Manufacturing service cooperation negotiation method under dynamic pressure

Abstract

The invention discloses a manufacturing service cooperation negotiation method under dynamic pressure, which comprises the following steps: constructing a three-level pressure evaluation model under dynamic pressure based on the number of negotiators, proposal situations, demand urgency and other characteristics in manufacturing service cooperation; constructing a multi-attribute incomplete information game model based on the attribute change characteristics of the demander and the manufacturing service provider and a three-level pressure evaluation model; based on a Markov decision process, a Bayesian Nash equilibrium algorithm for seeking expected utility maximization is designed. The method considers that the demander and the service provider share the same manufacturing service negotiation scene, studies the multi-round negotiation problem under the incomplete information dynamic pressure condition formed by the demander and the manufacturing service provider, and has the advantages of improving the task completion rate, shortening the negotiation period of the manufacturing service and improving the negotiation efficiency of the manufacturing service. And meanwhile, the satisfaction degree of a demander and a manufacturing service provider is improved.

Description

Technical Field

[0001] The present invention belongs to the field of manufacturing service collaborative operation management, and in particular relates to a manufacturing service collaborative negotiation method under dynamic pressure. Background Art

[0002] With the deep integration of advanced digital technologies such as the Internet of Things, cloud computing, and artificial intelligence with the manufacturing industry, manufacturing service collaboration has become the main way to meet the growing demand for personalized and customized manufacturing tasks. Manufacturing service collaboration usually involves multiple stakeholders, including demanders, manufacturing service providers, and operators, each of whom has different collaboration goals and preferences. In order to balance the differentiated and even conflicting collaboration needs between demanders and manufacturing service providers, negotiation is one of the most effective methods.

[0003] Negotiation research in manufacturing service collaboration focuses on the formulation of negotiation mechanisms and strategies between demanders and manufacturing service providers to maximize their respective utilities. Most studies focus on how demanders and manufacturing service providers negotiate on static collaboration prices, delivery times, and throughput. However, manufacturing service collaboration is dynamic in nature. Faced with different levels of pressure in manufacturing service collaboration, influenced by the inherent conditions of demanders, manufacturing service providers, and the collaborative environment, the negotiators of manufacturing service collaboration will choose different strategies to achieve their respective optimal utilities.

[0004] Utility is an important indicator to measure the satisfaction of negotiators, which is determined by the negotiation results under different negotiation pressures. In manufacturing service collaboration, it is still a challenge to determine a negotiation result that can provide high utility and balance for all negotiators. In manufacturing service collaboration, the manufacturing tasks submitted by demanders are often complex and diverse. These tasks can be further decomposed into different types of subtasks, which are then executed by the corresponding manufacturing service providers. In response to different subtask requirements, demanders and manufacturing service providers usually have to conduct multiple rounds of negotiations to obtain the final task plan. Most existing studies optimize service prices and service quality from a demander-centered or manufacturing service provider-centered perspective, ignoring the potential for both parties to balance utility and achieve a win-win situation. In addition, the dynamic negotiation pressure conditions faced by demanders and manufacturing service providers in the multi-round negotiation process have not been fully considered, which to some extent hinders the balance and stability of manufacturing service negotiations. In the actual negotiation process, demanders and manufacturing service providers will engage in dynamic bargaining based on their own pressures, with the goal of both parties achieving their respective expected goals in the case of incomplete information. However, existing manufacturing service collaboration technical solutions make it difficult for demanders and manufacturing service providers to achieve their respective goals together. Therefore, the present invention proposes a manufacturing service collaboration negotiation method under dynamic pressure, which can achieve balanced negotiation between the two parties under dynamic pressure changes. Summary of the invention

[0005] In order to solve the above technical problems and based on the negotiation problem of multi-round manufacturing service collaboration under dynamic pressure conditions with incomplete information, the present invention proposes a manufacturing service collaboration negotiation method under dynamic pressure, which adopts a three-level pressure evaluation model and an incomplete information game model, and uses a membership function based on Gaussian distribution to evaluate the pressure level based on different negotiation pressure level indexes to guide further strategy selection. At the same time, conservative strategies, balanced strategies and aggressive strategies based on time concession strategies are proposed, and a dynamic negotiation optimization model in manufacturing service collaboration is constructed. On this basis, a Bayesian Nash equilibrium search method for deriving future expected utility is proposed. The dynamic negotiation in manufacturing service collaboration aims to reach a consensus on attributes such as price, delivery time, and throughput through negotiation between demanders and manufacturing service providers, and maximize their respective expected utilities under different pressure conditions, thereby achieving efficient collaboration.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A manufacturing service collaborative negotiation method under dynamic pressure comprises the following steps:

