profit function is calculated based on multiple indicators of the department's data transmission conversion rate to reflect the benefits of each department under different strategies;

[0010] S3, implement a multi-stage optimization process through a multi-agent game model, including local optimization, cross-departmental cooperation and competition, and global optimization stages;

[0011] S4. In the local optimization stage, each department independently optimizes its department strategy based on the real-time updated cross-departmental data of the enterprise;

[0012] S5. In the cross-departmental cooperation and competition stage, departments optimize total costs through cooperation based on shared data and compete to obtain limited departmental strategies;

[0013] S6. In the global optimization stage, the distributed system continuously updates the data of each department and feeds back to the multi-agent game model in the system. The multi-agent game model adjusts the strategies of each department and optimizes the overall coordination capability of the enterprise.

[0014] S7. Apply dynamic programming algorithms to dynamically adjust the overall enterprise collaboration strategy based on real-time data and historical data, optimize the decision-making process, and achieve optimal data conversion and collaboration under the continuous changes of enterprise data and the decision constraints specified by managers;

[0015] S8. In the distributed computing nodes, the strategy execution results of each department are monitored, and the strategy execution results are fed back to the multi-agent game model to adjust the game strategies of each department.

[0016] Optionally, the S1 includes the following steps:

[0017] S11. Deploy multiple distributed computing nodes in each department of the enterprise, and the distributed computing nodes assign a unique identifier ID to each department of the enterprise i , where i represents the department number, from 1 to n, and n is the total number of departments;

[0018] S12, connect distributed computing nodes through the network, build a distributed computing architecture, and carry out real-time data transmission between departments of the enterprise. The real-time data transmission rate is D T :

[0019] D T =(D i +D j ) / T;

[0020] Among them, D i and D j are the amount of data transmitted by department i and department j in time T respectively;

[0021] S13. Set the scope of data sharing and classify cross-departmental data of enterprises into public data. D and private data P D , O D represents a shareable dataset, P D Represents department-specific data sets, and manages data sharing permissions through access control lists;

[0022] S14, real-time tracking of the operation status and data transmission status of the computing nodes through the monitoring module, using the performance indicator P i To conduct an assessment:

[0023] P i =(U i / C i )*100;

[0024] Among them, U i is the actual utilization rate of node i, C i is the maximum carrying capacity of node i.

[0025] Optionally, S2 includes the following steps:

[0026] S21. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, and the strategy set of each participant is S i ;

[0027] S22. Set the profit function R for each game participant i , which is used to reflect the benefits of each department under different strategy choices:

[0028]

[0029] Among them, D T is the real-time data transmission rate, Q i is the product sales volume of department i, C i is the inventory level of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α, β, γ, δ are adjustment coefficients, and N i is the operating cost of department i;

[0030] S23. Evaluate the impact of different strategic choices on the profitability of each department of the enterprise by calculating the profitability function:

[0031]

[0032] Among them, E i represents the change in revenue of department i under different strategies, m is the total number of strategy choices, Θ j is the influence weight of strategy j.

[0033] Optionally, S4 includes the following steps:

[0034] S41. In the local optimization stage, through the multi-agent game model based on real-time updated cross-departmental data, each department independently analyzes its own inventory and procurement needs and defines the department strategy set S i , and set the decision variable X of each department i , X i represents the specific optimization strategy of department i;

[0035] S42. Set a local optimization objective function G for each department i , make the best choice of departmental strategy:

[0036]

[0037] in, represents the profit of department i in the kth strategy selection, is the product sales volume of department i in the kth strategy selection, is the operating cost of department i in the kth strategy selection, β is the weight coefficient, is the operating loss of department i in the kth strategy selection, δ is the loss coefficient;

[0038] S43, using the real-time data update mechanism to dynamically adjust the parameters in the local optimization objective function, and provide real-time feedback of the parameter value D T , P D and O D , so that the objective function G i Adapt to changes in the market and departmental needs;

[0039] S44. Combined with the enterprise's cross-departmental data sharing and synchronization mechanism, conduct deep learning analysis on the historical data of each department, identify potential optimization strategies, and generate optimization plans for each department based on the correlation and trend prediction of shared data;

[0040] S45. By comparing the objective function values ​​before and after optimization, evaluate the strategy improvement effect of each department in the local optimization stage and adjust the subsequent overall coordination strategy:

[0041]

[0042] Among them, E′ irepresents the revenue gain of department i after local optimization, λ is the adjustment coefficient, t is the number of strategy iterations, is the profit of department i in the kth optimization, is the loss in the k-th optimization.

