A petrochemical supply chain planning and scheduling scheme reliability analysis method and device

Abstract

The application discloses a petrochemical supply chain planning and scheduling scheme reliability analysis method and device, which quantitatively calculates the reliability of effective execution of a petrochemical supply chain planning and scheduling scheme by constructing a reliability analysis model of the supply chain and the planning and scheduling problem under the condition of considering uncertainty. In the first stage, the optimal scheme of the petrochemical supply chain planning and scheduling model is solved without considering uncertainty. In the second stage, the reliability measurement index of the scheme in response to various uncertain conditions is calculated according to the calculated index. If the index cannot meet the requirement, the reliability of the scheme execution can be improved by optimizing the nominal point of the uncertain parameter and the expected value of the performance index. The method can evaluate the reliability of the petrochemical supply chain planning and scheduling scheme in a quantitative analysis manner, optimize production decisions, and balance the optimal benefits of production decisions, the demand of energy saving and emission reduction and the reliability of scheme execution.

Description

Technical Field

[0001] This invention relates to the field of reliability analysis and optimization of supply chain and production planning, and in particular, to a method and apparatus for reliability analysis of petrochemical supply chain planning and scheduling schemes. Background Technology

[0002] In large-scale industrial construction, especially in the petrochemical industry, profits no longer solely rely on improvements in production processes. Scientific and effective production planning, scheduling, production allocation, and supply chain management often create greater profit margins through cost reduction. With increasing global demands for energy conservation and emission reduction, petrochemical companies can significantly reduce energy consumption and emissions during transportation by optimizing logistics routes and improving the energy efficiency of transportation vehicles. In the production process, adopting energy-efficient equipment and optimizing production processes not only helps reduce costs but also enhances the company's social image and competitiveness. Therefore, current corporate competition has gradually evolved into competition over supply chains. In the petrochemical industry, efficient supply chain management directly impacts cost control and efficiency improvement in key areas such as raw material procurement, production scheduling, and product distribution. With intensifying market competition, rising costs, and the accelerating globalization of manufacturing, supply, and distribution, the importance of supply chain management is increasingly prominent. This background has spurred the development of supply chain operations, design, and systems analysis, and the supply chain is increasingly viewed as an integrated system composed of multiple components and decision-making levels.

[0003] The reliability of supply chain planning and scheduling schemes is fundamental to a company's stable profitability and determines the smooth operation of the entire supply chain system. Supply chain system reliability not only affects logistics and market pricing but also profoundly impacts a company's macro-level decision-making; while the feasibility of the planning and scheduling scheme directly determines the effectiveness of production and processing. Whether in process or discrete manufacturing industries, the reliability of supply chain planning and scheduling schemes plays a crucial role in the entire systems engineering. For example, the application of energy-efficient production equipment and processes can effectively reduce energy consumption and pollutant emissions generated during energy production, which indirectly improves the stability and reliability of the supply chain, giving companies a greater advantage in responding to energy conservation and emission reduction policies.

[0004] Numerous variables influence the reliability of supply chain systems, including fluctuations in supply and demand, cost changes, production capacity adjustments, unstable energy supply, and unforeseen events that could disrupt the supply chain. This is particularly true in the petrochemical industry, where uncertainties in raw material procurement and energy supply are often key factors affecting supply chain stability. Under the backdrop of energy conservation and emission reduction, energy price fluctuations and production process adjustments due to environmental policies have also become significant influencing factors. Uncertainties in planning and scheduling are primarily reflected in the uncertainty of production costs and profits. Changes in equipment performance, material properties, chemical reaction mechanisms, and the unique fluctuations in raw material quality specific to the petrochemical industry also affect the feasibility of the plans. Furthermore, the development and application of energy-saving technologies also impact planning and scheduling; for example, new energy-saving materials may alter existing production rhythms and the connection between upstream and downstream supply chains. A stable supply chain is crucial for the stable production of manufacturers in the petrochemical industry, the sustained profitability of enterprises, the long-term stable development of the market, and the satisfaction of consumer demand. Supply chain stability is especially critical in handling raw material procurement, refining, and production. Moreover, a stable supply chain that emphasizes energy conservation and emission reduction helps promote the entire industry towards a green and sustainable development direction, aligning with global development trends.

[0005] To achieve precise cost control across the entire supply chain, this invention aims to systematically analyze each link of the supply chain, covering modeling and uncertainty analysis of the petrochemical industry's distribution, planning and scheduling, and processing and production links. By analyzing uncertain variables, the reliability indicators of the supply chain are evaluated, thereby quantifying the reliability of supply chain planning and scheduling schemes. Simultaneously, factors related to energy conservation and emission reduction are incorporated into the analysis system, enabling enterprises to fully consider the opportunities and challenges brought by energy conservation and emission reduction when formulating supply chain strategies, further enhancing the overall efficiency and competitiveness of the supply chain. Summary of the Invention

[0006] The purpose of this invention is to provide an effective method for evaluating petrochemical supply chain planning and scheduling schemes that do not involve loops, by using a reliability index to quantify the reliability of the scheme through reliability analysis.

