Profit distribution model construction and collaborative optimization method of industrial software service platform under multiple operation modes

Abstract

The invention discloses a benefit distribution model construction and collaborative optimization method under multiple operation modes of an industrial software service platform, and the method comprises the following steps: 1, analyzing an optimal decision based on a Stackelberg master-slave game decision model: 11) constructing a pricing model, profit functions and constraint conditions of the industrial software service platform and the industrial software developer are obtained respectively; 12) performing optimal game decision analysis; 121) analyzing optimal decisions of an industrial software developer and an industrial software service platform under an independent operation condition to obtain an optimal price and a maximum benefit; 122) analyzing the optimal decision of the industrial software developer and the industrial software service platform under the cooperative operation condition to obtain the overall maximum benefit; 2, establishing a benefit distribution model based on a Shapley value method; step 3, benefit decision optimization: 31) fitting a relation function between the price and the number of functional modules by using historical data; and 32) adopting a reinforcement learning method to decide an optimal benefit distribution scheme.

Description

Technical Field

[0001] The present invention relates to the field of industrial software development and operation technology, and specifically to a method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform. Background Art

[0002] The industrial software service platform plays a key role in the industrial Internet ecosystem. It is responsible for connecting industrial software developers and users. It is an important link for information transmission. Relying on its powerful information technology, it provides support for information sharing among industrial software developers, and has an important impact on the developers' operations and development.

[0003] However, there are many problems with the current profit distribution mechanism of industrial software service platforms. Under information symmetry, the multi-subject distribution mechanism of industrial software service platforms is not sound, which limits the sustainability of the platform. When regional small and medium-sized enterprise clusters carry out multi-mode collaborative applications in the industrial Internet environment, the profit model and profit incentive mechanism are not perfect.

[0004] 1. Limitations of the existing benefit distribution method

[0005] 1.1. Assumptions and limitations of the Stackelberg game model

[0006] (1) Demand stability assumption: When solving the problem of benefit distribution, the Stackelberg game model assumes that demand is stable. However, in the actual market, demand changes dynamically. This assumption ignores the impact of market demand fluctuations on the interests of various entities, resulting in the fact that the benefit distribution plan may not accurately reflect the actual contribution and deserved benefits of each entity in actual application.

[0007] (2) Simplifying realistic factors: To simplify calculations, the model assumes that realistic factors such as input and loss generated when the subjects cooperate do not exist. However, in actual cooperation, these factors have an important impact on the costs and benefits of each subject. Ignoring them will cause the distribution of benefits to deviate from the actual situation and fail to reasonably reflect the input and return of each subject in the cooperation.

[0008] 1.2. The basic assumptions of the Shapley value method are flawed

[0009] Assumption of equivalent risk, resource input and information transmission: When distributing benefits, the Shapley value method assumes that each subject has the same degree of risk-taking, resource input and information transmission. However, in actual industrial software service platforms, different subjects have significant differences in these aspects. For example, industrial software developers face technology research and development risks and a large amount of technology investment in the software development process, while industrial software service platforms need to bear risks such as server maintenance and data security, as well as a large amount of asset investment to build server clusters and data storage systems. In terms of information transmission, the value and amount of information mastered and transmitted by different subjects are also different. This assumption cannot accurately distinguish the contributions of each subject, making the distribution of benefits unfair and unreasonable.

[0010] 2. Imperfect profit model and profit incentive mechanism

[0011] 2.1. Single profit model: In the industrial software service platform environment, the existing profit model is often limited to traditional product sales and simple service charges, and lacks effective development of emerging profit channels such as data sharing value-added and platform personalized services. This has led to limited profitability of enterprises, and they are unable to fully utilize the advantages of the industrial Internet to achieve value-added, limiting the development space of platforms and enterprises.

[0012] 2.2. Insufficient profit incentive mechanism

[0013] (1) Weak incentives: The incentives for enterprises to participate in complex collaborative tasks and innovative activities are insufficient, and the incentive measures lack attractiveness, making it difficult to stimulate the enthusiasm and innovation drive of enterprises.

[0014] (2) Lack of specificity: It does not take into account the differences in scale, resources, development stage, etc. between SMEs and cannot meet the actual needs of different enterprises, resulting in some enterprises being unresponsive to the incentive mechanism.

[0015] (3) No long-term planning: Lacking long-term incentive planning, companies focus more on short-term benefits and ignore long-term technological innovation and industrial development investment, which is not conducive to the sustainable development of industrial clusters. Summary of the invention

[0016] In view of this, the purpose of the present invention is to provide a profit distribution model construction and collaborative optimization method under multiple operation modes of an industrial software service platform, so as to build a more reasonable profit distribution model and optimize the profit model and profit incentive mechanism.

