Intelligent collaborative operation system for supply chain visualization

Abstract

This invention belongs to the field of supply chain collaboration, specifically disclosing a visualized intelligent collaborative operation system for the supply chain. The system includes: a supply chain basic parameter construction module, a multi-objective collaborative delivery digital model construction module, a multi-objective hybrid selection module, and a visualization module. Based on the probability distribution characteristics of uncertain demand parameters, this solution employs a mixed-integer programming method to construct a multi-objective optimization overall model, achieving integrated modeling of the entire supply chain collaborative delivery process from design to operation. By integrating genetic algorithms and simulated annealing algorithms, a multi-objective hybrid selection method based on a parallel selection strategy is designed. A roulette wheel selection method is used to screen high-fitness individuals, and differentiated mutation is designed with a non-uniform mutation mechanism introduced. Simultaneously, the intermediate population is used as the initial solution for the simulated annealing algorithm for local optimization, significantly improving the rationality and timeliness of supply chain collaborative decision-making.

Description

Technical Field

[0001] This invention relates to the field of supply chain collaboration, specifically to a visual intelligent collaborative operation system for supply chains. Background Technology

[0002] Intelligent collaborative operation systems for supply chains utilize technologies such as the Internet of Things (IoT), big data, and artificial intelligence (AI) to collect and monitor various data within the supply chain in real time. This allows for automatic adjustments to supply chain plans and strategies, promoting collaboration among stakeholders and improving overall operational efficiency and responsiveness. However, traditional supply chain collaborative delivery modeling suffers from several drawbacks. These include fragmented design and operational models, insufficient quantitative analysis of demand uncertainty, a lack of scientific mathematical models supporting collaborative center location selection and supply-demand matching planning, failure to consider cost accounting deviations due to special cargo losses in inventory control, and the inability to achieve collaborative optimization through independent vehicle routing and loading planning. Furthermore, traditional supply chain collaborative decision-making relies on simplistic population initialization methods, lacks differentiated selection strategies for multi-objective optimization, and its crossover and mutation methods lack specificity and fail to adapt to the characteristics of different decision-making stages within the supply chain. This can easily lead to local optima, resulting in slow algorithm convergence and insufficient optimality and diversity of multi-objective decision solutions, making it unsuitable for the complex decision-making scenarios of intelligent supply chain collaboration. Summary of the Invention

[0003] To address the aforementioned issues and overcome the shortcomings of existing technologies, this invention provides a visualized intelligent collaborative operation system for the supply chain. It addresses the technical problems inherent in traditional supply chain collaborative delivery modeling, such as the disconnect between design and operation models, insufficient quantitative analysis of demand uncertainty, lack of scientific mathematical model support for collaborative center location selection and supply-demand matching planning, failure to consider cost accounting deviations due to special cargo losses in inventory control, and the inability to achieve collaborative optimization through independent design of vehicle routes and loading plans. This solution, based on the probability distribution characteristics of demand uncertainty parameters, accurately quantifies the expected order volume. It employs a mixed-integer programming method, relies on the economic order quantity model, and incorporates green computing and energy management requirements. It integrates all sub-models to construct a multi-objective optimization model centered on minimizing collaboration, cost, and vehicle sub-objectives, and maximizing delivery on-time rate. This achieves integrated modeling of the entire supply chain collaborative delivery process from design to operation, accurately solving the problem of order volume quantification under demand uncertainty, scientifically achieving optimal planning for collaborative center location selection and supply-demand matching, finely controlling inventory costs and reducing economic losses caused by special cargo losses, and achieving collaborative optimization of delivery routes and loading plans. This solution addresses the technical problems in traditional supply chain collaborative decision-making, such as the simplistic population initialization method, lack of differentiated selection strategies for multi-objective optimization, and the lack of targeted crossover and mutation methods that fail to adapt to the characteristics of different decision-making stages in the supply chain. These issues easily lead to local optima, resulting in slow algorithm convergence, insufficient optimality and diversity of multi-objective decision solutions, and an inability to adapt to the complex decision-making scenarios of intelligent supply chain collaboration. This solution integrates genetic algorithms and simulated annealing algorithms, designing a multi-objective hybrid selection method based on parallel selection strategies. It employs a roulette wheel selection method to screen high-fitness individuals and designs differentiated mutation with a non-uniform mutation mechanism. Simultaneously, it uses the intermediate population as the initial solution for simulated annealing algorithm for local optimization, achieving an efficient fusion of the global search advantage of genetic algorithms and the local optimization advantage of simulated annealing algorithm. This effectively avoids the algorithm getting trapped in local optima and accelerates its convergence speed. It can accurately adapt to the decision-making characteristics of different stages in supply chain system design, inventory control, and vehicle routing optimization, significantly improving the rationality and timeliness of supply chain collaborative decision-making. From the decision-making level, it ensures the efficient collaborative operation of all participants in the supply chain, including manufacturing, warehousing, and distribution.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides a supply chain visualization intelligent collaborative operation system, which includes a supply chain basic parameter construction module, a multi-objective collaborative distribution digital model construction module, a multi-objective hybrid selection module, and a visualization display module;