[0043] Optionally, S5 includes the following steps:

[0044] S51. In the inter-departmental cooperation and competition stage, each department uses a multi-agent game model based on a shared data set to analyze the possibility of cooperation and define a set of cooperation strategies C. i ;

[0045] S52. Set the inter-department cooperation benefit function H i j, evaluate the benefits of cooperation between department i and department j:

[0046]

[0047] in, is the real-time data transmission rate between department i and department j, Q i and Q j are the product sales of department i and department j respectively, C i and C j are the operating costs of department i and department j respectively, β is the weight coefficient, L i and L j are the operating losses of department i and department j respectively;

[0048] S53, through the cooperation benefit function H i j is calculated, each department selects a cooperation strategy, optimizes the total cost, and defines the total cost function Z:

[0049]

[0050] Among them, n is the total number of departments participating in cooperation, and m is the total number of departments participating in competition;

[0051] S54. In the competition stage, each department allocates resources and adjusts strategies according to the results of cooperation under the set strategic constraints, so that the department can achieve optimal benefits when obtaining limited resources. The cooperation benefit function is adjusted by evaluating the cooperation and competition effects of each department.

[0052] Optionally, the S6 includes the following steps:

[0053] S61. In the global optimization stage, the distributed system continuously monitors and updates the inventory level, procurement demand, sales situation and market feedback of each department in real time to form an updated data set U i , where U i represents the latest operating data of department i;

[0054] S62, update the data set U i Feedback to the multi-agent game model to recalculate the profit function R of each department i , evaluate the profit performance of each department in the current market environment, and recalculate the profit function R′ i for:

[0055]

[0056] in, is the real-time data transmission rate of department i, A′ i is the product sales volume of department i under the new data conditions, C′ i is the updated operating cost of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α′, β′, γ′, δ′ are new adjustment coefficients, and N′ i is the updated operating cost of department i;

[0057] S63. Based on the recalculated profit function R′ i , each department readjusts its strategy set S in the multi-agent game model i , maximize departmental benefits and optimize overall synergy capabilities, and implement dynamic strategies to optimize and adapt to changes in market demand;

[0058] S64. Introduce a feedback mechanism to calculate performance evaluation indicators by monitoring the effects of strategy adjustments in each department and define effect feedback F i :

[0059] F i =R′ i -R i ;

[0060] Among them, F i It reflects the performance changes of each department after the strategy adjustment;

[0061] S65, combined effect feedback function F i , each department will adjust its strategy again based on the goal of global optimization, forming a closed-loop optimization process and optimizing the overall collaborative capabilities of the enterprise.

[0062] Optionally, the S7 includes the following steps:

[0063] S71. Apply dynamic programming algorithm to construct decision state space based on real-time update of cross-departmental data of the enterprise. The state of decision state space is defined as the current inventory level, purchase demand and sales data of each department, denoted as S t ;

[0064] S72, set the state transfer equation T(S t ,a t ), used to describe the decision action a t Next, the status of each department changes from S t Transfer to S t+1 , where a t represents the optimization strategy selected by each department at time t;

[0065] S73, define the objective function F(S t ,a t ), the objective function is to maximize the overall profit of the enterprise:

[0066]

[0067] Among them, C total The overall operating costs of the enterprise;

[0068] S74, use dynamic programming algorithm to solve multi-stage decision-making, and calculate the strategy a of each department at different time stages t. t The best choice, update the optimal strategy set A*.

[0069] Optionally, the S8 includes the following steps:

[0070] S81. In the distributed computing nodes, monitor the policy execution results of each department in real time and record the actual benefits R″ of each department in policy execution. i and its corresponding decision variable X′ i ;

[0071] S82. Compare the strategy execution results of each department with the expected returns through the feedback mechanism and define the deviation function D i Evaluate the effectiveness of your strategy:

[0072] D i =R″ i -R′ i ;

[0073] S83, the deviation function D i Feedback to the multi-agent game model to analyze the effectiveness of each department's strategy execution and adjust the game strategy set S of each department i :

[0074]

[0075] Among them, γ is the adjustment coefficient, is the deviation function D i The gradient of

[0076] S84. By dynamically adjusting the game strategy set S i, real-time optimization of the strategic choices of each department, so that the strategic choices are in line with market demand and actual execution effects:

[0077]

[0078] Among them, H ij is the cooperation benefit function between department i and department j;

[0079] S85. In the process of adjusting the game strategy, the collaboration and competition between departments should be combined to maximize the overall benefits and promote the synergy between departments:

[0080]

[0081] Among them, Z 1 is the overall benefit, λ is the synergy coefficient, R′ i is the operating loss of department i.