[0017] In order to achieve the above object, the present invention provides the following technical solutions:

[0018] A method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform includes the following steps:

[0019] Step 1: Analyze the optimal decision based on Stackelberg's master-slave game decision model

[0020] 11) Construct a pricing model to obtain the profit function and constraints of the industrial software service platform and industrial software developers respectively;

[0021] 12) Optimal game decision analysis based on pricing model

[0022] 121) Analyze the optimal decision of industrial software developers and industrial software service platforms under independent operation conditions, and obtain the optimal price and maximum benefit of industrial software developers and industrial software service platforms;

[0023] 122) Analyze the optimal decision of industrial software developers and industrial software service platforms under cooperative operation conditions to obtain the overall maximum benefits of industrial software developers and industrial software service platforms;

[0024] Step 2: Establish a benefit distribution model based on the Shapley value method;

[0025] Step 3: Benefit Decision Optimization

[0026] 31) Use historical data to fit the relationship function between price and the number of functional modules;

[0027] 32) Use reinforcement learning method to decide the optimal benefit distribution plan.

[0028] Furthermore, in step 11), the profit function of the industrial software service platform is:

[0029]

[0030] Where: platform is the profit of the industrial software service platform; C0 is the total cost of the project; r platform is the proportion of the operating cost of the industrial software service platform to the total project revenue; E is the market demand elasticity coefficient; V avg is the average price of similar projects in the market; k is the comprehensive cost coefficient; P1 is the deposit paid by users in the early stage of the project; P2 is the deposit paid by users in the middle stage of the project; P3 is the balance paid by users after the project is completed;

[0031] The profit function of industrial software developers is:

[0032]

[0033] Where: developer is the profit function of industrial software developers; r d The actual profit retention ratio of industrial software developers after deducting various costs; C dis the software development cost of industrial software developers; E d is the market demand elasticity coefficient; V avgd The average price of similar projects in the market;

[0034] For industrial software developers, their profits must meet the minimum expected profit requirements:

[0035]

[0036] in: It is the minimum profit bottom line that developers are willing to take on a project;

[0037] For users, the perceived value of customized software must be greater than the user's psychological expected price:

[0038]

[0039] Where: V is the price of customized software; The price expected by the user.

[0040] Furthermore, in step 121), the optimal decision for the industrial software platform to be independently operated is:

[0041]

[0042] in: To maximize the profits of the industrial software service platform under independent operation; The best price for an industrial software platform operating independently; The best price for early stage users to pay deposits for projects under independent operation of industrial software platforms; The best price for mid-term project users to pay deposits under independent operation of industrial software platforms; The best price for users to pay the balance after the project is completed under the independent operation of the industrial software platform.

[0043] Furthermore, in step 121), the optimal decision for the industrial software developer to operate independently is:

[0044]

[0045] in: To maximize the profits of industrial software developers operating independently; The best price for industrial software developers operating independently; The best price for early stage users to pay deposits for projects operated independently by industrial software developers; The best price for mid-term users to pay deposits for projects under independent operation of industrial software developers; The best price for users to pay the final payment after the project is completed when the industrial software developer is independently operated.

[0046] Further, in step 122), the profit function of the industrial software developer and the industrial software service platform under the cooperative operation conditions is:

[0047]

[0048] in: To maximize profits under cooperative management conditions; The best price under cooperative operation conditions; The best price for the early users of the project to pay the deposit under cooperative operation conditions; The best price for the mid-term users of the project to pay the deposit under cooperative operation conditions; The optimal price for users to pay the balance after the project is completed under cooperative operation conditions.

[0049] Furthermore, in step 2, the method for establishing the benefit distribution model based on the Shapley value method is:

[0050] The Shapley value benefit distribution model indicates that:

[0051]

[0052] in: is the distribution obtained by the i-th member in the cooperation mode; |S| is the weighting factor; S is the set of all alliances that contain member i; |S| is the number of members in alliance S; v(S) is the maximum benefit of alliance S; v(S) is the maximum benefit of alliance S after removing member i; n is the number of members;

[0053] Substituting the overall maximum benefits of industrial software developers and industrial software service platforms into the Shapley value benefit distribution model, we can obtain the Shapley value of benefit distribution under the cooperation between industrial software developers and industrial software service platforms and the Shapley value of benefit distribution under the cooperation between industrial software service platforms and users:

[0054]

[0055] in: The Shapley value for the benefits of cooperation between industrial software developers and industrial software service platforms; The Shapley value for the benefit distribution between the industrial software service platform and users: To maximize the profits of industrial software developers operating independently; To maximize the profits of the industrial software service platform under independent operation; It is the maximum common profit under the cooperative operation conditions between industrial software developers and industrial software service platforms.

