Resource task collaborative decision-making method in complex manufacturing network
By adopting resource task collaborative decision-making methods in intelligent manufacturing systems, using game models and quasi-Newtonian optimization algorithms, the resource allocation of mechanical processing equipment is optimized, and the problem of customer personalized needs and low equipment utilization is achieved, and efficient resource utilization is achieved and balance of interests.
Patent Information
- Application Number
- CN202411070647.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-19
AI Technical Summary
In intelligent manufacturing systems, the existing technology is difficult to effectively solve the problems of dynamic changes in customer personalized needs and low resource utilization of mechanical processing equipment, especially when order demand information changes and resource allocation is complex.
The resource task collaborative decision-making method in complex manufacturing networks is used to determine the available time period of the machining equipment, calculate the usage price and profit function of the machining equipment according to the weight allocation of the order process and the game start time, and use quasi-Newtonian second-order optimization The algorithm solves the optimization model, optimizes resource configuration to meet order needs and improves equipment utilization.
It realizes flexible processing of order needs in a dynamic environment, improves resource utilization of mechanical processing equipment, reduces computing complexity, enhances scalability, balances customer interests with manufacturers' interests, and promotes supply chain collaboration.
Smart Images

Figure CN118982199A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent manufacturing systems, and in particular to a resource task collaborative decision-making method in a complex manufacturing network. Background Art
[0002] The production and processing of complex products are constantly moving towards dynamic decision-making and autonomous control. The problems faced by manufacturing systems, such as complex and changing demands, high frequency of collaboration, and many disturbance factors, are becoming increasingly prominent, bringing huge challenges to production control systems in terms of large-scale resource allocation and adaptive collaboration.
[0003] Large-scale production systems include massive manufacturing machinery and processing equipment resources with different process capabilities, and also receive order demand information from all over the world. The production resource allocation aims to achieve efficient matching between order demand and machinery and processing equipment resources, dynamically optimize resource utilization while ensuring customer needs, and enhance the market competitiveness of the production system.
[0004] In terms of product manufacturing process, the variable delay and mixed integer linear programming model used in Chinese patent CN118114808A provides a process planning and resource scheduling integrated optimization method for batch tasks. When the mechanical processing equipment resources and task information are determined and unchanged, the above method can effectively achieve resource allocation with the shortest completion time as the goal. However, this method is difficult to use to solve the problem of online scheduling where customer personalized demand information is dynamically generated or disappears.
[0005] At the same time, Chinese patent CN117909043A uses deep reinforcement learning technology to realize the online scheduling problem in a cloud manufacturing environment. However, the training process of the deep reinforcement learning framework in this method requires high-performance computing resources. When the number of resources or orders is large, the solution process is complicated, and the dynamic performance of the model lacks explainability.
[0006] Moreover, Chinese patent CN118195557A uses a two-layer game model to solve the problem of joint design optimization of product family configurations in the manufacturing process of personalized customized products. This method is mainly used for the production management process of the same type of products with customizable modules, and is not easy to expand to the production resource allocation system with different types of products.
[0007] For an online system where order demand information arrives dynamically, each order will apply to its target manufacturing mechanical processing equipment resources for a certain length of available processing time based on its different requirements for production cost, time consumption, quality and other indicators; at the same time, the available time of each mechanical processing equipment resource will be occupied by several order processes. Figure 1As shown, when an order occupies the available processing time, it may cause short time fragments in the available time period of the machining equipment. These time fragments are too short to be used by other orders, resulting in reduced utilization of the machining equipment. Summary of the invention
[0008] The technical problem to be solved by the present invention is: in the resource optimization configuration of the intelligent manufacturing system, how to achieve both meeting the customer order requirements in the online resource configuration process and improving the resource utilization of mechanical processing equipment. In order to overcome the defects of the above-mentioned prior art, the present invention provides a resource task collaborative decision-making method in a complex manufacturing network.