[0008] Step (1), based on the negotiation scenario of a demander and multiple manufacturing service providers, the negotiation pressure sources of manufacturing services are analyzed, and a negotiation pressure level index is introduced. The membership function based on Gaussian distribution is used to evaluate the probability under different pressure levels. The negotiation pressure of manufacturing service collaboration is divided into three levels: low, medium and high.

[0009] Step (2): Based on the attribute change characteristics of the demander and the manufacturing service provider and the evaluation results of the three-level pressure, a negotiation strategy model based on the time concession principle, a utility function model based on negotiation pressure-negotiation strategy matching, and an incomplete information game model are constructed;

[0010] Step (3): Based on the negotiation pressure changes and dynamic pressure evaluation of manufacturing services, the negotiation process of manufacturing services is modeled as a Markov decision process, and on this basis, a Bayesian Nash equilibrium algorithm that maximizes utility is sought. Based on the Bayesian Nash equilibrium algorithm, the optimal strategy of each negotiator in the negotiation process of multiple rounds of manufacturing services is determined.

[0011] Furthermore, the step (1) comprises:

[0012] Step (1.1) is based on a negotiation scenario between a demander and multiple manufacturing service providers, analyzing detailed parameters under different pressure sources in manufacturing service collaboration, including the number of negotiation opponents, differences in competitor proposals, and inherent conditions, and based on this, calculating the pressure level index faced by the demander and the manufacturing service provider;

[0013] Step (1.2) uses a membership function based on Gaussian distribution to divide the negotiation pressure of manufacturing services into three levels: low, medium, and high pressure. The pressure level index is used as a measure of the pressure of demanders and manufacturing service providers. Based on the negotiation pressure level index of different manufacturing services, the probability distribution of different pressure levels is evaluated. The pressure evaluation conditional distribution function based on the negotiation pressure level index is:

[0014] (3)

[0015] in, is the pressure level, is the stress level index, , , Represent the probability distribution under low pressure, medium pressure and high pressure respectively.

[0016] Furthermore, the step (2) comprises:

[0017] Step (2.1) is based on the three-level stress assessment results of step (1.2), and sets two scenarios for modeling according to the attribute change characteristics of the demander and the manufacturing service provider. Scenario 1 is to model with the demander's attributes and price / time as the change quantity, or to model with the manufacturing service provider's attributes and throughput as the change quantity; Scenario 2 is to model with the demander's attributes and throughput as the change quantity, or to model with the manufacturing service provider's attributes and price / time as the change quantity.

[0018] Furthermore, the step (2) further includes:

[0019] Step (2.2) considers that under the three-level pressure of manufacturing service collaboration, the utility functions of demanders and manufacturing service providers are different at different pressure levels, and negotiators of different manufacturing services usually have different utility functions according to their attribute characteristics; demanders and manufacturing service providers select the optimal negotiation strategy by evaluating the negotiation pressure of manufacturing service collaboration, and different pressure-strategy matching relationships lead to different utilities; when the negotiation strategy of manufacturing services matches the negotiation pressure level, the utility function is monotonically increasing or monotonically decreasing; when the negotiation strategy of manufacturing services does not match the negotiation pressure level, the utility function will be adjusted by adding reward or penalty clauses.

[0020] Furthermore, the step (2) further includes:

[0021] Step (2.3) Under the three-level pressure in manufacturing service collaboration, the demander and the manufacturing service provider establish a negotiation strategy set, which includes conservative strategy, balanced strategy and aggressive strategy; the choice of negotiation strategy is based on negotiation time and is adopted in the negotiation process of manufacturing service collaboration.