[0082] The beneficial effects of the present invention are:

[0083] (1) The present invention introduces a multi-agent game model, which enables each department to optimize its strategy independently and efficiently. By dynamically adjusting the strategy, the enterprise can flexibly respond to various challenges under the conditions of real-time updated market data, so that the benefits of each department under different strategies can be quantified through a clear benefit function, which greatly improves the scientificity and rationality of decision-making and ensures the optimization of resource allocation.

[0084] (2) The present invention uses a dynamic programming algorithm to adjust the overall collaborative strategy of the enterprise in real time, optimize the decision-making process, and can calculate the optimal data conversion and collaborative strategy in real time under the continuously changing data environment and the decision constraints set by the manager. Compared with the traditional static decision-making model, it significantly improves the response speed and adaptability of the enterprise, ensuring that in a rapidly changing market environment, the enterprise can quickly adjust its strategy and achieve effective use of resources.

[0085] (3) The present invention can optimize the total cost of an enterprise in cross-departmental cooperation and competition. It not only takes into account the independent decision-making of each department, but also integrates the synergy effect from a global perspective. By dynamically monitoring the strategy execution results of each department and providing timely feedback to the game model, the game strategy of each department is adjusted, thereby improving the overall collaborative ability of the enterprise and optimizing the operational efficiency of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0087] Figure 1A flow chart of a method for collaborative optimization of enterprise cross-departmental data based on a distributed architecture proposed by the present invention;

[0088] Figure 2 This is a schematic diagram of constructing a multi-agent game model for an enterprise cross-departmental data collaborative optimization method based on a distributed architecture proposed in the present invention. DETAILED DESCRIPTION

[0089] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0090] refer to Figure 1-2 , a method for enterprise cross-departmental data collaborative optimization based on a distributed architecture, comprising the following steps:

[0091] S1. Deploy distributed computing nodes in various departments of the enterprise. The distributed computing nodes are connected through the network to form a distributed architecture to share and synchronize enterprise cross-departmental data in real time;

[0092] S2. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, define a set of data conversion strategies for each department, and set a profit function for each game participant. The profit function is calculated based on multiple indicators of the department's data transmission conversion rate to reflect the benefits of each department under different strategies;

[0093] S3, implement a multi-stage optimization process through a multi-agent game model, including local optimization, cross-departmental cooperation and competition, and global optimization stages;

[0094] S4. In the local optimization stage, each department independently optimizes its department strategy based on the real-time updated cross-departmental data of the enterprise;

[0095] S5. In the cross-departmental cooperation and competition stage, departments optimize total costs through cooperation based on shared data and compete to obtain limited departmental strategies;

[0096] S6. In the global optimization stage, the distributed system continuously updates the data of each department and feeds back to the multi-agent game model in the system. The multi-agent game model adjusts the strategies of each department and optimizes the overall coordination capability of the enterprise.

[0097] S7. Apply dynamic programming algorithms to dynamically adjust the overall enterprise collaboration strategy based on real-time data and historical data, optimize the decision-making process, and achieve optimal data conversion and collaboration under the continuous changes of enterprise data and the decision constraints specified by managers;

[0098] S8. In the distributed computing nodes, the strategy execution results of each department are monitored and fed back to the multi-agent game model to adjust the game strategies of each department.

[0099] In this implementation, S1 includes the following steps:

[0100] S11. Deploy multiple distributed computing nodes in each department of the enterprise, and the distributed computing nodes assign a unique identifier ID to each department of the enterprise i , where i represents the department number, from 1 to n, and n is the total number of departments;

[0101] S12, connect distributed computing nodes through the network, build a distributed computing architecture, and carry out real-time data transmission between departments of the enterprise. The real-time data transmission rate is D T :

[0102] D T =(D i +D j ) / T;

[0103] Among them, D i and D j are the amount of data transmitted by department i and department j in time T respectively;

[0104] S13. Set the scope of data sharing and classify cross-departmental data of enterprises into public data. D and private data P D , O D represents a shareable dataset, P D Represents department-specific data sets, and manages data sharing permissions through access control lists;

[0105] S14, real-time tracking of the operation status and data transmission status of the computing nodes through the monitoring module, using the performance indicator P i To conduct an assessment:

[0106] P i =(U i / C i )*100;

[0107] Among them, U i is the actual utilization rate of node i, C i is the maximum carrying capacity of node i.