[0056] The beneficial effects of the present invention are:

[0057] The profit distribution model construction and collaborative optimization method of the industrial software service platform under multiple operation modes of the present invention comprehensively considers the impact of multiple factors such as resource input, risk bearing and information transmission on profit distribution, overcomes the defects of existing profit distribution methods that only consider some factors or oversimplify assumptions, makes profit distribution more fair and reasonable, and conforms to the complex actual operation of the industrial Internet platform; by using the historical data fitting function of the interaction between the developer and the platform and strengthening learning to make decisions, it can more accurately determine the price and interest relationship, optimize the platform's operational decisions, and improve the platform's economic benefits and sustainable development capabilities. And has the following advantages:

[0058] (1) Build a more reasonable benefit distribution model

[0059] The present invention aims to overcome the defects of the existing industrial software platform profit distribution method and build a more reasonable, fair and accurate profit distribution model that reflects the contribution of each subject. By comprehensively considering various factors, such as resource input, risk taking and information transmission, the profit distribution results are more in line with the actual situation and promote the cooperation and development among the various subjects of the industrial Internet platform.

[0060] (2) Optimize profit model and profit incentive mechanism

[0061] In view of the problems of single existing profit model and insufficient profit incentive mechanism, the purpose of this invention is to develop a diversified profit model and fully tap the potential profit points of the industrial Internet platform, such as data sharing value-added, platform personalized services, etc. At the same time, a more attractive and targeted profit incentive mechanism is designed to encourage enterprises to actively participate in collaborative tasks and innovation activities and promote the sustainable development of industrial clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0063] Figure 1 A flowchart of the profit distribution model construction and collaborative optimization method of the industrial software service platform under multiple operation modes of the present invention;

[0064] Figure 2 The schematic diagram of the relationship function between price and the number of functional modules using historical data;

[0065] Figure 3 Schematic diagram of the optimal benefit distribution plan using reinforcement learning method. DETAILED DESCRIPTION

[0066] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0067] like Figure 1 As shown, the profit distribution model construction and collaborative optimization method under multiple operation modes of the industrial software service platform in this embodiment includes the following steps.

[0068] Step 1: Analyze the optimal decision based on Stackelberg's master-slave game decision model

[0069] 1.1. Stackelberg-based master-slave game decision model

[0070] The Stackelberg game model is a leader-follower dynamic game model. In the multi-operation model of the industrial software service platform, the industrial software platform, as the leader, first determines the pricing strategy of the project (including the deposit and final payment amount at each stage), and the software developer, as the follower, determines its own development investment and effort based on the pricing decision of the platform. By optimizing the pricing strategy, the platform aims to maximize its own profits while taking into account the interests of developers and the acceptance of users to achieve coordination and optimization of the entire industrial chain.

[0071] 1.2 Analysis of the cooperation mechanism among industrial software platforms, users and developers

[0072] 1.2.1 Analysis of Cooperation between Industrial Software Platform and Users

[0073] In the business model of industrial software platforms, the platform cooperates with users to provide users with multiple service options such as package services and one-time services. The platform is responsible for the development, maintenance and update of the software, and provides technical support and after-sales service. For package services, it is suitable for corporate users with long-term demand for software. The subscription system is adopted. Users pay fees regularly to obtain continuous software updates and technical support, reduce software upgrade and maintenance costs, and the platform can also obtain stable income and improve user stickiness. For one-time services, it is suitable for users with specific project needs. According to the complexity of the project and the workload, the pricing is based on a fixed price or project quotation method, allowing users to choose services on demand to avoid unnecessary expenses, and the platform can flexibly undertake projects to expand market share. The platform and users understand the needs by establishing effective communication channels, customize different services, sign cooperation agreements to clarify the rights and obligations of both parties, and continuously improve service quality and software performance based on user feedback. In terms of revenue distribution, the revenue of package services comes from user subscription fees, part of which is used for continuous development and maintenance of software, and preferential rewards can be given according to user conditions to increase stickiness; the revenue of one-time services is obtained through project quotations, which are used for project implementation and team incentives, and can also cooperate with developers to share revenue to increase their enthusiasm. At the same time, the platform controls risks by paying attention to market trends, conducting continuous technological research and development, and complying with laws and regulations to ensure smooth cooperation with users and achieve win-win results for both parties.