[0022] Furthermore, the step (2) further includes:

[0023] Step (2.4) Based on the negotiation strategy and expected utility of manufacturing services, the expected utility function of the demander and the manufacturing service provider is defined as:

[0024] (4)

[0025] in, is the individual utility function, Proposals for individuals The utility of It is the negotiation strategy of individuals under different stress levels. To choose a negotiation strategy The generated proposals, i represents the individual number, j represents the strategy number, is the pressure level, Conditional distribution function for stress assessment based on negotiated stress level index;

[0026] The optimization model belongs to multi-objective and multi-constraint optimization. The negotiation optimization model and constraints of dynamic manufacturing services are defined as:

[0027] (5)

[0028] (6)

[0029] (7)

[0030] (8)

[0031] (9)

[0032] (10)

[0033] Among them, argmax is a function that finds the parameter that makes the function value the largest. Refers to the individual utility of the demander. refers to the individual utility of the provider, Choose a negotiation strategy for the demander The generated proposals, Selecting a negotiation strategy for a provider The generated proposals, , Represents the stress level index of demanders and providers respectively, , Represent the satisfaction of demanders and providers on price attributes, , Respectively represent the satisfaction of demanders and providers with respect to time attributes, , They represent the satisfaction of the demander and provider on the throughput attribute, , Represent the lower and upper price requirements respectively, Represent the lower and upper limits of the delivery time, respectively. , They represent the lower and upper requirements of throughput respectively, , , They represent the price, delivery time, and throughput of proposal generation respectively;

[0034] Formula (5) and Formula (6) respectively express the optimization objectives of the demander and the manufacturing service provider, maximizing their expected utility; Formula (7) expresses the conditions for the demander and the manufacturing service provider to reach a negotiation agreement; Formula (8)-Formula (10) ensure that the proposal does not exceed their respective expected ranges.

[0035] Furthermore, the step (3) includes:

[0036] Step (3.1) seeks the Bayesian Nash equilibrium state in the manufacturing service collaboration based on the above pressure assessment model and optimization model; for the Markov decision process, the state set is described as the negotiation information, the action set is described as the negotiation strategy that the negotiator of the manufacturing service can choose in each state, the state transition probability is described as the change of negotiation pressure, and the reward function is the expected utility of each negotiation round; the goal of the optimization problem is to find the optimal strategy that maximizes the reward; in the dynamic negotiation process, the Bayesian Nash equilibrium state is:

[0037] (11)

[0038] in, is the individual utility function, are negotiation strategies under different stress levels, , Select negotiation strategies for , Proposals generated; is the optimal negotiation strategy that can bring the greatest benefit in each round of negotiation. Other negotiation strategies in each negotiation round;

[0039] Find the optimal solution through changes in negotiation strategies for manufacturing service collaboration and dynamic negotiation pressure assessment.

[0040] Furthermore, the step (3) includes:

[0041] Step (3.2) is based on the Bayesian Nash equilibrium search method of future expected utility to negotiate the manufacturing service collaboration. The demander and the manufacturing service provider first determine the demand and service information; next, the demander and the manufacturing service provider will put forward preliminary proposals to complete each subtask; if the initial proposal for a specific subtask enables the manufacturing service negotiator to successfully reach a negotiation, the proposal will be added to the final negotiation result list; otherwise, the negotiation between the demander and the manufacturing service provider of the subtask will continue.

[0042] Furthermore, in the negotiation of dynamic manufacturing service collaboration, the negotiators first conduct a negotiation pressure assessment; then, based on the assessed pressure and the incomplete information obtained, they determine the optimal strategy for the current round; finally, they select the relatively optimal strategy that can produce greater expected utility; the negotiation process of manufacturing services continues to cycle and the model is updated until a successful negotiation is reached, and then the negotiation results are added to the final negotiation result list. When the negotiations of all subtasks are completed, the negotiation process ends.

[0043] The beneficial effects of the present invention compared with the prior art are:

[0044] The design strategy of the present invention can select effective ways to achieve a win-win situation in terms of price and service quality attributes. It evaluates the probability under different pressure levels based on the negotiation pressure level values ​​of different manufacturing services to guide further strategic choices, proposes conservative strategies, balanced strategies and radical strategies based on time concession strategies, and constructs a dynamic negotiation optimization model in manufacturing service collaboration. On this basis, a Bayesian Nash equilibrium search algorithm for deriving future expected utility is proposed. The present invention is of great value in improving the task completion rate, significantly shortening the negotiation cycle of manufacturing services, and improving the satisfaction of demanders and manufacturing service providers. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flowchart of a manufacturing service collaborative negotiation method under dynamic pressure of the present invention;