[0007] According to a first aspect of the present invention, a method for reliability analysis of petrochemical supply chain planning and scheduling schemes is provided, the method comprising the following steps:

[0008] S1, establish a mathematical model for supply chain planning and scheduling that increases revenue and reduces energy consumption in the petrochemical supply chain and planning and scheduling process;

[0009] S2, without considering uncertainties, solve the optimal supply chain planning and scheduling scheme through the optimization proposition of the mathematical model, and establish the expected values ​​of various performance indicators and corresponding concession constraints based on the optimal scheme;

[0010] S3: Set the fluctuation range of various uncertain parameters in the supply chain layer and the planning and scheduling layer, establish a reliability analysis model, and conduct a feasibility test to determine whether the optimal solution is completely reliable within the given fluctuation range of uncertain parameters. If it is not completely reliable, then execute S4.

[0011] S4, calculate the reliability index of the reliability analysis model and use it as a quantitative indicator of the reliability of the scheme.

[0012] S5 determines whether the reliability index meets the user's reliability requirements. If it does, the solution is reliable, and the reliability index is returned to the user. If it does not meet the requirements, there are two cases:

[0013] For uncertain parameters that can have a nominal point specified, use the derivative-free optimization method to find the optimal nominal point of the uncertain parameter, then mark the uncertain parameter as not being able to specify a nominal point, and return to S4;

[0014] For uncertain parameters that do not have a specified nominal point, identify the key constraints that play a limiting role, determine whether the key constraints can be relaxed, and return to S4 if they can be relaxed; otherwise, it indicates that the solution is unreliable.

[0015] Furthermore, the supply chain component of the mathematical model is a multi-level model comprising four stages: suppliers, factories, distributors, and customers. The supply chain model is constrained by the following related constraints:

[0016]

[0017]

[0018] Equation (1) represents the mass balance constraint of material s in factory m during period t, where, This represents the inventory of material s at factory m at the beginning of period t. F represents the inventory of material s at factory m at the end of period t; d,d′,s,t η represents the amount of material s transported from scenario d to scenario d' during cycle t; m,s Ba represents the conversion efficiency of material s in factory m between output and input. m,t Equation (2) represents the batch size of factory m in cycle t; Equation (2) represents the quality balance constraint of distributor v. This represents the quantity of material s for distributor v at the beginning of period t. This represents the quantity of material s held by distributor v at the end of period t.v,s,t Let represent the amount of material s delivered by distributor v to customer s in period t; M represents the set of all factories m, H represents the set of all suppliers h, V represents the set of all distributors v, S represents the set of all materials s, and T represents the set of all periods t.

[0019]

[0020] Equations (3) and (4) represent periodic inventory constraints, i.e., the coupling relationship between material quantity constraints of different cycles, factories, and distributors. Material may be lost due to storage or transportation during cycle changes, with a loss rate of α. s In equation (4) This represents the amount of material s in scenario d at the initial time.

[0021]

[0022] Equations (3)-(7) represent the upper and lower limits of transportation volume and inventory volume, respectively, where F h,m,s,t Sp represents the amount of material s transported by supplier h to factory m during period t; h,s,t This represents the maximum supply quantity of material s by supplier h during period t. This represents the upper limit of the inventory capacity for material s at node d.

[0023] Furthermore, the planning and scheduling part of the mathematical model includes constraints at the planning layer and constraints at the scheduling layer. The planning layer is responsible for formulating the overall production plan, while the scheduling layer focuses on specific resource allocation and execution details. The planning layer and the scheduling layer are coupled through shared variables and synchronization constraints to form a comprehensive planning and scheduling model.

[0024] The specific program constraints are as follows:

[0025]

[0026] in, This represents the inventory level of material s in period t. This represents the demand for material s during period t. This represents the stock shortage quantity of material s in period t. This represents the actual delivery quantity of material s in period t. This represents the planned production quantity of material s in period t. Let |N| represent the state of material s at operation node n in period t, where |N| represents the total number of operation nodes. S represents the state quantity of material s at the initial time of period t. P Represents a collection of products;

[0027] The scheduling constraints are specifically as follows:

[0028]

[0029] Equations (12) and (13) characterize the processing logic and capacity constraints of the equipment executing production tasks, where J represents the set of generating and processing equipment and I represents the set of tasks; It is a 0-1 binary variable that indicates whether device j executes task i at operator node n during period t. This represents the batch size of task i executed by device j at operation node n during period t. and I represents the lower and upper limits of batch processing for device j to execute task i, respectively. j J represents the set of tasks that device j can perform. i Let N represent the set of devices capable of performing task i, and let N represent the set of operation nodes; Equations (14) and (15) embody the material state transformation in the production process, where This represents the consumption rate of material s per unit batch size during the execution of task i. I represents the production rate of material s per unit batch size during the execution of task i. s This represents the set of tasks related to the processing of material s.

[0030] Furthermore, in S2, for the optimal supply chain planning and scheduling scheme, a comprehensive analysis of performance indicators such as total profit, total sales, production cost, transportation cost, inventory cost, stockout cost, maintenance cost, energy consumption, and pollutant emissions needs to be considered. When quantifying the reliability of the supply chain planning and scheduling scheme, it is necessary to set the expected profit value, the expected cost budget value, the maximum energy consumption value, and the maximum pollutant emissions value, and then construct the corresponding performance indicator concession constraints.

[0031] Furthermore, the uncertain parameters in the supply chain layer include uncertain parameters on the supply side, uncertain parameters on the demand side, and transportation costs; the uncertain parameters in the planning and scheduling layer include batch production conversion rate, pollutant emission rate, and batch production cost.

[0032] Furthermore, the concession constraints, uncertain parameters and their fluctuation range established in S2 are introduced into the supply chain planning and scheduling mathematical model in S1 to construct a reliability analysis model, and then the feasibility of the uncertain parameters is tested.