[0005] The supply chain basic parameter construction module constructs supply chain basic parameters, which include node attribute parameters, transportation and loading parameters, demand uncertainty parameters, inventory control parameters, and collaborative operation parameters.

[0006] The multi-objective collaborative delivery digital model construction module, based on the basic parameters of the supply chain, constructs a multi-objective optimization overall model through an integrated multi-objective collaborative delivery optimization mathematical model method;

[0007] The multi-objective hybrid selection module, based on the above-mentioned multi-objective optimization overall model, integrates genetic algorithms and simulated return algorithms to design a multi-objective hybrid selection method based on a parallel selection strategy, and makes decisions on intelligent collaboration in the supply chain.

[0008] The visualization module displays key supply chain information, including inventory levels, order status, and logistics transportation routes, in the form of intuitive charts, maps, and reports.

[0009] Furthermore, in the multi-objective collaborative delivery digital model construction module, the integrated multi-objective collaborative delivery optimization mathematical model method specifically includes the following steps:

[0010] Step A1: Construct a retailer order volume calculation model. Based on the probability distribution characteristics of the demand uncertainty parameters, calculate the retailer's expected order volume within the cycle period, which serves as the basis for subsequent cost calculation and optimization decisions.

[0011] Step A2: Constructing the collaborative optimization sub-model. Using a 0-1 mixed-integer programming method, optimize the location scheme of the collaborative center and the supply-demand matching relationship between the manufacturing plant and the collaborative center. The sub-objective is to minimize the total cost in the collaborative system design phase. The formula used for the collaborative sub-objective function is as follows: ;

[0012] In the formula, It is a collaborative sub-objective function. It is the number of candidate points for the collaboration center. Yes Traversal index, It refers to the number of manufacturing plants. Yes Traversal index, It is a 0-1 decision variable; a value of 1 indicates that a candidate node for the collaboration center is selected, and a value of 0 indicates that no node is selected. It is a candidate point for the collaboration center. Fixed construction costs, It is a manufacturing plant To Collaboration Center Candidate Points Supply volume It is a manufacturing plant The unit production cost It is a manufacturing plant To Collaboration Center Candidate Points The unit transportation cost;

[0013] Step A3: Construction of the operational inventory control sub-model. Based on the economic order quantity model, calculate the inventory costs of retailers and collaboration centers, while also considering the special goods loss costs, and construct the operational inventory control sub-model. The inventory costs of retailers and collaboration centers include the retailer's economic order quantity, the retailer's inventory holding cost, the retailer's order processing cost, the special goods loss cost, and the collaboration center's inventory holding cost.