[0108] In this implementation, S2 includes the following steps:

[0109] S21. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, and the strategy set of each participant is S i ;

[0110] S22. Set the profit function R for each game participant i , which is used to reflect the benefits of each department under different strategy choices:

[0111]

[0112] Among them, D T is the real-time data transmission rate, Q i is the product sales volume of department i, C i is the inventory level of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α, β, γ, δ are adjustment coefficients, and N i is the operating cost of department i;

[0113] S23. Evaluate the impact of different strategic choices on the profitability of each department of the enterprise by calculating the profitability function:

[0114]

[0115] Among them, E i represents the change in revenue of department i under different strategies, m is the total number of strategy choices, Θ j is the influence weight of strategy j.

[0116] In this implementation, S4 includes the following steps:

[0117] S41. In the local optimization stage, through the multi-agent game model based on real-time updated cross-departmental data, each department independently analyzes its own inventory and procurement needs and defines the department strategy set S i , and set the decision variable X of each department i , X i represents the specific optimization strategy of department i;

[0118] S42. Set a local optimization objective function G for each department i , make the best choice of departmental strategy:

[0119]

[0120] in, represents the profit of department i in the kth strategy selection, is the product sales volume of department i in the kth strategy selection, is the operating cost of department i in the kth strategy selection, β is the weight coefficient, is the operating loss of department i in the kth strategy selection, δ is the loss coefficient;

[0121] S43, using the real-time data update mechanism to dynamically adjust the parameters in the local optimization objective function, and provide real-time feedback of the parameter value D T , P D and O D , so that the objective function G i Adapt to changes in the market and departmental needs;

[0122] S44. Combined with the enterprise's cross-departmental data sharing and synchronization mechanism, conduct deep learning analysis on the historical data of each department, identify potential optimization strategies, and generate optimization plans for each department based on the correlation and trend prediction of shared data;

[0123] S45. By comparing the objective function values ​​before and after optimization, evaluate the strategy improvement effect of each department in the local optimization stage and adjust the subsequent overall coordination strategy:

[0124]

[0125] Among them, E′ i represents the revenue gain of department i after local optimization, λ is the adjustment coefficient, t is the number of strategy iterations, is the profit of department i in the kth optimization, is the loss in the k-th optimization.

[0126] In this implementation, S5 includes the following steps:

[0127] S51. In the inter-departmental cooperation and competition stage, each department uses a multi-agent game model based on a shared data set to analyze the possibility of cooperation and define a set of cooperation strategies C. i ;

[0128] S52. Set the inter-department cooperation benefit function H i j, evaluate the benefits of cooperation between department i and department j:

[0129]

[0130] in, is the real-time data transmission rate between department i and department j, Q i and Q j are the product sales of department i and department j respectively, C i and C j are the operating costs of department i and department j respectively, β is the weight coefficient, L i and L j are the operating losses of department i and department j respectively;

[0131] S53, through the cooperation benefit function H ij is calculated, each department selects a cooperation strategy, optimizes the total cost, and defines the total cost function Z:

[0132]

[0133] Among them, n is the total number of departments participating in cooperation, and m is the total number of departments participating in competition;

[0134] S54. In the competition stage, each department allocates resources and adjusts strategies according to the results of cooperation under the set strategic constraints, so that the department can achieve optimal benefits when obtaining limited resources. The cooperation benefit function is adjusted by evaluating the cooperation and competition effects of each department.

[0135] In this implementation, S6 includes the following steps:

[0136] S61. In the global optimization stage, the distributed system continuously monitors and updates the inventory level, procurement demand, sales situation and market feedback of each department in real time to form an updated data set U i , where U i represents the latest operating data of department i;

[0137] S62, update the data set U i Feedback to the multi-agent game model to recalculate the profit function R of each department i , evaluate the profit performance of each department in the current market environment, and recalculate the profit function R′ i for:

[0138]

[0139] in, is the real-time data transmission rate of department i, A′ i is the product sales volume of department i under the new data conditions, C′ i is the updated operating cost of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α′, β′, γ′, δ′ are new adjustment coefficients, and N′ i is the updated operating cost of department i;

[0140] S63. Based on the recalculated profit function R′ i , each department readjusts its strategy set S in the multi-agent game model i , maximize departmental benefits and optimize overall synergy capabilities, and implement dynamic strategies to optimize and adapt to changes in market demand;

[0141] S64. Introduce a feedback mechanism to calculate performance evaluation indicators by monitoring the effects of strategy adjustments in each department and define effect feedback F i :

[0142]

[0143] Among them, F i It reflects the performance changes of each department after the strategy adjustment;

[0144] S65, combined effect feedback function F i , each department will adjust its strategy again based on the goal of global optimization, forming a closed-loop optimization process and optimizing the overall collaborative capabilities of the enterprise.