[0046] Figure 2 Flowchart of the negotiation process for seeking manufacturing services based on the Bayesian Nash equilibrium search algorithm for deriving future expected utility. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] like Figure 1 As shown, a manufacturing service collaborative negotiation method under dynamic pressure of the present invention specifically includes the following steps:

[0049] Step (1): Based on the number of negotiators, proposal situations, and demand urgency in manufacturing service collaboration, a three-level pressure assessment model under dynamic pressure is constructed, including:

[0050] Based on the negotiation scenario between a demander and multiple manufacturing service providers, the negotiation pressure sources of manufacturing services are analyzed. The negotiation pressure level index is introduced. The membership function based on Gaussian distribution is used to divide the negotiation pressure of manufacturing services into three levels: low, medium and high. Based on the values ​​of negotiation pressure levels of different manufacturing services, the probabilities under different pressure levels are evaluated.

[0051] Step (2): Based on the attribute change characteristics of the demander and the manufacturing service provider, a negotiation strategy model based on the time concession principle and a utility function model based on negotiation pressure-negotiation strategy matching are constructed. Based on the attribute change characteristics of the demander and the manufacturing service provider, as well as the evaluation results of the three-level pressure, an incomplete information game model is constructed.

[0052] Step (3): Based on the Markov decision process, a Bayesian Nash equilibrium algorithm is designed to maximize the expected utility, including:

[0053] Based on the negotiation pressure changes and dynamic pressure evaluation of manufacturing services, the negotiation process of manufacturing services is modeled as a Markov decision process, and on this basis, a Bayesian Nash equilibrium algorithm with utility maximization is sought. The optimal strategy of each negotiator in the negotiation process of multiple rounds of manufacturing services is determined based on the Bayesian Nash equilibrium algorithm.

[0054] Furthermore, the step (1) specifically includes:

[0055] Step (1.1) is based on a negotiation scenario between a demander and multiple manufacturing service providers. The detailed parameters under different pressure sources such as the number of negotiation opponents for manufacturing services, differences in competitor proposals, inherent conditions (such as differences in negotiation rounds and initial proposals for manufacturing services) are analyzed. Based on this, the pressure level index faced by the demander and the manufacturing service provider is calculated:

[0056] (1)

[0057] (2)

[0058] in, is the demander stress level index, To create a stress level index for service providers, Indicates the number of negotiation opponents faced by the demander, Indicates that demanders and manufacturing service providers The difference in proposals Indicates the remaining negotiation time. Indicates the urgency of manufacturing service negotiation, Manufacturing service provider and demanders in The difference in proposals Manufacturing service provider and other competitors in The difference in proposals Indicates the remaining negotiation time. Indicates the urgency of successful negotiation; , , , is the weight coefficient of the four negotiation pressure factors of the demander; , , , The weight coefficients of the four negotiation pressure factors for service providers.

[0059] Step (1.2) Since the three-level stress assessment model has high requirements for accuracy, a membership function based on Gaussian distribution is selected to divide the negotiation pressure of manufacturing services into three levels: low, medium, and high pressure. The stress level index is used as a measure of the stress of demanders and manufacturing service providers; based on the stress level index of negotiators of different manufacturing services, the probability distribution under different stress levels is evaluated.

[0060] Therefore, the conditional distribution function of the stress assessment of the negotiator stress level index based on manufacturing services is:

[0061] (3)

[0062] in, is the pressure level, is the stress level index, , , Represent the probability distribution under low pressure, medium pressure and high pressure respectively.

[0063] Furthermore, the step (2) specifically includes:

[0064] Step (2.1) is based on the three-level stress assessment results of step (1.2), and two scenarios are set for modeling according to the attribute change characteristics of the demander and the manufacturing service provider. Scenario 1 is to model with the demander's attributes and price / time as the change amount, or to model with the manufacturing service provider's attributes and throughput as the change amount; Scenario 2 is to model with the demander's attributes and throughput as the change amount, or to model with the manufacturing service provider's attributes and price / time as the change amount;

[0065] Step (2.2) considers that under the three-level pressure of manufacturing service collaboration, the utility functions of demanders and manufacturing service providers are different at different pressure levels, and negotiators of different manufacturing services usually have different utility functions according to their attribute characteristics; demanders and manufacturing service providers select the optimal negotiation strategy by evaluating the negotiation pressure of manufacturing service collaboration, and different pressure-strategy matching relationships lead to different utilities; when the negotiation strategy of manufacturing services matches the negotiation pressure level, the utility function is monotonically increasing or monotonically decreasing; when the negotiation strategy of manufacturing services does not match the negotiation pressure level, the utility function will be adjusted by adding reward or penalty clauses.