[0009] The present invention provides a resource task collaborative decision-making method in a complex manufacturing network, comprising the following steps:
[0010] S1: determining multiple currently available time periods of mechanical processing equipment in a manufacturing network environment, and obtaining a set of intervals corresponding to the currently available time periods;
[0011] S2: Select all available time periods whose length is not less than the time taken by the current order process from the interval set, and obtain the start time period corresponding to the current order process in each selected available time period according to the time distance between the current moment and each selected available time period and the logistics and lead time of the current order process;
[0012] S3: according to the weight distribution of time, cost and quality of the current order, the ideal start time of the current order process corresponding to each selected available time period is obtained from the start time period corresponding to each selected available time period obtained in step S2;
[0013] S4: taking the game start time of the current order process as an independent variable, obtaining the use price of the mechanical processing equipment corresponding to each selected available time period by determining the use price of the cooperative game between the mechanical processing equipment and the current order process using a floating pricing strategy;
[0014] S5: Taking the game start time of the current order process as an independent variable, based on the comparison relationship between the independent variable and the ideal start time of the current order process corresponding to the selected available time period obtained in step S3, the use price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4, and the sales price of the product processed by the mechanical processing equipment, respectively obtain the profit function of the current order process corresponding to each selected available time period;
[0015] S6: based on the use price of the machining equipment corresponding to each selected available time period obtained in step S4 and the opportunity loss of machining tasks caused by time period fragmentation generated by the machining equipment, respectively obtain the benefit function of the machining equipment corresponding to each selected available time period;
[0016] S7: The sum of the revenue function of the current order process corresponding to each selected available time period and the revenue function of the machining equipment corresponding to the selected available time period is taken as the overall revenue corresponding to each selected available time period, and the objective function corresponding to each selected available time period is obtained by taking the overall revenue corresponding to each selected available time period, and the starting time period corresponding to each selected available time period of the current order process obtained in step S2 is taken as the constraint condition corresponding to the selected available time period, and the optimization model corresponding to each selected available time period is obtained;
[0017] S8: Solve the optimization model corresponding to each selected available time period respectively by using the quasi-Newton second-order optimization algorithm to obtain the start time point of the order process corresponding to each selected available time period at the machining equipment.
[0018] The resource task collaborative decision-making method in a complex manufacturing network disclosed by the present invention takes the order process and the mechanical processing equipment as participants in the game process. The order process focuses on its current demand, with the goal of meeting the expected start time of its application; the mechanical processing equipment focuses on its long-term utilization, with the goal of avoiding the generation of short-length time fragments. In the process of the game, after obtaining the optimization model, this method solves the online resource allocation problem through the quasi-Newton second-order optimization algorithm, and can flexibly handle order demand information in a dynamic environment. The model avoids pre-training or complex parameter configuration process, has clear implementation logic, low computational complexity, and strong scalability. In addition, through this method, it is achieved to meet the order demand to the greatest extent while ensuring the resource utilization of the equipment. According to the relative priority of the order and the relative importance of the equipment, the present invention effectively balances the interests of customers and manufacturers, which is conducive to the healthy development of supply chain collaboration.
[0019] In a possible implementation manner, step S2 includes the following steps:
[0020] S21: Selecting all available time periods whose lengths are not less than the time consumption of the current order process from the interval set;
[0021] S22: obtaining the time distance between the current time and each selected available time period by subtracting the current time from the starting time of each available time period selected in step S21;
[0022] S23: Compare the time distance between the current time obtained in step S22 and each selected available time period with the logistics and lead time of the current order process:
[0023] If the time distance between the current time and a selected available time period is not less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows:
[0024] The start time is the start time of the selected available time period;
[0025] The end time is the minimum value between the end time of the selected available time period minus the time consumption of the current order process and the latest start time of the current order process;
[0026] If the time distance between the current time and a selected available time period is less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows:
[0027] The starting time is the end time of the previous order process of the current order process plus the logistics and pre-preparation time of the current order process;
[0028] The end time is the minimum value between the end time of the selected available time period minus the time consumption of the current order process and the latest start time of the current order process;
[0029] This technical solution ultimately determines the optional time for the order process on the mechanical processing equipment based on the fact that preparatory work such as material handling and tooling adjustment is required before the order process starts, thereby ensuring the accurate determination of the optional time.