[0033] The feasibility test is used to verify whether, within a given space of uncertain parameters, the optimal solution has at least one set of control variables z for any set of uncertain parameters θ, such that all constraints f... g Both (z, θ) are feasible, and the mathematical expression is as follows:

[0034]

[0035] Where G represents the set of mathematical model constraints in S1 and concession constraints established based on the optimal solution in S2;

[0036] For a given uncertain parameter θ, the fluctuation range θ N -Δθ - ≤θ≤θ N +Δθ + , where θ N For the nominal point of the uncertain parameter θ, Δθ - ,Δθ + Let θ be the minimum and maximum fluctuation values ​​of the uncertain parameter θ, respectively. If the uncertain parameter θ satisfies equation (16) within the fluctuation range, the feasibility test of the optimal solution is passed, indicating that the optimal solution is completely reliable within the given fluctuation range of the uncertain parameter. If the feasibility test is not passed, then S4 is executed.

[0037] Furthermore, the reliability index is a quantitative indicator used to measure the ratio between a fully feasible space of uncertain parameters and the expected range of uncertain parameter fluctuations, and its mathematical expression is as follows:

[0038] F = maxδ

[0039] stmax θ∈T(δ) min z max g∈G f g (z,θ)≤0

[0040] T(δ) = {θ|θ N -δΔθ - ≤θ≤θ N +δΔθ +}

[0041] Where δ is the scaling factor of the uncertain parameter space; T(δ) represents the scaled uncertain parameter space;

[0042] The solution to the above problem is transformed into the following equivalent convex optimization form:

[0043] F = min k∈K δ k

[0044] stδ k =max δ,z δ

[0045] stf g (z,θ)≤0,g∈G

[0046] θ=θ N +δΔθk

[0047] Where k is a vertex in the uncertain parameter space, K is the set of vertices in the uncertain parameter space, and δ k Δθ represents the scaling factor corresponding to vertex k. k f represents the maximum fluctuation value of the uncertain parameter corresponding to vertex k; g (z,θ) represents the inherent constraints generated during the construction of the mathematical model in S1, and the concession constraints based on performance indicators in S2; θ is a given uncertain parameter; z is a control variable.

[0048] Furthermore, when there are uncertain parameters in S5 that can be specified at a nominal point, a derivative-free optimization method is used to find the nominal point θ of the uncertain parameter that maximizes the reliability index. N The specific optimization propositions are as follows:

[0049]

[0050] T(δ) = {θ|θ N -δΔθ - ≤θ≤θ N +δΔθ +}

[0051] z L ≤z≤z U

[0052] Where, θ U and θ L Representing the nominal point θ of the uncertain parameter respectively N The upper and lower bounds can vary; if some uncertain parameters cannot be specified at a nominal point, then their upper and lower bounds are equal to the nominal point, i.e., θ. L =θ N =θ U ;z U and z L The upper and lower bounds of the control variable z are represented; the optimal nominal point of the uncertain parameter is determined by the derivative-free optimization method, while maximizing the reliability index. The hyperrectangle calculated based on the optimal nominal point is the largest space of uncertain parameters.

[0053] Furthermore, if there are no uncertain parameters in S5 that can specify a nominal point, then an active constraint strategy needs to be applied to identify the active constraints, that is, to restrict one or more key constraints on the reliability index of the scheme within the scaled uncertain parameter space; based on the performance index corresponding to the returned constraints, the decision-maker dynamically adjusts the value of the performance index to achieve a balance between performance index and reliability requirements, thereby improving the reliability of the scheme.

[0054] According to a second aspect of the present invention, a petrochemical supply chain planning and scheduling scheme reliability analysis apparatus is provided. The apparatus includes a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it is used to implement the above-described petrochemical supply chain planning and scheduling scheme reliability analysis method.

[0055] The beneficial effects of this invention are:

[0056] 1. The reliability analysis method for petrochemical supply chain planning and scheduling schemes proposed in this invention can handle the reliability calculation problem of petrochemical supply chain planning and scheduling scheme design parameters under multidimensional uncertain parameter conditions, and provide decision-making assistance suggestions for decision-makers.

[0057] 2. The nominal point tuning method in this invention can find a better nominal point for uncertain parameters that can be manually intervened. This has a good guiding role in decision-making when there are problems such as parameter tuning and pricing under uncertain disturbances.

[0058] 3. This invention addresses the reliability space corresponding to uncertain parameters by finding activation constraints and relaxation quantities. This helps decision-makers improve the reliability of supply chain planning and scheduling schemes by adjusting performance index values, while taking into account both revenue and energy conservation and emission reduction requirements. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 A flowchart illustrating the reliability analysis method for supply chain planning and scheduling schemes provided in this embodiment of the invention;

[0061] Figure 2 This is a schematic diagram of a planned scheduling and processing model for a petrochemical process industry, provided as an embodiment of the present invention.

[0062] Figure 3 This is a schematic diagram of the reliability analysis device provided in an embodiment of the present invention. Detailed Implementation

[0063] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0064] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0065] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0066] like Figure 1 As shown in the figure, this invention provides a method for reliability analysis of petrochemical supply chain planning and scheduling schemes, which includes the following steps:

[0067] S1, establish a mathematical model for supply chain planning and scheduling that increases revenue and reduces energy consumption in the petrochemical supply chain and planning and scheduling process;

[0068] S2, without considering uncertainty, solve the optimal supply chain planning and scheduling scheme through the optimization proposition of the mathematical model of supply chain planning and scheduling, and establish the expected values ​​of various performance indicators and corresponding concession constraints based on the optimal scheme;

[0069] S3: Set the fluctuation range of various uncertain parameters in the supply chain layer and the planning and scheduling layer, establish a reliability analysis model, and conduct a feasibility test to determine whether the optimal solution is completely reliable within the given fluctuation range of uncertain parameters. If it is not completely reliable, then execute S4.