[0145] In this implementation, S7 includes the following steps:

[0146] S71. Apply dynamic programming algorithm to construct decision state space based on real-time update of cross-departmental data of the enterprise. The state of decision state space is defined as the current inventory level, purchase demand and sales data of each department, denoted as S t ;

[0147] S72, set the state transfer equation T(S t ,a t ), used to describe the decision action a t Next, the status of each department changes from S t Transfer to S t+1 , where a t represents the optimization strategy selected by each department at time t;

[0148] S73, define the objective function F(S t ,a t ), the objective function is to maximize the overall profit of the enterprise:

[0149]

[0150] Among them, C total The overall operating costs of the enterprise;

[0151] S74, use dynamic programming algorithm to solve multi-stage decision-making, and calculate the strategy a of each department at different time stages t. t The best choice, update the optimal strategy set A*.

[0152] In this implementation, S8 includes the following steps:

[0153] S81. In the distributed computing nodes, monitor the policy execution results of each department in real time and record the actual benefits R″ of each department in policy execution. i and its corresponding decision variable X′ i ;

[0154] S82. Compare the strategy execution results of each department with the expected returns through the feedback mechanism and define the deviation function D i Evaluate the effectiveness of your strategy:

[0155] D i =R″ i -R′ i ;

[0156] S83, the deviation function D i Feedback to the multi-agent game model to analyze the effectiveness of each department's strategy execution and adjust the game strategy set S of each department i :

[0157]

[0158] Among them, γ is the adjustment coefficient, is the deviation function D i The gradient of

[0159] S84. By dynamically adjusting the game strategy set S i , real-time optimization of the strategic choices of each department, so that the strategic choices are in line with market demand and actual execution effects:

[0160]

[0161] Among them, H ij is the cooperation benefit function between department i and department j;

[0162] S85. In the process of adjusting the game strategy, the collaboration and competition between departments should be combined to maximize the overall benefits and promote the synergy between departments:

[0163]

[0164] Among them, Z 1 is the overall benefit, λ is the synergy coefficient, L′ i is the operating loss of department i.

[0165] Embodiment 1:

[0166] In a large e-commerce enterprise, daily operations involve multiple departments, including sales, procurement, logistics and finance. Departments need to efficiently share and synchronize data in order to make accurate decisions in a rapidly changing market environment. However, the existing technology has obvious deficiencies in data integration and collaborative decision-making between departments, resulting in the formation of information islands and reducing overall operational efficiency. By implementing the enterprise cross-departmental data sharing and synchronization method based on a distributed architecture of the present invention, the enterprise can effectively solve this problem.

[0167] In this embodiment, an e-commerce enterprise faces problems of insufficient inventory, untimely procurement and transportation delays in its promotion activities before holidays. In order to cope with these challenges, the enterprise decides to implement the method based on the present invention to ensure that various departments can effectively collaborate based on real-time data.

[0168] The sales department reports the estimated sales volume of promotional products to the system based on market demand forecasts. The sales department estimates that the sales volume of a certain smartphone will be 5,000 units. By combining it with market data, the system generates a real-time data set that includes historical sales, competitor pricing information, and market feedback. Based on this data, the sales department sets its strategy for collective sales and decides to set the promotion price at 2,000 yuan to attract more consumers.

[0169] After receiving the estimated sales data from the sales department, the purchasing department uses the multi-agent game model of the present invention to adjust its strategy. The purchasing department analyzes the current inventory level and supply chain status and finds that the current inventory is 3,000 units and it is expected that 2,000 units will be needed to meet sales demand. Therefore, the purchasing department sets its strategy set procurement to increase the purchase order by 5,000 units, and considers adding bulk discounts to the order to optimize the procurement cost.

[0170] At the same time, after receiving the information, the logistics department evaluated the transportation capacity and existing transportation routes. Since the promotional activities will lead to a surge in orders, the logistics department decided to adjust its strategy and planned to add temporary transportation shifts during the promotion period to ensure timely delivery. In this case, the logistics department set its strategy to add 5 temporary transportation routes.