[0030] In a possible implementation manner, step S4 includes the following steps:
[0031] S41: Obtaining a general process duration based on an average duration of a plurality of processes previously processed by the machining equipment;
[0032] S42: taking the minimum value between the general process duration and the process duration of the current order as the process duration;
[0033] S43: subtracting the starting time of each selected available time period from the game start time of the current order process to obtain the time difference corresponding to each selected available time period;
[0034] S44: Determine whether the time difference corresponding to each selected available time period is less than the process duration;
[0035] If the time difference corresponding to a selected available time period is not less than the process duration, the use price of the mechanical processing equipment corresponding to the selected available time period is set to the standard price per unit time of the mechanical processing equipment multiplied by the process duration of the current order;
[0036] If the time difference corresponding to a selected available time period is less than the process duration, the usage price of the mechanical processing equipment corresponding to the selected available time period is set to the sum of the current order process duration and the time difference corresponding to the selected available time period multiplied by the standard price per unit time of the mechanical processing equipment.
[0037] This technical solution avoids, to the greatest extent possible, dividing the available time period of the equipment into short time fragments, thereby improving the utilization rate of the mechanical processing equipment.
[0038] In a possible implementation manner, in step S5, the profit function formula of the current order process corresponding to any selected available time period is as follows:
[0039] P Order(p) =Value Order(p) -Cost Order(p) -Aux Order(p) ,
[0040]
[0041] In the formula,
[0042] P Order(p) Represents the revenue of the current order process corresponding to the selected available time period;
[0043] Value Order(p) Represents the value of the products currently processed by the machining equipment;
[0044] Cost Order(p) The usage price of the mechanical processing equipment corresponding to the selected available time period;
[0045] Aux Order(p) Represents the additional storage or penalty costs incurred due to early or delayed production of products currently processed by the above-mentioned mechanical processing equipment;
[0046] t p Represents the game start time of the current order process;
[0047] T LS(p) Represents the latest start time of the current order process;
[0048] W p Represents the storage cost per unit time of the current order process;
[0049] T Start(p) Represents the ideal start time of the current order process corresponding to the selected available time period;
[0050] Pen p Represents the delay penalty rate of the current order process;
[0051] This scheme determines the profit function of the order process in the game process based on the comparison between the independent variable and the ideal start time of the current order process corresponding to the selected available time period, which can ensure the orderly construction and solution of the optimization model.
[0052] In a possible implementation manner, in step S6, the profit function formula corresponding to the machining equipment in any selected available time period is as follows:
[0053] P m =Cost Order(p) -Opploss Order(p) ,
[0054]
[0055] In the formula,
[0056] P m Represents the revenue of the machining equipment corresponding to the selected available time period;
[0057] Opploss Order(p) represents the quantified opportunity loss of machining tasks due to time fragmentation of the machining equipment;
[0058] Prob arr represents the probability that at least one new task arrives at the machining equipment within ΔSlot;
[0059] Prob RW represents the probability that the new task randomly walks and visits the machining equipment at least once within ΔSlot;
[0060] E(I' p ) represents the mathematical expectation of the priority of the new task;
[0061] I p (Order(p)) represents the priority of the current order process;
[0062] in,
[0063] ΔSlot=TN i -TL i -T Order(p) ,
[0064] In the formula,
[0065] TN i Represents the end time of the selected available time period;
[0066] TL i represents the initial time of the selected available time period;
[0067] T Order(p) Represents the time taken for the current order process;
[0068] This solution creatively provides a formula for the profit function of the machining equipment in each selected available time period through a statistical method, thereby ensuring that the profit of the machining equipment in each selected available time period is obtained efficiently.