[0074] 1.2.2 Analysis of cooperation between industrial software platforms and software developers

[0075] In the cooperation model with the industrial software platform, the cooperation fee paid by the developer to the platform is of great significance. In terms of pricing strategy, there is a basic cooperation fee to establish a basic cooperation relationship, including project display and demand matching; the project cooperation fee is determined according to the scale, complexity and expected benefits, so that the platform can invest more resources; the value-added service fee is charged according to the specific service content and value. In terms of service content, the fee is used to enhance technical support, provide developers with efficient development tools, technical training and troubleshooting services; for market promotion to increase the platform's visibility and influence, bring more opportunities and resources to developers; and also for establishing a resource sharing platform to reduce the cost of developers. The cooperation agreement specifies the payment method, purpose and liability for breach of contract. It can be paid in one lump sum, in installments or according to progress, indicating that the fee is used for technology research and development, market promotion, etc., and there will be corresponding sanctions for breach of contract. In terms of income distribution, a risk fund is established to deal with project risks. When successful, the developer will receive income distribution according to the proportion of the fee, and failure can make up for part of the loss; the reward mechanism gives excellent developers rewards such as reduced fees and priority project opportunities based on the payment of cooperation fees and project results. In short, the cooperation costs are reflected in pricing, services, agreements and profit distribution, which promote win-win cooperation between the platform and developers and promote the continuous development and innovation of the industrial software platform.

[0076] 1.3 Pricing Model

[0077] A pricing model is constructed to obtain the profit function and constraints of the industrial software service platform and the industrial software developer. Specifically, in this embodiment, a Stackelberg pricing model is constructed based on cost estimation results and market factors, and the method steps are as follows.

[0078] Assume that the total value of the industrial software customization project is V, the total project cost is C, the industrial software platform is the leader, and the software developer is the follower.

[0079] Assume: The platform sets the initial user deposit as P1, the mid-term user deposit as P1, and the final payment after the project is completed as P3, and: P = P1 + P2 + P3. The scope and constraints of P1, P2 and P3 are:

[0080] Considering the user's liquidity pressure and trust building, P1 should not be too high, but should be enough to cover the developer's initial startup costs and part of the platform's management expenses;

[0081] The setting of P2 should correspond to key milestones in the middle of the project, such as completing the development of core software functions and passing internal testing, while motivating developers to continue to invest resources;

[0082] As the final payment, P3 must, under the premise of protecting the rights and interests of users, effectively constrain the developer to ensure that it completes the project's finishing work and after-sales support obligations.

[0083] In this way, the profit function of the industrial software service platform can be obtained as:

[0084]

[0085] Where: platform is the profit of the industrial software service platform; C0 is the total cost of the project; r platform is the proportion of the operating cost of the industrial software service platform to the total project revenue; E is the market demand elasticity coefficient; V avg is the average price of similar projects in the market; k is the comprehensive cost coefficient; P1 is the deposit paid by users in the early stage of the project; P2 is the deposit paid by users in the middle stage of the project; P3 is the balance paid by users after the project is completed;

[0086] The profit function of industrial software developers is:

[0087]

[0088] Where: I developer is the profit function of industrial software developers; r d The actual profit retention ratio of industrial software developers after deducting various costs; C dis the software development cost of industrial software developers; E d is the market demand elasticity coefficient; V avgd The average price of similar projects in the market;

[0089] For industrial software developers, their profits must meet the minimum expected profit requirements:

[0090]

[0091] in: It is the minimum profit bottom line that developers are willing to take on a project;

[0092] For users, the perceived value of customized software must be greater than the user's psychological expected price:

[0093]

[0094] Where: V is the price of customized software; The price expected by the user.

[0095] 1.4. Analyze the pricing model

[0096] In the profit function of the industrial software service platform, (P1+P2+P3-C0) represents the remaining part after deducting the basic cost from the project income, (1-r platform ) is the profit retention ratio after considering the platform operating costs, and k(P1+P2+P3) comprehensively considers the cost factors corresponding to the deposit at each stage. The market demand elasticity coefficient E is calculated by This item is used to adjust the relationship between price and demand. As P1+P2+P3 (i.e. the total project price) increases, The value of will change according to the size of E, thus reflecting the combined impact of price increases on market demand and platform revenue. If E is large, the value of this item will drop quickly when prices rise, which means that the platform needs to be more cautious about increasing the inhibitory effect of prices on demand to avoid a reduction in total revenue.

[0097] In the profit function of industrial software developers, r d It is used to calculate the actual profit retention ratio of the developer after deducting various costs. d ) is used to adjust the developer’s gross income to net profit after deducting costs, thereby more accurately reflecting the developer’s actual income from the project. d If the value is large, it means that the developer's cost accounts for a high proportion of its income, and its profit margin is relatively small; on the contrary, if r d The smaller the value of E, the higher the profit the developer can obtain at the same income level. dIt has a similar function to E, helping software developers to comprehensively consider the balance between their own profits and market demand in pricing decisions, and to formulate reasonable pricing strategies according to different market demand elasticity constraints to maximize profits.