[0171] In order to integrate the above information, the system uses a dynamic programming algorithm to calculate the overall coordination strategy in real time. After real-time data updates and adjustments to the multi-agent game model, the departments reached a consensus: the sales department's estimated sales, the procurement department's replenishment needs, and the logistics department's distribution strategy are coordinated to maximize the benefits of each department. In the end, the strategy sets of each department were integrated into a unified execution plan.

[0172] During the specific implementation process, the system monitored the strategy execution results of each department. The sales department's sales in the first two days after the event started were as follows:

[0173] Sales on the first day: 2,000 units (1,000 units remaining in stock);

[0174] Sales on the second day: 2,500 units (inadequate inventory, need to start replenishment);

[0175] Through real-time monitoring, the system found that sales were too hot and expected to need to be replenished quickly before the end of the promotion. The purchasing department immediately placed an order and successfully completed the purchase within 48 hours to ensure that sales demand was met. At the same time, the logistics department adjusted the delivery frequency during the promotion period, adding two shipments per day to ensure that the goods were delivered to customers in a timely manner.

[0176] After the event, the company collected operational data from each department and compared it with traditional methods. With traditional methods, each department made independent decisions without real-time data support, leading to the following problems:

[0177] Due to the lack of inventory information, the sales department was out of stock the next day and missed sales opportunities.

[0178] The purchasing department failed to respond to market changes in a timely manner, which delayed purchases and led to an increase in customer complaints.

[0179] The logistics department lacked information sharing with the sales and purchasing departments, and unreasonable transportation arrangements caused unnecessary delays.

[0180] Through the method of the present invention, the enterprise has achieved the following results in sales promotion activities:

[0181] Total sales: RMB 15 million, a 20% increase over the same period last year.

[0182] Customer satisfaction: reached 95%, 15% higher than traditional methods.

[0183] Logistics delivery time efficiency: The average delivery time was shortened by 30%, from 3 days to 2 days.

[0184] Comparative data show that under the traditional method, the total sales of the promotion was 12.5 million yuan, the customer satisfaction rate was only 80%, and the average logistics delivery time was 3 days. After implementing the method of the present invention, the company achieved significant improvements in sales, customer satisfaction and logistics efficiency.

[0185] This embodiment fully demonstrates the feasibility and effectiveness of the present invention, successfully solves the problems faced by enterprises in cross-departmental data sharing and synchronization, realizes efficient collaboration between departments through real-time data and dynamic adjustment strategies, and improves the overall operational efficiency and market responsiveness of the enterprise.

[0186] The present invention introduces a multi-agent game model, allowing each department to optimize its strategy independently and efficiently. By dynamically adjusting the strategy, the enterprise can flexibly respond to various challenges under the condition of real-time updated market data, so that the benefits of each department under different strategies can be quantified through a clear benefit function, which greatly improves the scientificity and rationality of decision-making and ensures the optimization of resource allocation.

[0187] The present invention utilizes a dynamic programming algorithm to adjust the overall collaborative strategy of the enterprise in real time, optimize the decision-making process, and can calculate the optimal data conversion and collaborative strategy in real time under the continuously changing data environment and the decision constraints set by managers. Compared with the traditional static decision-making model, it significantly improves the response speed and adaptability of the enterprise, ensuring that in a rapidly changing market environment, the enterprise can quickly adjust its strategy and achieve effective use of resources.

[0188] The present invention can optimize the total cost of an enterprise in cross-departmental cooperation and competition. It not only takes into account the independent decision-making of each department, but also integrates the synergy effect from a global perspective. By dynamically monitoring the strategy execution results of each department and providing timely feedback to the game model, the game strategy of each department is adjusted, thereby improving the overall collaborative ability of the enterprise and optimizing the operational efficiency of the enterprise.