[0069] In a possible implementation, step S6 includes the following steps:
[0070] S61: defining the process of the mechanical processing equipment receiving orders as a Poisson process, so as to obtain the probability that at least one new task arrives at the mechanical processing equipment within ΔSlot;
[0071] S62: Obtain the probability of the machining equipment being visited once in the manufacturing network based on the inter-node transition probability algorithm in the directed network, and obtain the probability of the new task randomly walking and visiting the machining equipment at least once within ΔSlot;
[0072] S63: Substitute the usage price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4 and the probabilities obtained in steps S61 and S62 into the profit function formula corresponding to the mechanical processing equipment in any selected available time period to obtain the profit function of the mechanical processing equipment in each selected available time period.
[0073] In a possible implementation, the quasi-Newton second-order optimization algorithm in step S8 is a quasi-Newton second-order optimization algorithm using a BFGS matrix; thereby, the efficiency of solving the model can be improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] Figure 1 A schematic diagram of available time periods and time period fragments of the mechanical processing equipment described in the background technology and embodiments of the present invention;
[0075] Figure 2 It is a flow chart of a resource task collaborative decision-making method in a complex manufacturing network disclosed in an embodiment of the present invention;
[0076] Figure 3 is a flow chart of step S2 disclosed in an embodiment of the present invention;
[0077] Figure 4is a flow chart of step S4 disclosed in an embodiment of the present invention;
[0078] Figure 5 This is a flow chart of step S6 disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0079] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.
[0080] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0081] See also Figure 1 to Figure 5 As shown, the embodiment of the present application discloses a resource task collaborative decision-making method in a complex manufacturing network, see Figure 2 As shown, the method comprises the following steps:
[0082] S1: Determine multiple currently available time periods of mechanical processing equipment in a manufacturing network environment, and obtain a set of intervals corresponding to the currently available time periods.
[0083] Machining equipment can be CNC machining centers, lathes, milling machines, planers, grinders, see Figure 1 As shown, for machining equipment m, the interval set corresponding to its current available time period can be expressed as:
[0084] AvailSlot m ={(TL1,TN1),(TL2,TN2),…,(TL n ,TN n )}, where the left endpoint of each open interval is the completion time TL of the previous process, and the right endpoint is the planned start time TN of the next process.
[0085] S2: Select all available time periods whose length is not less than the time taken by the current order process from the interval set, and obtain the start time period corresponding to the current order process in each selected available time period based on the comparison results between the time distance between the current moment and each available time period and the logistics and lead preparation time of the current order process.
[0086] See also Figure 3 As shown, in this embodiment, step S2 includes the following steps:
[0087] S21: Select all available time periods whose length is not less than the duration of the current order process from the interval set.
[0088] S22: obtaining the time distance between the current time and each selected available time period by subtracting the current time from the start time of each available time period selected in step S21.
[0089] S23: Compare the time distance between the current time obtained in step S22 and each selected available time period with the logistics and lead time of the current order process:
[0090] If the time distance between the current time and a selected available time period is not less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows:
[0091] The start time is the start time of the selected available time period;
[0092] The end time is the minimum value between the result of subtracting the time consumption of the current order process from the end time of the selected available time period and the latest start time of the current order process.
[0093] If the time distance between the current time and a selected available time period is less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows:
[0094] The starting time is the end time of the previous order process of the current order process plus the logistics and pre-preparation time of the current order process;
[0095] The end time is the minimum value between the end time of the selected available time period minus the time consumption of the current order process and the latest start time of the current order process.