[0098] 1.5. Optimal game decision analysis based on pricing model

[0099] In the Stackelberg game, since the core enterprise is the industrial software service platform, the "leader" is set as the industrial software service platform, and the "follower" is the industrial software developer. According to the Stackelberg game principle, the industrial software service platform has a certain advantage as a leader. It can first observe the market environment and formulate its own strategy. After the leader makes a decision, the industrial software developer makes its own strategy based on the leader's decision. By establishing the interest model of the two subjects, the optimal price, maximum benefit and overall maximum benefit of both can be finally obtained.

[0100] 1.5.1. Analyze the optimal decision of industrial software developers and industrial software service platforms under independent operation conditions to obtain the optimal price and maximum benefit of industrial software developers and industrial software service platforms.

[0101] 1.5.1.1. Optimal Decision Making for Independent Operation of Industrial Software Platforms

[0102] Under the above constraints, the platform maximizes its own profit Π by optimizing pricing (i.e. determining the values ​​of P1, P2, and P3). platform .

[0103] The profit function of the industrial software service platform is:

[0104]

[0105] (1) The best price for users to pay deposits in the early stages of the project

[0106] Yes platform Taking the first-order derivative of P1 we get:

[0107]

[0108] Again platform Taking the second-order derivative of P1 we get:

[0109]

[0110] According to the strict definition of concave function: if in (a, b) Then f(x) is a strictly concave function in (a, b). Since E>0, we know that So platformis a strictly concave function of the industrial software platform price P1, Π platform There is a maximum value, namely ∏ platform When the first-order derivative function is zero, the optimal price of the industrial software platform can be obtained. According to the Stackelberg game, P1 at this time is the optimal price.

[0111] (2) The best price for users to pay deposits in the middle of the project

[0112] Yes platform Taking the first-order derivative of P2 we get:

[0113]

[0114] Then take the second-order derivative of P2 and we get:

[0115]

[0116] Since E>0, we know So platform is a strictly concave function of the industrial software platform price P2, Π platform There is a maximum value, namely Π platform When the first-order derivative function is zero, the optimal price of the industrial software platform can be obtained. According to the Stackelberg game, P2 is the optimal price.

[0117] (3) The best price for the user to pay the balance after the project is completed

[0118] Yes platform Taking the first-order derivative of P3 we get:

[0119]

[0120] Then take the second-order derivative of P3 and we get:

[0121]

[0122] Since E>0, we know So platform is a strictly concave function of the industrial software platform price P3, Π platform There is a maximum value, namely Π platform When the first-order derivative function is zero, the optimal price of industrial software developers can be obtained. According to Stackelberg game, P3 is the optimal price.

[0123] To simplify the calculation, assume:

[0124]

[0125] We can get:

[0126]

[0127] Will Substitute it into the profit function of the industrial software service platform to obtain the maximum benefit of the industrial software platform

[0128]

[0129] in: To maximize the profits of the industrial software service platform under independent operation; The best price for an industrial software platform operating independently; The best price for early stage users to pay deposits for projects under independent operation of industrial software platforms; The best price for mid-term project users to pay deposits under independent operation of industrial software platforms; The best price for users to pay the balance after the project is completed under the independent operation of the industrial software platform.

[0130] 1.5.1.2. Optimal Decisions for Industrial Software Developers Under Independent Operation

[0131] The profit function of industrial software developers is:

[0132]

[0133] (1) The best price for users to pay deposits in the early stages of the project

[0134] Yes developer Taking the first-order derivative of P1 we get:

[0135]

[0136] Then take the second-order derivative of P1 of the function, and we get:

[0137]

[0138] Since E>0, we know So dveloper is a strictly concave function of the price P1 of industrial software developers, Π developer There is a maximum value, namely Π developer When the first-order derivative function is zero, the optimal price of industrial software developers can be obtained. According to the Stackelberg game, P1 is the optimal price.

[0139] (2) The best price for users to pay deposits in the middle of the project

[0140] Yesdeveloper Taking the first-order derivative of P2 we get:

[0141]

[0142] Taking the second-order derivative of P2 for this function, we get:

[0143]

[0144] Since E>0, we know So developer is a strictly concave function of the price P2 of industrial software developers, Π developer There is a maximum value, namely Π developer When the first-order derivative function is zero, the optimal price of industrial software developers can be obtained. According to Stackelberg game, P2 is the optimal price.

[0145] (3) The best price for the user to pay the balance after the project is completed

[0146] Yes developer Taking the first-order derivative of P3 we get:

[0147]

[0148] Taking the second-order derivative of P2 for this function, we get:

[0149]

[0150] Since E>0, we know So developer is a strictly concave function of the price P3 of industrial software developers, ∏ developer There is a maximum value, namely ∏ developer When the first-order derivative function is zero, the optimal price of industrial software developers can be obtained. According to Stackelberg game, P3 is the optimal price.