[0189] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for collaborative optimization of enterprise cross-departmental data based on a distributed architecture, characterized in that: The steps include: S1. Deploy distributed computing nodes in various departments of the enterprise. The distributed computing nodes are connected through the network to form a distributed architecture to share and synchronize enterprise cross-departmental data in real time; S2. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, define a set of data conversion strategies for each department, and set a profit function for each game participant. The profit function is calculated based on multiple indicators of the department's data transmission conversion rate to reflect the benefits of each department under different strategies; S3, implement a multi-stage optimization process through a multi-agent game model, including local optimization, cross-departmental cooperation and competition, and global optimization stages; S4. In the local optimization stage, each department independently optimizes its department strategy based on the real-time updated cross-departmental data of the enterprise; S5. In the cross-departmental cooperation and competition stage, departments optimize total costs through cooperation based on shared data and compete to obtain limited departmental strategies; S6. In the global optimization stage, the distributed system continuously updates the data of each department and feeds back to the multi-agent game model in the system. The multi-agent game model adjusts the strategies of each department and optimizes the overall coordination capability of the enterprise. S7. Apply dynamic programming algorithms to dynamically adjust the overall enterprise collaboration strategy based on real-time data and historical data, optimize the decision-making process, and achieve optimal data conversion and collaboration under the continuous changes of enterprise data and the decision constraints specified by managers; S8. In the distributed computing nodes, the strategy execution results of each department are monitored, and the strategy execution results are fed back to the multi-agent game model to adjust the game strategies of each department.

2. According to the distributed architecture-based enterprise cross-departmental data collaborative optimization method of claim 1, it is characterized by: The S1 comprises the following steps: S11. Deploy multiple distributed computing nodes in each department of the enterprise, and the distributed computing nodes assign a unique identifier ID to each department of the enterprise i , where i represents the department number, from 1 to n, and n is the total number of departments; S12, connect distributed computing nodes through the network, build a distributed computing architecture, and carry out real-time data transmission between departments of the enterprise. The real-time data transmission rate is D T : D T =(D i +D j ) / T; Among them, D i and D j are the amount of data transmitted by department i and department j in time T respectively; S13. Set the scope of data sharing and classify cross-departmental data of enterprises into public data. D and private data P D , O D represents a shareable dataset, P D Represents department-specific data sets, and manages data sharing permissions through access control lists; S14, real-time tracking of the operation status and data transmission status of the computing nodes through the monitoring module, using the performance indicator P i To conduct an assessment: P i =(U i / C i )*100; Among them, U i is the actual utilization rate of node i, C i is the maximum carrying capacity of node i.

3. According to the distributed architecture-based enterprise cross-departmental data collaborative optimization method of claim 1, it is characterized by: The S2 comprises the following steps: S21. Establish a multi-agent game model, treat each department of the enterprise as an independent game participant, and the strategy set of each participant is S i ; S22. Set the profit function R for each game participant i , which is used to reflect the benefits of each department under different strategy choices: Among them, D T is the real-time data transmission rate, Q i is the product sales volume of department i, C i is the inventory level of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α, β, γ, δ are adjustment coefficients, and N i is the operating cost of department i; S23. Evaluate the impact of different strategic choices on the profitability of each department of the enterprise by calculating the profitability function: Among them, E i represents the change in revenue of department i under different strategies, m is the total number of strategy choices, Θ j is the influence weight of strategy j.

4. According to the distributed architecture-based enterprise cross-departmental data collaborative optimization method of claim 1, it is characterized by: The S4 comprises the following steps: S41. In the local optimization stage, through the multi-agent game model based on real-time updated cross-departmental data, each department independently analyzes its own inventory and procurement needs and defines the department strategy set S i , and set the decision variable X of each department i , X i represents the specific optimization strategy of department i; S42. Set a local optimization objective function G for each department i , make the best choice of departmental strategy: in, represents the profit of department i in the kth strategy selection, is the product sales volume of department i in the kth strategy selection, is the operating cost of department i in the kth strategy selection, β is the weight coefficient, is the operating loss of department i in the kth strategy selection, δ is the loss coefficient; S43, using the real-time data update mechanism to dynamically adjust the parameters in the local optimization objective function, and provide real-time feedback of the parameter value D T , P D and O D , so that the objective function G i Adapt to changes in the market and departmental needs; S44. Combined with the enterprise's cross-departmental data sharing and synchronization mechanism, conduct deep learning analysis on the historical data of each department, identify potential optimization strategies, and generate optimization plans for each department based on the correlation and trend prediction of shared data; S45. By comparing the objective function values ​​before and after optimization, evaluate the strategy improvement effect of each department in the local optimization stage and adjust the subsequent overall coordination strategy: Among them, E′ i represents the revenue gain of department i after local optimization, λ is the adjustment coefficient, t is the number of strategy iterations, is the profit of department i in the kth optimization, is the loss in the k-th optimization.