[0070] S4, calculate the reliability index of the reliability analysis model and use it as a quantitative indicator of the reliability of the supply chain planning and scheduling scheme;

[0071] S5 determines whether the reliability index meets the user's reliability requirements. If it does, i.e., the reliability index is greater than or equal to the preset reliability index threshold, then the solution is reliable, and the reliability index is returned to the user. If it does not meet the requirements, there are two cases:

[0072] For uncertain parameters that can have a nominal point specified, use the derivative-free optimization method (DFO) to solve for the optimal nominal point of the uncertain parameter, then mark the uncertain parameter as not being able to have a nominal point specified, and return to S4;

[0073] For uncertain parameters that do not have a specified nominal point, identify the key constraints that play a limiting role, determine whether the key constraints can be relaxed, and return to S4 if they can be relaxed; otherwise, it indicates that the solution is unreliable.

[0074] Furthermore, in step S1, the supply chain component of the mathematical model is a multi-level model comprising four stages: suppliers, factories, distributors, and customers. Suppliers provide the raw materials for petrochemical product production, the products are processed through factories at various levels, and the products are distributed to customers through distributors at various levels. The supply chain model requires the following related constraints:

[0075]

[0076] Equation (1) represents the mass balance constraint of material s in factory m during period t, where, This represents the inventory of material s at factory m at the beginning of period t. F represents the inventory of material s at factory m at the end of period t; d,d′,s,t This represents the amount of material s transported from scenario d to scenario d' in cycle t. Here, d and d' represent scenarios, which, based on their respective sets, can refer to factory m, distributor v, and supplier h; η m,s Ba represents the conversion efficiency of material s in factory m between output and input. m,t This indicates the batch size of factory m in cycle t; Equation (2) represents the quality balance constraint of distributor v, where there is no material conversion process within distributor v. This represents the quantity of material s for distributor v at the beginning of period t. This represents the quantity of material s in distributor v at the end of period t. v,s,t Let represent the amount of material s delivered by distributor v to customer s in period t; M represents the set of all factories m, H represents the set of all suppliers h, V represents the set of all distributors v, S represents the set of all materials s, and T represents the set of all periods t.

[0077]

[0078] Equations (3) and (4) represent periodic inventory constraints, i.e., the coupling relationship between material quantity constraints of different cycles, factories, and distributors. Among them, materials may suffer certain material losses due to storage or transportation during cycle changes, with a loss rate of α. s In equation (4) This represents the amount of material s in scenario d at the initial time.

[0079]

[0080] Equations (3)-(7) represent the upper and lower limits of the transport volume F and the inventory volume Inv, respectively, where F h,m,s,t Sp represents the amount of material s transported by supplier h to factory m during period t; h,s,t This represents the maximum supply quantity of material s by supplier h during period t. This represents the upper limit of the inventory capacity of material s at node d; in addition, constraints such as the upper and lower limits of delivery quantity can be used as additional constraints in the supply chain model, specifically including constraints such as customer order fulfillment rate and delivery time window.

[0081] Furthermore, in step S1, the planning and scheduling part of the mathematical model includes constraints at the planning layer and constraints at the scheduling layer. The planning layer is responsible for formulating the overall production plan, while the scheduling layer focuses on specific resource allocation and execution details. The planning layer and the scheduling layer establish a coupling relationship through shared variables and synchronization constraints to form a comprehensive planning and scheduling model. This comprehensive model can adopt a planning and scheduling model based on State Task Network (STN) or Resource Task Network (RTN), which can effectively describe and optimize resource scheduling and production task allocation in the petrochemical supply chain to meet various production and energy efficiency requirements.

[0082] ① The planning constraints need to include equations (8) to (11):

[0083]

[0084] in, This represents the inventory level of material s in period t. This represents the demand for material s during period t. This represents the stock shortage quantity of material s in period t. This represents the actual delivery quantity of material s in period t. This represents the planned production quantity of material s in period t. Let |N| represent the state quantity of material s at operation node n in period t (i.e., the quantity of inventory or work-in-process at that node), and |N| represent the total number of operation nodes. S represents the state quantity of material s at the initial time of period t. P S represents a set of products. P All materials listed are products that can be sold.

[0085] ② Scheduling constraints are mainly logical constraints and production constraints, specifically:

[0086]

[0087]

[0088] Equations (12) and (13) characterize the processing logic and capacity constraints of the equipment executing production tasks, where J represents the set of generating and processing equipment and I represents the set of tasks; It is a 0-1 binary variable that indicates whether device j executes task i at operator node n during period t. This represents the batch size of task i executed by device j at operation node n during period t. and I represents the lower and upper limits of batch processing for device j to execute task i, respectively. j J represents the set of tasks that device j can perform. i Let N represent the set of devices capable of performing task i, and let N represent the set of operation nodes. Equations (14) and (15) embody the material state transitions in the production process, where... This represents the consumption rate of material s per unit batch size during the execution of task i. I represents the production rate of material s per unit batch size during the execution of task i. s This represents the set of tasks related to the processing of material s. These scheduling constraints are determined by factors such as the specific processing tasks, equipment availability, and the order of operations in the production environment.