[0096] From a mathematical perspective, the current order process is in the selected available time period (TL i ,TN i ) The formula for the corresponding start time period is as follows:
[0097]
[0098] In the formula,
[0099] T LS(p) Represents the latest start time of the current order process;
[0100] T Finish(p-1) Represents the end time of the previous order process of the current order process;
[0101] T log(p-1,p) The logistics and lead time of the current order process;
[0102] T Order(p) Represents the time taken for the current order process;
[0103] ξI Represents the current time and the selected available time period (TL i ,TN i ) is less than the logistics and lead time of the current order process, that is
[0104] TL i -t <T log(p-1,p) ;
[0105] ξ S Represents the current time and the selected available time period (TL i ,TN i ) is not less than the logistics and lead time of the current order process, that is
[0106] TL i -t≥T log(p-1,p) .
[0107] S3: According to the weight distribution of time, cost and quality of the current order, the ideal start time of the current order process corresponding to each selected available time period is obtained from the start time period corresponding to each selected available time period obtained in step S2.
[0108] S4: Taking the game start time of the current order process as the independent variable, the use price of the mechanical processing equipment corresponding to each selected available time period is obtained by determining the use price of the cooperative game between the mechanical processing equipment and the current order process through a floating pricing strategy.
[0109] See also Figure 4 As shown, in this embodiment, step S4 includes the following steps:
[0110] S41: The average duration of multiple processes previously processed by the machining equipment is taken as the general process duration.
[0111] S42: The minimum value between the general process duration and the current order process duration is taken as the process duration.
[0112] S43: The time difference corresponding to each selected available time period is obtained by subtracting the starting time of each selected available time period from the game start time of the current order process.
[0113] S44: Determine whether the time difference corresponding to each selected available time period is less than the process duration;
[0114] If the time difference corresponding to a selected available time period is not less than the duration of the process, the use price of the mechanical processing equipment corresponding to the selected available time period is set to the standard price per unit time of the mechanical processing equipment multiplied by the duration of the current order process;
[0115] If the time difference corresponding to a selected available time period is less than the process duration, the usage price of the mechanical processing equipment corresponding to the selected available time period is set to the sum of the current order process duration and the time difference corresponding to the selected available time period multiplied by the standard price per unit time of the mechanical processing equipment.
[0116] From the perspective of mathematical expression, the selected available time period (TL i ,TN i ) The formula for the use price of the corresponding mechanical processing equipment is as follows:
[0117]
[0118] Δt=t p -TL i ,
[0119] ProcLen=min(T Order(p) ,genProcLen),
[0120] In the formula,
[0121] Cost Order(p) Represents the selected available time period (TL i ,TN i ) The price of the corresponding mechanical processing equipment;
[0122] StdCost represents the standard price per unit time of use of machining equipment;
[0123] t p Represents the game start time of the current order process, which is also an independent variable;
[0124] genProcLen represents the general process length;
[0125] ProcLen represents the process length.
[0126] S5: Taking the game start time of the current order process as the independent variable, based on the comparative relationship between the ideal start time of the current order process corresponding to the selected available time period obtained in step S3 and the independent variable, the usage price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4 and the sales price of the products processed by the mechanical processing equipment, the profit function of the current order process corresponding to each selected available time period is obtained respectively.
[0127] From the perspective of mathematical expression, the selected available time period (TL i ,TN i ) The profit function formula of the current order process corresponding to is as follows:
[0128] P Order(p) =Value Order(p) -CostOrde(p) -Aux Order(p) ,
[0129]
[0130] In the formula,
[0131] P Order(p) Represents the revenue of the current order process corresponding to the selected available time period;
[0132] Value Order(p) Represents the value of the products processed by the current machining equipment;
[0133] Aux Order(p) Represents the additional storage or penalty costs incurred due to early or delayed production of products processed by current machining equipment;
[0134] T LS(p) Represents the latest start time of the current order process;
[0135] W p Represents the storage cost per unit time of the current order process;
[0136] T Start(p) represents the ideal start time of the current order process corresponding to the selected available time period determined in step S3;
[0137] Pen p Represents the delay penalty rate of the current order process.
[0138] S6: Based on the usage price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4 and the opportunity loss of processing tasks caused by time period fragmentation generated by the mechanical processing equipment, the profit function corresponding to each selected available time period is obtained respectively.