[0151] To simplify the calculation, assume:

[0152]

[0153] but:

[0154]

[0155] Will Substituting this into the profit function of the industrial software developer, we can get the maximum profit of the industrial software developer under independent operation.

[0156]

[0157] in: To maximize the profits of industrial software developers operating independently; The best price for industrial software developers operating independently; The best price for early stage users to pay deposits for projects operated independently by industrial software developers; The best price for mid-term users to pay deposits for projects under independent operation of industrial software developers; The best price for users to pay the final payment after the project is completed when the industrial software developer is independently operated.

[0158] 1.5.2. Optimal Decisions under Cooperation between Software Developers and Industrial Software Platforms

[0159] Analyze the optimal decisions of industrial software developers and industrial software service platforms under cooperative operation conditions to obtain the overall maximum benefits of industrial software developers and industrial software service platforms.

[0160] In the case of full cooperation, the industrial software developer and the industrial software service platform form an alliance, and the three parties jointly decide At this point, the overall interest function of industrial software developers and industrial software service platforms is:

[0161]

[0162] (1) The best price for users to pay deposits in the early stages of the project

[0163] Yes + Taking the first-order derivative of P1 we get:

[0164]

[0165] Yes + Taking the second-order derivative of P1 we get:

[0166]

[0167] Since E>0, we know So + is a strictly concave function of the price P1 of industrial software developers, Π + There is a maximum value, namely Π + When the first-order derivative is zero, the optimal price for cooperation can be obtained. According to Stackelberg game, P1 is the optimal price.

[0168] (2) The best price for users to pay deposits in the middle of the project

[0169] Yes + Taking the first-order derivative of P2 we get:

[0170]

[0171] Yes + Taking the second-order derivative of P2 we get:

[0172]

[0173] Since E>0, we know So + is a strictly concave function of the price P2 of industrial software developers, Π + There is a maximum value, namely Π + When the first-order derivative is zero, the optimal price for cooperation can be obtained. According to the Stackelberg game, P2 is the optimal price.

[0174] (3) The best price for the user to pay the balance after the project is completed

[0175] Yes + Taking the first-order derivative of P3 we get:

[0176]

[0177] Yes + Taking the second-order derivative of P3 we get:

[0178]

[0179] Since E>0, we know So + is a strictly concave function of the price P3 of industrial software developers, Π + There is a maximum value, namely ∏ + When the first-order derivative is zero, the optimal price for cooperation can be obtained. According to Stackelberg game, P3 is the optimal price.

[0180] To simplify the calculation, assume:

[0181]

[0182] but:

[0183]

[0184] Will Substituting this into the overall benefit function of industrial software developers and industrial software service platforms, we get:

[0185]

[0186] in: To maximize profits under cooperative management conditions; The best price under cooperative operation conditions; The best price for the early users of the project to pay the deposit under cooperative operation conditions; The best price for the mid-term users of the project to pay the deposit under cooperative operation conditions; The optimal price for users to pay the balance after the project is completed under cooperative operation conditions.

[0187] Step 2: Establish a benefit distribution model based on the Shapley value method.

[0188] Through the analysis of the cooperation results of various stakeholders, it can be seen that when industrial software developers and industrial software service platforms form a cooperative alliance, the benefits are the highest, and the benefits obtained by each stakeholder after the cooperation are also higher than the benefits when operating alone. Therefore, the industrial software service platform plays an important role as the core link between industrial software developers and users. Under information symmetry, the various subjects on the industrial Internet platform can maximize the overall benefits under centralized decision-making, and the interests of each subject in the industrial Internet platform need to be reasonably distributed. The distribution of benefits should not only meet the requirements that the benefits obtained by each subject in the industrial chain are greater than the benefits obtained under their independent decision-making, but also meet the fair distribution mechanism of equal contribution and benefit size. Based on the situation of cooperation within the industrial Internet platform under information symmetry, the maximum benefits generated by the three major subjects under different combinations are obtained by using the Stackelberg game as mentioned above as a known condition, and the Shapley value method is used to calculate the benefit distribution model of the initial industrial software developer and the industrial software service platform.

[0189] 2.1 Model Assumptions

[0190] The profit distribution of the industrial Internet platform considered in this embodiment is based on the cooperation of various entities, so the profit distribution of the entities needs to show the contribution of each entity to the total profit. The problem studied in this embodiment is the initial profit distribution plan of the various entities of the industrial Internet platform. The profit distribution needs to be calculated by the Shapley value method, and the assumptions required for model establishment are supplemented:

[0191] (1) Under information symmetry, the cooperative entities of the industrial Internet platform are industrial software developers and industrial software service platforms.