[0089] In addition, the planning and scheduling model needs to construct an objective function that integrates the needs for energy conservation and emission reduction with the needs for economic benefits. Depending on the specific circumstances, the planning and scheduling model can construct corresponding energy consumption constraints and pollutant emission constraints.

[0090] Furthermore, step S2 specifically includes:

[0091] To determine the optimal supply chain planning and scheduling scheme, a comprehensive analysis of various performance indicators is required, including total profit, total sales, production costs, transportation costs, inventory costs, stockout costs, maintenance costs, energy consumption, and pollutant emissions. When considering uncertain parameter interference, the optimal solution may not be achievable, as parameter fluctuations can cause the plan to deviate from its optimal state. Quantifying the reliability of the supply chain planning and scheduling scheme requires setting expected profit P*, expected cost budget C*, maximum energy consumption Q*, and maximum pollutant emissions S*. After setting these expected and maximum values, concession constraints for the corresponding performance indicators are constructed. These concession constraints are a necessary component of the reliability quantification calculation for the supply chain planning and scheduling scheme.

[0092] Furthermore, step S3 specifically includes:

[0093] Uncertain parameters in the supply chain layer include those on the supply side (such as supply quantity and raw material costs), the demand side (such as demand quantity and sales price), and transportation costs. Uncertain parameters in the planning and scheduling layer include batch production conversion rate, pollutant emission rate, and batch production cost.

[0094] By introducing the concession constraints, uncertain parameters, and their fluctuation range established in S2 into the supply chain planning and scheduling mathematical model in S1, a reliability analysis model is constructed.

[0095] When discussing the impact of uncertain parameters, it is first necessary to conduct a feasibility test on the uncertain parameters to ensure that all constraints can be met under different parameter conditions.

[0096] Feasibility testing is a branch of reliability analysis methods used to verify whether, within a given space of uncertain parameters, the optimal solution has at least one set of control variables z such that all constraints f... g Both (z, θ) are feasible, and the mathematical expression is as follows:

[0097]

[0098] Where G represents the set of mathematical model constraints in S1 and concession constraints established based on the optimal solution in S2.

[0099] For a given uncertain parameter θ, the fluctuation range θ N -Δθ - ≤θ≤θ N +Δθ = , where θ N For the nominal point of the uncertain parameter θ, Δθ - ,Δθ + Let be the minimum and maximum fluctuation values ​​of the uncertain parameter θ, respectively. If the uncertain parameter θ satisfies equation (16) within the fluctuation range, the feasibility test of the optimal solution is passed, indicating that the optimal solution is completely reliable within the given fluctuation range of the uncertain parameter. If the feasibility test is not passed, then S4 is executed.

[0100] Furthermore, step S4 specifically includes:

[0101] The reliability index calculation method is an extension based on feasibility testing, further quantifying the reliability of the uncertain parameter space. The reliability index F is a quantitative indicator used to measure the ratio between a fully feasible uncertain parameter space and the expected range of uncertain parameter fluctuations, reflecting the supply chain planning and scheduling scheme's ability to withstand uncertainty. Its mathematical expression is as follows:

[0102] F = maxδ

[0103] stmax θ∈T(δ) min z max g∈G f g (z,θ)≤0

[0104] T(δ) = {θ|θ N -δΔθ - ≤θ≤θ N +δΔθ +}

[0105] Where δ is the scaling factor of the uncertain parameter space; T(δ) represents the scaled uncertain parameter space.

[0106] The solution to the above problem can be transformed into the following equivalent convex optimization form, which provides the same solution under the same constraints:

[0107] F = min k∈K δ k

[0108] stδ k =max δ,z δ

[0109] stf g (z,θ)≤0,g∈G

[0110] θ=θ N +δΔθ k

[0111] Where k is a vertex in the uncertain parameter space, K is the set of vertices in the uncertain parameter space, and δ k Δθ represents the scaling factor corresponding to vertex k. k f represents the maximum fluctuation value of the uncertain parameter corresponding to vertex k; g (z,θ) represents the inherent constraints generated during the construction of the mathematical model in S1, and the additional concession constraints based on performance indicators added in S2; θ is a given uncertain parameter; z is a control variable. For the supply chain layer, the control variables refer to transportation volume, delivery volume, emissions volume, and production volume. For the planning and scheduling layer, the control variables have been initially determined by the scheme and do not need to be considered separately.

[0112] The reliability index F calculated by the calculation method designed according to the present invention can effectively quantify the reliability of the supply chain planning and scheduling scheme. The larger the F value, the higher the resistance of the scheme to the disturbance of uncertain parameters. Conversely, if the F value is too small, it means that the scheme may not be able to achieve the expected effect under the current uncertain parameter disturbance, which is not conducive to stable production. It is necessary to consider changing the parameters of the supply chain planning and scheduling process design or reallocating the corresponding performance indicators, and recalculate on this basis.

[0113] Furthermore, step S5 specifically includes:

[0114] Verify whether there are situations where a nominal point can be specified for uncertain parameters, especially in pricing issues, where it is necessary to determine whether certain key uncertain parameters can be manually set to increase reliability.