[0139] From a mathematical perspective, the machining equipment is i ,TN i ) The corresponding profit function formula is as follows:
[0140] P m =Cost Order(p) -Opploss Order(p) ,
[0141]
[0142] In the formula,
[0143] P m Represents the revenue of the machining equipment in the selected available time period;
[0144] Opploss Order(p) Represents the quantified opportunity loss of machining tasks due to time fragmentation of machining equipment;
[0145] Prob arr represents the probability that at least one new task arrives at the machining equipment within ΔSlot;
[0146] Prob RW represents the probability that a new task randomly walks and visits a machining device at least once within ΔSlot;
[0147] E(I' p ) represents the mathematical expectation of the priority of the new task;
[0148] I p (Order(p)) represents the priority of the current order process;
[0149] in,
[0150] ΔSlot=TN i -TL i -T Order(p) .
[0151] In order to calculate specifically, the machining equipment in the selected available time period (TL i ,TN i ) corresponding to the income, see Figure 5 As shown, the adopted step S6 includes the following steps:
[0152] S61: The process of receiving orders for machining equipment is defined as a Poisson process, so as to obtain the probability that at least one new task arrives at the machining equipment within ΔSlot:
[0153]
[0154] In the formula,
[0155] λ a Represents the average number of order arrivals per unit time of the machining equipment, which can be calculated by the arrival time of N processes recently processed by the machining equipment, where N is the number of processes previously processed by the machining equipment extracted in step S41.
[0156] S62: Based on the inter-node transfer probability algorithm in the directed network, the probability of the mechanical processing equipment being visited once in the manufacturing network is obtained, and the probability of the new task randomly walking and visiting the mechanical processing equipment at least once within ΔSlot is obtained accordingly.
[0157] The mathematical expression of this step is as follows:
[0158] The probability of a machining device m being visited once is the sum of the probabilities of jumping to the machining device m after all machining device nodes in the manufacturing network are visited, that is:
[0159]
[0160] In the formula,
[0161] SNR m Represents the set of all machining equipment points in the manufacturing network where the machining equipment m is located;
[0162] I m (a) represents the influence or importance of machining equipment a in the manufacturing network where machining equipment m is located, that is, the PageRank value of node a.
[0163] Furthermore, the probability that the new task randomly walks and visits the machining equipment at least once within ΔSlot is:
[0164]
[0165] In the formula,
[0166] ceil() is a round-up function. is the number of processes that can be performed within the ΔSlot time period, and is also the number of jumps that occur in the random walk.
[0167] S63: Substitute the usage price of the machining equipment corresponding to each selected available time period obtained in step S4 and the probability obtained in steps S61 and S62 into the usage price of the machining equipment in the selected available time period (TL i ,TN i ) to obtain the corresponding profit function of the machining equipment in each selected available time period.
[0168] S7: The sum of the profit function of the current order process corresponding to each selected available time period and the profit function of the machining equipment corresponding to the selected available time period is taken as the overall profit corresponding to each selected available time period, and the objective function corresponding to each selected available time period is obtained by using the overall profit corresponding to each selected available time period, and the start time period corresponding to each selected available time period of the current order process obtained in step S2 is taken as the constraint condition corresponding to the selected available time period to obtain the optimization model corresponding to each selected available time period.
[0169] Considering the production value for a specific machining equipment Order(p) It is a constant in the game process and does not affect the solution result, so the selected available time period (TL i ,TN i) The corresponding objective function can be expressed as follows:
[0170]
[0171] Thus, the selected available time period (TL i ,TN i ) The corresponding optimization model is as follows:
[0172] minimize y=-Payoff(t p ),
[0173] stt p ∈ActSlot m .
[0174] S8: The optimization model corresponding to each selected available time period is solved respectively by a quasi-Newton second-order optimization algorithm to obtain the start time point of the order process corresponding to each selected available time period on the machining equipment. The quasi-Newton second-order optimization algorithm is a quasi-Newton second-order optimization algorithm using a BFGS matrix.