5. According to the distributed architecture-based enterprise cross-departmental data collaborative optimization method of claim 1, it is characterized by: The S5 comprises the following steps: S51. In the inter-departmental cooperation and competition stage, each department uses a multi-agent game model based on a shared data set to analyze the possibility of cooperation and define a set of cooperation strategies C. i ; S52. Set the inter-department cooperation benefit function H i j, evaluate the benefits of cooperation between department i and department j: in, is the real-time data transmission rate between department i and department j, Q i and Q j are the product sales of department i and department j respectively, C i and C j are the operating costs of department i and department j respectively, β is the weight coefficient, L i and L j are the operating losses of department i and department j respectively; S53, through the cooperation benefit function H i j is calculated, each department selects a cooperation strategy, optimizes the total cost, and defines the total cost function Z: Among them, n is the total number of departments participating in cooperation, and m is the total number of departments participating in competition; S54. In the competition stage, each department allocates resources and adjusts strategies according to the results of cooperation under the set strategic constraints, so that the department can achieve optimal benefits when obtaining limited resources. The cooperation benefit function is adjusted by evaluating the cooperation and competition effects of each department.

6. The enterprise cross-department data collaborative optimization method based on distributed architecture according to claim 1 is characterized in that: The S6 comprises the following steps: S61. In the global optimization stage, the distributed system continuously monitors and updates the inventory level, procurement demand, sales situation and market feedback of each department in real time to form an updated data set U i , where U i represents the latest operating data of department i; S62, update the data set U i Feedback to the multi-agent game model to recalculate the profit function R of each department i , evaluate the profit performance of each department in the current market environment, and recalculate the profit function R′ i for: in, is the real-time data transmission rate of department i, A′ i is the product sales volume of department i under the new data conditions, C′ i is the updated operating cost of department i, P D is the procurement cost of the department, O D is the amount of public data shared by the department, α′, β′, γ′, δ′ are new adjustment coefficients, and N′ i is the updated operating cost of department i; S63. Based on the recalculated profit function R′ i , each department readjusts its strategy set S in the multi-agent game model i , maximize departmental benefits and optimize overall synergy capabilities, and implement dynamic strategies to optimize and adapt to changes in market demand; S64. Introduce a feedback mechanism to calculate performance evaluation indicators by monitoring the effects of strategy adjustments in each department and define effect feedback F i : F i =R′ i -R i ; Among them, F i It reflects the performance changes of each department after the strategy adjustment; S65, combined effect feedback function F i , each department will adjust its strategy again based on the goal of global optimization, forming a closed-loop optimization process and optimizing the overall collaborative capabilities of the enterprise.

7. The enterprise cross-department data collaborative optimization method based on distributed architecture according to claim 1 is characterized in that: The S7 comprises the following steps: S71. Apply dynamic programming algorithm to construct decision state space based on real-time update of cross-departmental data of the enterprise. The state of decision state space is defined as the current inventory level, purchase demand and sales data of each department, denoted as S t ; S72, set the state transfer equation T(S t ,a t ), used to describe the decision action a t Next, the status of each department changes from S t Transfer to S t+1 , where a t represents the optimization strategy selected by each department at time t; S73, define the objective function F(S t ,a t ), the objective function is to maximize the overall profit of the enterprise: Among them, C total The overall operating costs of the enterprise; S74, use dynamic programming algorithm to solve multi-stage decision-making, and calculate the strategy a of each department at different time stages t. t The best choice, update the optimal strategy set A*.

8. The enterprise cross-department data collaborative optimization method based on distributed architecture according to claim 1 is characterized in that: The S8 comprises the following steps: S81. In the distributed computing nodes, monitor the policy execution results of each department in real time and record the results of each department. The actual profit of the gate in strategy execution R i and its corresponding decision variable X i ; S82. Compare the strategy execution results of each department with the expected returns through the feedback mechanism and define the deviation function D i Evaluate the effectiveness of your strategy: ″′ D i =R i -R i ; S83, the deviation function D i Feedback to the multi-agent game model to analyze the effectiveness of each department's strategy execution and adjust the game strategy set S of each department i : Among them, γ is the adjustment coefficient, is the deviation function D i The gradient of S84. By dynamically adjusting the game strategy set S i , real-time optimization of the strategic choices of each department, so that the strategic choices are in line with market demand and actual execution effects: Among them, H ij is the cooperation benefit function between department i and department j; S85. In the process of adjusting the game strategy, the collaboration and competition between departments should be combined to maximize the overall benefits and promote the synergy between departments: Among them, Z1 is the overall benefit, λ is the synergy coefficient, L′ i is the operating loss of department i.

Images