[0115] S5.1 When there is an uncertain parameter for which a nominal point can be specified, the uncertain parameter for which a nominal point can be specified is denoted as θ. ′The derivativeless optimization method (DFO) is used to find the nominal point θ of the uncertain parameter that maximizes the reliability index. N The specific optimization propositions are as follows:

[0116]

[0117] θ L ≤θ N ≤θ U

[0118]

[0119] T(δ) = {θ|θ N -δΔθ - ≤θ≤θ N +δΔθ +}

[0120] z L ≤z≤z U

[0121] Where, θ U and θ L Representing the nominal point θ of the uncertain parameter respectively N The upper and lower bounds can vary; if some uncertain parameters cannot be specified at a nominal point, then their upper and lower bounds are equal to the nominal point, i.e., θ. L =θ N =θ U ;z U and z L The upper and lower bounds of the control variable z are represented; the optimal nominal point of the uncertain parameter is determined by the derivative-free optimization method, while maximizing the reliability index. The hyperrectangle calculated based on the optimal nominal point is the largest space of uncertain parameters.

[0122] S5.2 When there are no uncertain parameters for which a nominal point can be specified, meaning that all uncertain parameters cannot be manually intervened to reduce uncertainty, an active constraint strategy needs to be applied to identify active constraints. This involves imposing one or more key constraints on the reliability index of the solution within the scaled space of uncertain parameters. Based on the performance indices corresponding to the returned constraints, decision-makers can dynamically adjust the values ​​of these performance indices to achieve a balance between performance and reliability requirements, thereby improving the reliability of the solution.

[0123] The following is a concrete implementation example. The structure of a planning and scheduling processing model for a certain petrochemical process industry is as follows: Figure 2 As shown, the system consists of three raw materials, two products, three intermediate materials, and five types of processing tasks, with corresponding parameters shown in Tables 1-7. The total scheduling cycle is 24 hours.

[0124] Table 1

[0125] pyrolysis reaction Polymerization reaction Hydrogen cracking hydrogenation reaction Deep processing Fixed production costs 1500 1100 1000 900 1600 Variable production costs 11 6 4 5 10

[0126] Table 2

[0127]

[0128]

[0129] Table 3

[0130] Material conversion rate pyrolysis reaction Polymerization reaction Hydrogen cracking hydrogenation reaction Deep processing Ethane -1 naphtha -0.5 hydrogen -0.5 -0.2 ethylene 1 -0.4 Unreacted materials 0.6 -0.8 0.1 propylene 1 -0.6 Saturated hydrocarbons 1 -1 ethylene-propylene copolymer 0.4 Isobutane 0.9

[0131] Table 4

[0132] Inventory costs Shortage transfer costs Selling price ethylene-propylene copolymer 12 95 880 Isobutane 14 105 1120

[0133] Table 5

[0134] Ethane naphtha hydrogen Unit cost of raw material procurement 310 190 390

[0135] Table 6

[0136] ethylene Unreacted materials propylene Saturated hydrocarbons Inventory capacity limit 100 200 150 100

[0137] Table 7

[0138] need First cycle Second cycle ethylene-propylene copolymer 210 260 Isobutane 330 290

[0139] Based on the original plan scheduling optimization proposition, solve for the optimal supply chain planning and scheduling scheme and calculate the corresponding value parameters. The total profit is 625,483.33, of which sales volume is 1,108,000.00, production cost is 78,633.33, inventory cost and stockout cost are 0, and raw material cost is 403,883.33. Note: In this scheme, there is no inventory backlog or stockout transfer. Assume the decision-maker's expected minimum total profit is 550,000, where sales cannot be lower than 1,050,000, production cost cannot be higher than 85,000, inventory cost cannot be higher than 2,000, stockout cost cannot be higher than 10,000, and raw material cost cannot be higher than 430,000. Unreacted materials are special intermediate products; if not completely consumed, they will pollute the environment. The emission of unreacted materials cannot exceed 200 units.

[0140] At this point, fixed production costs, variable production costs, raw material purchase prices, commodity sales prices, unit inventory costs, unit stockout transfer costs, and commodity demand are all uncertain parameters, with nominal points as shown in the table, and a range fluctuation of 10%. The reliability index for solving the problem is 0.428. Since there are no control variables in the planning and scheduling model, according to the active constraint algorithm, only one inequality constraint will trigger the boundary conditions. In this embodiment, the first constraint to trigger the allowable boundary conditions is the total profit constraint, with sales still having a growth potential of 10531.26. Production costs have a margin of 2997.87, inventory costs have a margin of 2000, stockout costs have a margin of 2657.30, and purchase costs have a margin of 8813.57. The emission of unreacted materials is 164 units, meeting the pollutant emission requirements.

[0141] A reliability index of at least 0.5 is required for the proposed solution to be adopted. To improve reliability while keeping the solution unchanged, we first consider whether there are any uncertain parameters among a series of uncertain parameters for which the decision-maker can specify a nominal point. In this case, only the commodity selling price is subject to human intervention. According to our analysis, a more suitable nominal point needs to be provided for the decision-maker. With all uncertain parameters remaining within their fluctuation range, based on the pricing recommendation obtained using the derivative-free optimization method, the reliability index of the solution reaches its maximum value of 0.575 when the initial price of ethylene-propylene copolymer is 869.72 and the initial price of isobutane is 1227.56.

[0142] Corresponding to the aforementioned embodiments of the petrochemical supply chain planning and scheduling scheme reliability analysis method, the present invention also provides embodiments of the petrochemical supply chain planning and scheduling scheme reliability analysis device.