[0192] (2) Real factors such as resource input, risk assumption, and information transfer that arise when entities cooperate do not exist.

[0193] (3) The main players in the industrial chain all aim to maximize their own interests. They are all rational economic people with independent decision-making power.

[0194] (4) All members of the industry chain are risk neutral.

[0195] 2.2. Establishment of benefit distribution model

[0196] Assume that the two entities, industrial software developer and industrial software service platform, are numbered 1 and 2 respectively. The maximum benefits of the two entities under different cooperation situations are obtained through Stackelberg game, and then substituted into the Shapley value benefit allocation model. Specifically, the Shapley value benefit allocation model is expressed as:

[0197]

[0198] in: is the allocation obtained by the i-th member in the cooperation mode; w(|S|) is the weighting factor; S is the set of all alliances that include member i; |S| is the number of members in alliance S; v(S) is the maximum benefit of alliance S; v(S) is the result of removing member i from alliance S; n is the number of members.

[0199] 2.3. Obtaining Shapley values ​​for the distribution of benefits between industrial software developers and industrial software service platforms under different cooperation models

[0200] The overall maximum benefits of industrial software developers and industrial software service platforms are substituted into the Shapley value benefit distribution model to obtain the Shapley value of benefit distribution under the cooperation between industrial software developers and industrial software service platforms and the Shapley value of benefit distribution under the cooperation between industrial software service platforms and users.

[0201] Table 1 Benefit analysis of cooperation between industrial software developers and industrial software service platforms

[0202]

[0203] The above data can be used to obtain the Shapley value of the benefit distribution between industrial software developers and industrial software service platforms:

[0204]

[0205] in: The Shapley value for the benefits of cooperation between industrial software developers and industrial software service platforms; To maximize the profits of industrial software developers operating independently; To maximize the profits of the industrial software service platform under independent operation; It is the maximum common profit under the cooperative operation conditions between industrial software developers and industrial software service platforms.

[0206] Table 2 Benefit analysis of cooperation between industrial software service platform and industrial software developers

[0207]

[0208]

[0209] The above data can be used to obtain the Shapley value of the benefit distribution between the industrial software service platform and users:

[0210]

[0211] Among them: Among them: The Shapley value for the benefits of cooperation between industrial software developers and industrial software service platforms; The Shapley value for the benefit distribution between the industrial software service platform and users: To maximize the profits of industrial software developers operating independently; To maximize the profits of the industrial software service platform under independent operation; It is the maximum common profit under the cooperative operation conditions between industrial software developers and industrial software service platforms.

[0212] The Shapley value method is used to distribute the interests of industrial software developers and industrial software service platforms in the industrial Internet platform under the cooperative game situation. This distribution method takes into account the contribution rate of each subject, making the distribution plan more fair and reasonable.

[0213] Step 3: Benefit Decision Optimization

[0214] 31) Use historical data to fit the relationship function between price and the number of functional modules.

[0215] The historical data of the interaction between the developer and the platform is used to fit the relationship function between the price and the number of functional modules using a neural network. When the number of functional modules and the number of software are input, the price given by the platform to the developer can be obtained. The historical data of the interaction between the platform and the user is used to fit the relationship function between the number of functional modules, the purchase quantity and the price. When the number of functional modules and the purchase quantity are input, the price given by the platform to the user can be obtained, such as Figure 2 shown.

[0216] 32) Use reinforcement learning method to decide the optimal benefit distribution plan.

[0217] Use reinforcement learning (actor-critic) to make decisions and get the optimal benefits of the platform. The specific process is as follows: Function point analysis and pricing: The software functions are divided into multiple types such as external input (EI) and external output (EO). Different weights are assigned to each type of function point according to its complexity and the data items it contains, and then the number of unadjusted function points is calculated. Simple input functions have lower weights and correspondingly fewer unadjusted function points; complex calculation and output functions have higher weights and more unadjusted function points. Considering factors such as the technical characteristics and business complexity of the software, an adjustment factor is determined. These factors include system stability, security requirements, data processing complexity, user interface friendliness, etc. For example, if the software has very high requirements for data accuracy or needs to interact with multiple complex external systems, the adjustment factor will be increased accordingly. Multiply the number of unadjusted function points by the adjustment factor to get the final number of function points. The number of function points multiplied by the unit price of function points is the price of software development.

[0218] Price fitting based on historical data and neural networks: Use the historical data of interaction between developers and the platform to use neural networks to fit the relationship function between price and the number of functional modules. When the number of functional modules and the number of software are input, the price the platform gives to developers can be obtained. Use the historical data of interaction between the platform and users to fit the relationship function between the number of functional modules, purchase quantity and price. Similarly, when the number of functional modules and purchase quantity are input, the price the platform gives to users can be obtained.