[0066] Step (2.3) Under the three-level pressure in manufacturing service collaboration, demanders and manufacturing service providers establish a negotiation strategy set, which includes conservative strategy, balanced strategy and radical strategy. The selection of these concession negotiation strategies is mainly based on negotiation time and is widely adopted in the negotiation process of manufacturing services.

[0067] Step (2.4) Because under different pressures, the demander and the manufacturing service provider adopt different negotiation strategies for manufacturing services, and will obtain different utilities. The expected utility of each strategy effectively reflects the average utility that the manufacturing service negotiator can obtain by choosing this strategy under dynamic pressure. Therefore, based on the negotiation strategy and expected utility of manufacturing services, the expected utility function of the demander and the manufacturing service provider can be defined as:

[0068] (4)

[0069] in, is the individual utility function, Proposals for individuals The utility of It is the negotiation strategy of individuals under different stress levels. To choose a negotiation strategy The generated proposals, i represents the individual number, j represents the strategy number, is the pressure level, Conditional distribution function for stress assessment based on the negotiated stress level index.

[0070] Since it involves multiple negotiators of manufacturing services and negotiation of multiple attributes, the optimization model of the present invention belongs to multi-objective and multi-constraint optimization. The negotiation optimization model and constraints of dynamic manufacturing services can be defined as:

[0071] (5)

[0072] (6)

[0073] (7)

[0074] (8)

[0075] (9)

[0076] (10)

[0077] Among them, argmax is a function that finds the parameter that makes the function value the largest. Refers to the individual utility of the demander. refers to the individual utility of the provider, Choose a negotiation strategy for the demander The generated proposals, Selecting a negotiation strategy for a provider The generated proposals, , Represents the stress level index of demanders and providers respectively, , Represent the satisfaction of demanders and providers on price attributes, , Respectively represent the satisfaction of demanders and providers with respect to time attributes, , They represent the satisfaction of the demander and provider on the throughput attribute, , Represent the lower and upper price requirements respectively, Represent the lower and upper limits of the delivery time, respectively. , They represent the lower and upper requirements of throughput respectively, , , They represent the price, delivery time, and throughput of proposal generation respectively;

[0078] Formula (5) and Formula (6) respectively express the optimization objectives of the demander and the manufacturing service provider, maximizing their expected utility; Formula (7) expresses the conditions for the demander and the manufacturing service provider to reach a negotiation agreement; Formula (8)-Formula (10) ensure that the proposal does not exceed their respective expected ranges.

[0079] Furthermore, the step (3) specifically includes:

[0080] Step (3.1) seeks the Bayesian Nash equilibrium state in the manufacturing service collaboration based on the above pressure assessment model and optimization model; for the Markov decision process, the state set is described as the negotiation information, the action set is described as the negotiation strategy that the negotiator of the manufacturing service can choose in each state, the state transition probability is described as the change of negotiation pressure, and the reward function is the expected utility of each negotiation round; the goal of the optimization problem is to find the optimal strategy that maximizes the reward; in the dynamic negotiation process, the Bayesian Nash equilibrium state is:

[0081] (11)

[0082] in, is the individual utility function, are negotiation strategies under different stress levels, , Select negotiation strategies for , Generated proposals. is the optimal negotiation strategy that can bring the greatest benefit in each round of negotiation. Other negotiation strategies in each round of negotiation.

[0083] The optimal solution is sought through changes in the negotiation strategy of manufacturing service collaboration and dynamic pressure evaluation. Due to the impact of time on utility, the utility obtained during negotiation will decay as the number of negotiation rounds increases. In the case of incomplete information, the negotiators of manufacturing service collaboration may judge that the optimal strategy in the Bayesian Nash equilibrium of the current round cannot make the negotiation of manufacturing service collaboration successful. However, other non-optimal strategies in the current round may lead to success. In this case, the negotiators of manufacturing services will compare the future expected utility of the optimal strategy with the non-optimal strategy. Then, choose the strategy that can bring greater utility after the negotiation is successful.