[0175] The resource task collaborative decision-making method in a complex manufacturing network disclosed in the embodiment of the present application adopts a non-zero-sum cooperative game model, taking the order process and mechanical manufacturing equipment resources as participants in the game process. The order process focuses on its current needs, with the goal of meeting the expected start time of its application; the mechanical manufacturing equipment resources focus on their long-term utilization, with the goal of avoiding the generation of shorter time period fragments. In the process of the game, the technical solution of this embodiment comprehensively considers the priority of the order process and the importance of equipment resources, and uses methods such as cost-benefit calculation, penalty function, network probability model, and income expectation model to achieve a dynamic balance between customer satisfaction and resource efficiency.
[0176] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0177] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
Claims
1. A resource task collaborative decision-making method in a complex manufacturing network, characterized in that: The steps include: S1: determining multiple currently available time periods of mechanical processing equipment in a manufacturing network environment, and obtaining a set of intervals corresponding to the currently available time periods; S2: Select all available time periods whose length is not less than the time taken by the current order process from the interval set, and obtain the start time period corresponding to the current order process in each selected available time period according to the time distance between the current moment and each selected available time period and the logistics and lead time of the current order process; S3: according to the weight distribution of time, cost and quality of the current order, the ideal start time of the current order process corresponding to each selected available time period is obtained from the start time period corresponding to each selected available time period obtained in step S2; S4: taking the game start time of the current order process as an independent variable, obtaining the use price of the mechanical processing equipment corresponding to each selected available time period by determining the use price of the cooperative game between the mechanical processing equipment and the current order process using a floating pricing strategy; S5: Taking the game start time of the current order process as an independent variable, based on the comparison relationship between the ideal start time of the current order process corresponding to the selected available time period obtained in step S3 and the independent variable, the use price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4, and the sales price of the product processed by the mechanical processing equipment, the profit function of the current order process corresponding to each selected available time period is obtained; S6: based on the use price of the machining equipment corresponding to each selected available time period obtained in step S4 and the opportunity loss of machining tasks caused by time period fragmentation generated by the machining equipment, respectively obtain the benefit function of the machining equipment corresponding to each selected available time period; S7: The sum of the revenue function of the current order process corresponding to each selected available time period and the revenue function of the machining equipment corresponding to the selected available time period is taken as the overall revenue corresponding to each selected available time period, and the objective function corresponding to each selected available time period is obtained by taking the overall revenue corresponding to each selected available time period, and the starting time period corresponding to each selected available time period of the current order process obtained in step S2 is taken as the constraint condition corresponding to the selected available time period, and the optimization model corresponding to each selected available time period is obtained; S8: Solve the optimization model corresponding to each selected available time period respectively by using the quasi-Newton second-order optimization algorithm to obtain the start time point of the order process corresponding to each selected available time period at the machining equipment.
2. The resource task collaborative decision-making method in a complex manufacturing network according to claim 1 is characterized in that: The step S2 comprises the following steps: S21: Selecting all available time periods whose lengths are not less than the time consumption of the current order process from the interval set; S22: obtaining the time distance between the current time and each selected available time period by subtracting the current time from the starting time of each available time period selected in step S21; S23: Compare the time distance between the current time obtained in step S22 and each selected available time period with the logistics and lead time of the current order process: If the time distance between the current time and a selected available time period is not less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows: The start time is the start time of the selected available time period; The end time is the minimum value between the end time of the selected available time period minus the time consumption of the current order process and the latest start time of the current order process; If the time distance between the current time and a selected available time period is less than the logistics and lead time of the current order process, the start time period corresponding to the selected available time period is set as follows: The starting time is the end time of the previous order process of the current order process plus the logistics and pre-preparation time of the current order process; The end time is the minimum value between the result of subtracting the time consumption of the current order process from the end time of the selected available time period and the latest start time of the current order process.