[0143] See Figure 3 The petrochemical supply chain planning and scheduling scheme reliability analysis device provided in this embodiment of the invention includes a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it is used to implement the petrochemical supply chain planning and scheduling scheme reliability analysis method in the above embodiment.

[0144] The embodiments of the petrochemical supply chain planning and scheduling scheme reliability analysis device of the present invention can be applied to any device with data processing capabilities, such as a computer. The device embodiments can be implemented through software, hardware, or a combination of both. Taking software implementation as an example, as a logical device, it is formed by the processor of any data processing device loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, such as... Figure 3The diagram shown is a hardware structure diagram of any data processing-capable device, including the petrochemical supply chain planning and scheduling scheme reliability analysis device of the present invention. (Except for...) Figure 3 In addition to the processor, memory, network interface, and non-volatile memory shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0145] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0146] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0147] This invention also provides a computer-readable storage medium storing a program thereon, which, when executed by a processor, implements the petrochemical supply chain planning and scheduling scheme reliability analysis method described in the above embodiments.

[0148] The computer-readable storage medium can be an internal storage unit of any data processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device of any data processing device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices of any data processing device. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0149] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the scope of protection of one or more embodiments of this specification.

Claims

1. A petrochemical supply chain planning and scheduling scheme reliability analysis method, characterized in that, The method comprises the following steps: S1, a supply chain planning and scheduling mathematical model for increasing revenue and reducing energy consumption is established for a petrochemical supply chain and a planning and scheduling process; S2, an optimal supply chain planning and scheduling scheme is solved by optimization proposition of the mathematical model without considering uncertainty, and expected values of various performance indexes and corresponding concession constraints are established according to the optimal scheme; S3, a fluctuation range of various uncertain parameters in the supply chain layer and the planning and scheduling layer is set, a reliability analysis model is established, and a feasibility test is performed to determine whether the optimal scheme is completely reliable within the given fluctuation range of the uncertain parameters, and if not, S4 is executed; specifically: The concession constraints established in S2, the set uncertain parameters and the set fluctuation range of the uncertain parameters are introduced into the supply chain planning and scheduling mathematical model in S1 to construct a reliability analysis model, and then the feasibility of the uncertain parameters is tested; The feasibility test is used to check whether the optimal solution has at least one set of control variables z for any set of uncertain parameters θ, such that all the constraints f g (z, θ) are feasible, mathematically expressed as follows: Wherein, G represents a set of constraint conditions of the mathematical model in S1 and the concession constraints established in S2 according to the optimal scheme; for a given fluctuation range θ N - Δθ - ≤ θ ≤ θ N + Δθ + , wherein θ N is a nominal point of the uncertain parameter θ, Δθ - , Δθ + are respectively a minimum fluctuation value and a maximum fluctuation value of the uncertain parameter θ, if the uncertain parameter θ in the fluctuation range satisfies formula (16), then the feasibility test of the optimal scheme passes, which indicates that the optimal scheme is completely reliable in the given fluctuation range of the uncertain parameter; if the feasibility test fails, then S4 is executed; S4, the reliability index of the reliability analysis model is calculated and used as a quantitative index of the reliability of the scheme; S5, whether the reliability index meets the reliability requirements of the user is determined, if yes, it means that the scheme is reliable, and the reliability index is returned to the user; if not, two cases are considered: For the uncertain parameters that can specify a nominal point, a derivative-free optimization method is used to solve the optimal nominal point of the uncertain parameters, and then the uncertain parameters are marked as unable to specify a nominal point, and returned to S4; For the uncertain parameters that cannot specify a nominal point, key constraint conditions that play a limiting role are identified, and whether the key constraint conditions can be relaxed is determined, if yes, it is returned to S4, otherwise, it means that the scheme is unreliable.

2. The petrochemical supply chain planning and scheduling scheme reliability analysis method of claim 1, wherein, The supply chain part of the mathematical model is a multi-level model including four links of suppliers, factories, distributors and customers, and has related constraints as follows to form a supply chain model: Equation (1) represents the mass balance constraint of material s in factory m during period t, where, This represents the inventory of material s at factory m at the beginning of period t. F represents the inventory of material s at factory m at the end of period t; d,d′,s,t η represents the amount of material s transported from scenario d to scenario d' during cycle t; m,s Ba represents the conversion efficiency of material s in factory m between output and input. m,t Equation (2) represents the batch size of factory m in cycle t; Equation (2) represents the quality balance constraint of distributor v. This represents the quantity of material s for distributor v at the beginning of period t. This represents the quantity of material s held by distributor v at the end of period t. v,s,t Let represent the amount of material s delivered by distributor v to customer s in period t; M represents the set of all factories m, H represents the set of all suppliers h, V represents the set of all distributors v, S represents the set of all materials s, and T represents the set of all periods t. Equations (3) and (4) represent periodic inventory constraints, i.e., the coupling between the material quantities of different periods, plants, and distributors, where material loss due to storage or transportation can occur at the change of period, with a loss rate of a s , where represents the s material quantity of the d scene at the initial time. Equations (3)-(7) represent upper and lower bounds on the transportation and inventory quantities, respectively, where F h,m,s,t represents the transportation quantity of material s from supplier h to factory m at period t; Sp h,s,t represents the maximum supply quantity of material s from supplier h at period t, represents the upper bound on the inventory capacity of material s at node d.