[0219] like Figure 3 As shown, reinforcement learning decision-making: using the reinforcement learning (actor-critic) algorithm, according to the above-fitted price relationship function and the interaction data between the platform and users, continuously learn and optimize the decision-making strategy to obtain the optimal benefit of the platform.

[0220] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform, characterized in that: The steps include: Step 1: Analyze the optimal decision based on Stackelberg's master-slave game decision model 11) Construct a pricing model to obtain the profit function and constraints of the industrial software service platform and industrial software developers respectively; 12) Optimal game decision analysis based on pricing model 121) Analyze the optimal decision of industrial software developers and industrial software service platforms under independent operation conditions, and obtain the optimal price and maximum benefit of industrial software developers and industrial software service platforms; 122) Analyze the optimal decision of industrial software developers and industrial software service platforms under cooperative operation conditions to obtain the overall maximum benefits of industrial software developers and industrial software service platforms; Step 2: Establish a benefit distribution model based on the Shapley value method; Step 3: Optimize benefit decisions 31) Use historical data to fit the relationship function between price and the number of functional modules; 32) Use reinforcement learning method to decide the optimal benefit distribution plan.

2. According to the method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform in claim 1, it is characterized by: In step 11), the profit function of the industrial software service platform is: Where: platform is the profit of the industrial software service platform; C0 is the total cost of the project; r platform is the proportion of the operating cost of the industrial software service platform to the total project revenue; E is the market demand elasticity coefficient; V avg is the average price of similar projects in the market; k is the comprehensive cost coefficient; P1 is the deposit paid by users in the early stage of the project; P2 is the deposit paid by users in the middle stage of the project; P3 is the balance paid by users after the project is completed; The profit function of industrial software developers is: Among them: developer is the profit function of industrial software developers; r d The actual profit retention ratio of industrial software developers after deducting various costs; C d is the software development cost of industrial software developers; E d is the market demand elasticity coefficient; V avgd The average price of similar projects in the market; For industrial software developers, their profits must meet the minimum expected profit requirements: in: It is the minimum profit bottom line that developers are willing to take on a project; For users, the perceived value of customized software must be greater than the user's psychological expected price: Where: V is the price of customized software; The price expected by the user.

3. The method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform according to claim 2 is characterized in that: In step 121), the optimal decision for the industrial software platform to be independently operated is: in: To maximize the profits of the industrial software service platform under independent operation; The best price for an industrial software platform operating independently; The best price for early stage users to pay deposits for projects under independent operation of industrial software platforms; The best price for mid-term users to pay deposits for projects under independent operation of industrial software platforms; The best price for users to pay the balance after the project is completed under the independent operation of the industrial software platform.

4. The method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform according to claim 2 is characterized in that: In step 121), the optimal decision for the industrial software developer to operate independently is: in: To maximize the profits of industrial software developers operating independently; The best price for industrial software developers operating independently; The best price for early stage users to pay deposits for projects operated independently by industrial software developers; The best price for mid-term users to pay deposits for projects under independent operation of industrial software developers; The best price for users to pay the final payment after the project is completed when the industrial software developer is independently operated.

5. The method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform according to claim 2 is characterized in that: In step 122), the profit function of the industrial software developer and the industrial software service platform under cooperative operation conditions is: in: To maximize profits under cooperative management conditions; The best price under cooperative operation conditions; The best price for the early users of the project to pay the deposit under cooperative operation conditions; The best price for the mid-term users of the project to pay the deposit under cooperative operation conditions; The optimal price for users to pay the balance after the project is completed under cooperative operation conditions.

6. The method for constructing a profit distribution model and collaborative optimization under multiple operation modes of an industrial software service platform according to claim 1 is characterized in that: In the step 2, the method for establishing the benefit distribution model based on the Shapley value method is: The Shapley value benefit distribution model indicates that: in: is the allocation obtained by the i-th member in the cooperation mode; w(|S|) is the weighting factor; S is the set of all alliances that contain member i; |S| is the number of members in alliance S; v(S) is the maximum benefit of alliance S; v(S\i) is the maximum benefit of alliance S without member i; n is the number of members; Substituting the overall maximum benefits of industrial software developers and industrial software service platforms into the Shapley value benefit distribution model, we can obtain the Shapley value of benefit distribution under the cooperation between industrial software developers and industrial software service platforms and the Shapley value of benefit distribution under the cooperation between industrial software service platforms and users: in: The Shapley value for the benefits of cooperation between industrial software developers and industrial software service platforms; The Shapley value for the benefit distribution between the industrial software service platform and users: To maximize the profits of industrial software developers operating independently; To maximize the profits of the industrial software service platform under independent operation; It is the maximum common profit under the cooperative operation conditions between industrial software developers and industrial software service platforms.

Images