[0084] Step (3.2) Based on the Bayesian Nash equilibrium search method of future expected utility, the negotiation of manufacturing service collaboration is carried out. The demander and the manufacturing service provider first determine the demand and service information. Next, the demander and the manufacturing service provider will put forward preliminary proposals to complete each subtask. If the initial proposal for a particular subtask enables the negotiators of the manufacturing service collaboration to successfully reach a negotiation, the proposal will be added to the final negotiation result list. Otherwise, the negotiation between the demander and the manufacturing service provider of the subtask will continue. In the negotiation of dynamic manufacturing service collaboration, the negotiators first conduct a negotiation pressure assessment; then, based on the assessed pressure and the incomplete information obtained, the optimal strategy for the current round is determined. Finally, the relatively optimal strategy that can generate greater expected utility is selected. The negotiation process of manufacturing services continues to cycle and update the Bayesian model until a successful negotiation is reached, and then the negotiation results are added to the final negotiation result list. When all negotiations for all subtasks are completed, the negotiation process ends.

[0085] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in this field.

[0086] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A manufacturing service collaboration and negotiation method under dynamic pressure, characterized in that: The steps include: Step (1), based on the negotiation scenario of a demander and multiple manufacturing service providers, the negotiation pressure sources of manufacturing services are analyzed, and a negotiation pressure level index is introduced. The membership function based on Gaussian distribution is used to evaluate the probability under different pressure levels. The negotiation pressure of manufacturing service collaboration is divided into three levels: low, medium and high. Step (2): Based on the attribute change characteristics of the demander and the manufacturing service provider and the evaluation results of the three-level pressure, a negotiation strategy model based on the time concession principle, a utility function model based on negotiation pressure-negotiation strategy matching, and an incomplete information game model are constructed; Step (3): Based on the negotiation pressure changes and dynamic pressure evaluation of manufacturing services, the negotiation process of manufacturing services is modeled as a Markov decision process, and on this basis, a Bayesian Nash equilibrium algorithm that maximizes utility is sought. Based on the Bayesian Nash equilibrium algorithm, the optimal strategy of each negotiator in the negotiation process of multiple rounds of manufacturing services is determined.

2. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 1, characterized in that: The step (1) comprises: Step (1.1) is based on a negotiation scenario between a demander and multiple manufacturing service providers, analyzing detailed parameters under different pressure sources in manufacturing service collaboration, including the number of negotiation opponents, differences in competitor proposals, and inherent conditions, and based on this, calculating the pressure level index faced by the demander and the manufacturing service provider; Step (1.2) uses a membership function based on Gaussian distribution to divide the negotiation pressure of manufacturing services into three levels: low, medium, and high pressure. The pressure level index is used as a measure of the pressure of demanders and manufacturing service providers. Based on the negotiation pressure level index of different manufacturing services, the probability distribution of different pressure levels is evaluated. The pressure evaluation conditional distribution function based on the negotiation pressure level index is: （3） in, is the pressure level, is the stress level index, , , Represent the probability distribution under low pressure, medium pressure and high pressure respectively.

3. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 2 is characterized in that: The step (2) comprises: Step (2.1) is based on the three-level stress assessment results of step (1.2), and sets two scenarios for modeling according to the attribute change characteristics of the demander and the manufacturing service provider. Scenario 1 is to model with the demander's attributes and price / time as the change quantity, or to model with the manufacturing service provider's attributes and throughput as the change quantity; Scenario 2 is to model with the demander's attributes and throughput as the change quantity, or to model with the manufacturing service provider's attributes and price / time as the change quantity.

4. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 3 is characterized in that: The step (2) further includes: Step (2.2) considers that under the three-level pressure of manufacturing service collaboration, the utility functions of demanders and manufacturing service providers are different at different pressure levels, and negotiators of different manufacturing services usually have different utility functions according to their attribute characteristics; demanders and manufacturing service providers select the optimal negotiation strategy by evaluating the negotiation pressure of manufacturing service collaboration, and different pressure-strategy matching relationships lead to different utilities; when the negotiation strategy of manufacturing services matches the negotiation pressure level, the utility function is monotonically increasing or monotonically decreasing; when the negotiation strategy of manufacturing services does not match the negotiation pressure level, the utility function will be adjusted by adding reward or penalty clauses.

5. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 4, characterized in that: The step (2) further includes: Step (2.3) Under the three-level pressure in manufacturing service collaboration, the demander and the manufacturing service provider establish a negotiation strategy set, which includes conservative strategy, balanced strategy and aggressive strategy; the choice of negotiation strategy is based on negotiation time and is adopted in the negotiation process of manufacturing service collaboration.

6. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 5, characterized in that: The step (2) further includes: Step (2.4) Based on the negotiation strategy and expected utility of manufacturing services, the expected utility function of the demander and the manufacturing service provider is defined as: （4） in, is the individual utility function, Proposals for individuals The utility of It is the negotiation strategy of individuals under different stress levels. To choose a negotiation strategy The generated proposals, i represents the individual number, j represents the strategy number, is the pressure level, Conditional distribution function for stress assessment based on negotiated stress level index; The optimization model belongs to multi-objective and multi-constraint optimization. The negotiation optimization model and constraints of dynamic manufacturing services are defined as: （5） （6） （7） （8） （9） （10） Among them, argmax is a function that finds the parameter that makes the function value the largest. Refers to the individual utility of the demander. refers to the individual utility of the provider, Choose a negotiation strategy for the demander The generated proposals, Selecting a negotiation strategy for a provider The generated proposals, , Represents the stress level index of demanders and providers respectively, , Represent the satisfaction of demanders and providers on price attributes, , Respectively represent the satisfaction of demanders and providers with respect to time attributes, , They represent the satisfaction of the demander and provider on the throughput attribute, , Represent the lower and upper price requirements respectively, Represent the lower and upper limits of the delivery time, respectively. , They represent the lower and upper requirements of throughput respectively, , , They represent the price, delivery time, and throughput of proposal generation respectively; Formula (5) and Formula (6) respectively express the optimization objectives of the demander and the manufacturing service provider, maximizing their expected utility; Formula (7) expresses the conditions for the demander and the manufacturing service provider to reach a negotiation agreement; Formula (8)-Formula (10) ensure that the proposal does not exceed their respective expected ranges.

7. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 6, characterized in that: The step (3) comprises: Step (3.1) seeks the Bayesian Nash equilibrium state in the manufacturing service collaboration based on the above pressure assessment model and optimization model; for the Markov decision process, the state set is described as the negotiation information, the action set is described as the negotiation strategy that the negotiator of the manufacturing service can choose in each state, the state transition probability is described as the change of negotiation pressure, and the reward function is the expected utility of each negotiation round; the goal of the optimization problem is to find the optimal strategy that maximizes the reward; in the dynamic negotiation process, the Bayesian Nash equilibrium state is: (11) in, is the individual utility function, are negotiation strategies under different stress levels, , Select negotiation strategies for , Proposals generated; is the optimal negotiation strategy that can bring the greatest benefit in each round of negotiation. Other negotiation strategies in each negotiation round; Find the optimal solution through changes in negotiation strategies for manufacturing service collaboration and dynamic negotiation pressure assessment.

8. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 7, characterized in that: The step (3) comprises: Step (3.2) is based on the Bayesian Nash equilibrium search method of future expected utility to negotiate the manufacturing service collaboration. The demander and the manufacturing service provider first determine the demand and service information; next, the demander and the manufacturing service provider will put forward preliminary proposals to complete each subtask; if the initial proposal for a specific subtask enables the manufacturing service negotiator to successfully reach a negotiation, the proposal will be added to the final negotiation result list; otherwise, the negotiation between the demander and the manufacturing service provider of the subtask will continue.

9. The manufacturing service collaborative negotiation method under dynamic pressure according to claim 8, characterized in that: In the negotiation of dynamic manufacturing service collaboration, the negotiators first evaluate the negotiation pressure; then, according to the evaluated pressure and the incomplete information obtained, they determine the optimal strategy for the current round; finally, they choose the relatively optimal strategy that can produce greater expected utility; The negotiation process of the manufacturing service continues to cycle and update the model until a successful negotiation is reached, and then the negotiation result is added to the final negotiation result list. When the negotiation of all subtasks is completed, the negotiation process ends.

Images