3. The petrochemical supply chain planning and scheduling scheme reliability analysis method of claim 2, wherein, The planning and scheduling part of the mathematical model includes constraints of the planning layer and constraints of the scheduling layer, the planning layer is responsible for the formulation of overall production planning, and the scheduling layer focuses on specific resource allocation and execution details; the planning layer and the scheduling layer are coupled through shared variables and synchronous constraints to form a comprehensive planning and scheduling model; The planning constraints are specifically as follows: wherein, denotes the inventory of material s at period t, denotes the demand of material s at period t, denotes the backorder of material s at period t, denotes the actual delivery of material s at period t, denotes the production plan of material s at period t, denotes the state of material s at operation node n at period t, |N| denotes the total number of operation nodes, denotes the state of material s at the beginning of period t, s P denotes the set of products; The scheduling constraints are specifically as follows: Equations (12) and (13) represent the processing logic and capacity constraint of the equipment performing the production tasks, where J represents the set of generating and processing equipment, and I represents the set of tasks; is a 0-1 binary variable indicating whether the equipment j performs the task i at the operation node n in period t, represents the batch size of the equipment j performing the task i at the operation node n in period t, and respectively represent the lower and upper bounds of the batch size of the equipment j performing the task i, I j represents the set of tasks that the equipment j can perform, J i represents the set of equipment that can perform the task i, and N represents the set of operation nodes; Equations (14) and (15) represent the material state transformation of the production process, where represents the consumption rate of material s per unit batch size of the task i in the execution process, represents the generation rate of material s per unit batch size of the task i in the execution process, I s represents the set of tasks related to the processing of material s.

4. The petrochemical supply chain planning and scheduling solution reliability analysis method of claim 1, wherein, In S2, for the optimal supply chain planning and scheduling scheme, comprehensive analysis of total profit, total sales, production cost, transportation cost, inventory cost, shortage cost, maintenance cost, energy consumption and pollutant emission is needed; when quantifying the reliability of the supply chain planning and scheduling scheme, the expected value of profit, the expected value of cost budget, the maximum value of energy consumption and the maximum value of pollutant emission are set, and then the concession constraints of the corresponding performance indexes are constructed.

5. The petrochemical supply chain planning and scheduling scheme reliability analysis method of claim 1, wherein, The uncertain parameters in the supply chain layer include supply-side uncertain parameters, demand-side uncertain parameters and transportation cost; the uncertain parameters in the planning and scheduling layer include batch processing production conversion rate, pollutant emission rate and batch processing production cost.

6. The petrochemical supply chain planning and scheduling solution reliability analysis method of claim 1, wherein, The reliability index is a quantitative index for measuring the proportion between the completely feasible uncertain parameter space and the expected uncertain parameter fluctuation range, and the mathematical expression is as follows: F = max δ s.t.max θ∈T(δ) min z max g∈G f g (z,θ)≤0 T(δ) = {θ | θ N - δ Δθ - ≤ θ ≤ θ N + δ Δθ +} Wherein, δ is the scaling coefficient of the uncertain parameter space; T(δ) represents the scaled uncertain parameter space; The solution of the above problem is converted into the following equivalent convex optimization form: F = min k∈K δ k s.t.δ k = max δ,z δ s.t.f g (z,θ)≤0,g∈G θ = θ N + δΔθ k where k is a vertex of the uncertain parameter space, K is a vertex set of the uncertain parameter space, δ k denotes the scaling factor corresponding to vertex k; Δθ k denotes the maximum fluctuation value of the uncertain parameter corresponding to vertex k; f g (z, θ) denotes the inherent constraints generated when the mathematical model is constructed in S1, and the performance index-based concession constraints in S2; θ is a given uncertain parameter; z is a control variable.

7. The petrochemical supply chain planning and scheduling scheme reliability analysis method of claim 6, wherein, When there are uncertain parameters that can specify the nominal point in S5, the non-derivative optimization method is used to find the uncertain parameter nominal point θ that can maximize the reliability index N The specific optimization proposition is as follows: θ L ≤θ N ≤θ U T(δ) = {θ | θ N - δ Δθ - ≤ θ ≤ θ N + δ Δθ +} z L ≤z≤z U where θ U and θ L denote the nominal point of uncertain parameters θ N and its upper and lower bounds, respectively; if the nominal point of some uncertain parameters cannot be specified, its upper and lower bounds are equal to the nominal point, i.e., θ L = θ N = θ U ; z U and z L denote the upper and lower bounds of control variables z; the optimal nominal point of uncertain parameters is determined by using the derivative-free optimization method, and the reliability index is maximized, and the hyper-rectangle calculated according to the optimal nominal point is the largest uncertain parameter space.

8. The petrochemical supply chain planning and scheduling solution reliability analysis method of claim 1, wherein, When there is no uncertain parameter capable of specifying the nominal point in S5, an active constraint identification strategy needs to be applied to identify the activated constraint, that is, one or more key constraint conditions of the scheme reliability index are limited in the scaled uncertain parameter space; according to the performance index corresponding to the returned constraint condition, the decision maker dynamically adjusts the value of the performance index to achieve a balance between the performance index and the reliability requirement, thereby improving the reliability of the scheme. 9.A petrochemical supply chain planning and scheduling scheme reliability analysis apparatus, comprising a memory and one or more processors, wherein the memory stores executable code, and the apparatus is characterized in that, The processor, when executing the executable code, is configured to implement the petrochemical supply chain planning and scheduling scheme reliability analysis method according to any one of claims 1-8.

Images