3. The resource task collaborative decision-making method in a complex manufacturing network according to claim 2 is characterized in that: The step S4 comprises the following steps: S41: Obtaining a general process duration based on an average duration of a plurality of processes previously processed by the machining equipment; S42: taking the minimum value between the general process duration and the process duration of the current order as the process duration; S43: subtracting the starting time of each selected available time period from the game start time of the current order process to obtain the time difference corresponding to each selected available time period; S44: Determine whether the time difference corresponding to each selected available time period is less than the process duration; If the time difference corresponding to a selected available time period is not less than the process duration, the use price of the mechanical processing equipment corresponding to the selected available time period is set to the standard price per unit time of the mechanical processing equipment multiplied by the process duration of the current order; If the time difference corresponding to a selected available time period is less than the process duration, the usage price of the mechanical processing equipment corresponding to the selected available time period is set to the sum of the current order process duration and the time difference corresponding to the selected available time period multiplied by the standard price per unit time of the mechanical processing equipment.
4. The resource task collaborative decision-making method in a complex manufacturing network according to claim 3 is characterized in that: In step S5, the profit function formula of the current order process corresponding to any selected available time period is as follows: P Order(p) =Value Order(p) -Cost Order(p) -To Order(p) , In the formula, P Order(p) Represents the revenue of the current order process corresponding to the selected available time period; Value Order(p) Represents the value of the products currently processed by the machining equipment; Cost Order(p) The usage price of the mechanical processing equipment corresponding to the selected available time period; Aux Order(p) Represents the additional storage or penalty costs incurred due to early or delayed production of products currently processed by the above-mentioned mechanical processing equipment; t p Represents the game start time of the current order process; T LS(p) Represents the latest start time of the current order process; W p Represents the storage cost per unit time of the current order process; T Start(p) Represents the ideal start time of the current order process corresponding to the selected available time period; Pen p Represents the delay penalty rate of the current order process.
5. The resource task collaborative decision-making method in a complex manufacturing network according to claim 4 is characterized in that: In step S6, the profit function formula corresponding to the machining equipment in any selected available time period is as follows: P m =Cost Order(p) -Opploss Order(p) , In the formula, P m Represents the revenue of the machining equipment corresponding to the selected available time period; Opploss Order(p) represents the quantified opportunity loss of machining tasks due to time fragmentation of the machining equipment; Prob arr represents the probability that at least one new task arrives at the machining equipment within ΔSlot; Prob RW represents the probability that the new task randomly walks and visits the machining equipment at least once within ΔSlot; E(I' p ) represents the mathematical expectation of the priority of the new task; I p (Order(p)) represents the priority of the current order process; in, ΔSlot=TN i -TL i -T Order(p) , In the formula, TN i Represents the end time of the selected available time period; TL i represents the initial time of the selected available time period; T Order(p) Indicates the time taken for the current order process.
6. The resource task collaborative decision-making method in a complex manufacturing network according to claim 5, characterized in that: The step S6 comprises the following steps: S61: defining the process of the mechanical processing equipment receiving orders as a Poisson process, so as to obtain the probability that at least one new task arrives at the mechanical processing equipment within ΔSlot; S62: Obtain the probability of the machining equipment being visited once in the manufacturing network based on the inter-node transition probability algorithm in the directed network, and obtain the probability of the new task randomly walking and visiting the machining equipment at least once within ΔSlot; S63: Substitute the usage price of the mechanical processing equipment corresponding to each selected available time period obtained in step S4 and the probabilities obtained in steps S61 and S62 into the profit function formula corresponding to the mechanical processing equipment in any selected available time period to obtain the profit function of the mechanical processing equipment in each selected available time period.
7. The resource task collaborative decision-making method in a complex manufacturing network according to claim 6, characterized in that: The quasi-Newton second-order optimization algorithm in step S8 is a quasi-Newton second-order optimization algorithm using a BFGS matrix.