[0014] Step A4: Design and construct an integrated inventory control model, fusing the collaborative optimization sub-model and the operational inventory control sub-model as the core sub-model for multi-objective optimization. Simultaneously, supplement the uniqueness constraint of retailer orders and integrate the cost objective function. The formula used for the cost objective function is as follows: ;

[0015] In the formula, It is the cost objective function. It is the number of retailers. Yes Traversal index, , and Retailers Inventory holding costs, order processing costs, and special goods loss costs, It is a candidate point for the collaboration center. Inventory holding costs;

[0016] Step A5: Construction of joint vehicle route optimization model. Combining green computing and energy management requirements, a joint vehicle route optimization model is constructed to achieve coordinated optimization of delivery route planning and loading planning. The vehicle sub-objectives are minimizing delivery operating costs and minimizing total carbon emissions, while also satisfying constraints on vehicle load, time window, and loading / unloading time.

[0017] Step A6: Construct a multi-objective optimization model. Integrate the above-mentioned retailer order volume calculation model, collaborative optimization sub-model, operational inventory control sub-model, design and operational inventory control integrated model, and vehicle route joint optimization model to construct a multi-objective optimization model with the objectives of minimizing collaborative sub-objectives, minimizing cost objectives, minimizing vehicle sub-objectives, and maximizing delivery on-time rate.

[0018] Furthermore, in the multi-objective hybrid selection module, the multi-objective hybrid selection method based on the parallel selection strategy specifically includes the following steps:

[0019] Step B1: Population initialization. Based on the decision variable characteristics of collaborative delivery, a chromosome structure is designed using a combination of real number encoding and 0-1 encoding. The chromosome is divided into three core segments: system design segment, inventory control segment, and vehicle route joint optimization segment. An initial population is generated using a combination of random initialization and heuristic initialization. In the initial population, heuristically initialized individuals account for 30%, while randomly initialized individuals account for 70%, generated based on the shortest distance principle and supply-demand balance principle.

[0020] Step B2: Parallel selection operation. The initial population is randomly divided into subpopulations. Each subpopulation corresponds to one of the four optimization objectives: minimizing the collaborative sub-objective, minimizing the cost objective, minimizing the vehicle sub-objective, and maximizing the on-time delivery rate. For each subpopulation, the fitness value of an individual is calculated based on its corresponding optimization objective. The fitness value of the cost objective is the reciprocal of the objective function value, and the fitness value of the on-time delivery rate objective is itself. The roulette wheel selection method is used for each subpopulation to select the top 50% of individuals with high fitness values ​​to form a candidate individual set. The candidate individual sets of the subpopulations are merged to obtain the parent population.

[0021] Step B3: Crossover operation. A segmented crossover strategy is adopted, with different crossover methods for different segments of the chromosome. The system design segment uses a single-point crossover method with crossover points randomly selected. The inventory control segment uses an arithmetic crossover method, where the gene value of the offspring is a weighted average of the gene values ​​of the parent individuals, with weights randomly generated. The vehicle path joint optimization segment uses an ordered crossover method to generate offspring individuals, which are then merged with the parent population to obtain an intermediate population.

[0022] Step B4: Mutation operation. Design a differentiated mutation strategy and introduce a non-uniform mutation mechanism. The system design section uses the bit flip mutation method to flip the 0-1 variables. The inventory control section uses the non-uniform mutation method. The vehicle route joint optimization section uses the route exchange mutation method to randomly exchange the delivery vehicle or route order of two retailers.

[0023] Step B5: Simulated annealing local optimization. Using the intermediate population as the initial solution of the simulated annealing algorithm, perform local optimization on each individual to obtain the optimized population;

[0024] Step B6: Population update. Based on the non-dominance of Pareto optimal solutions, the optimized population is screened, non-dominant individuals are retained, and new randomly generated individuals are added. The maximum number of evolutionary iterations is preset, and steps B2 to B6 are repeated until the maximum number of evolutionary iterations is reached.

[0025] The beneficial effects achieved by adopting the above solution are as follows:

[0026] (1) In the traditional supply chain collaborative delivery modeling process, there are technical problems such as the separation of design and operation models, insufficient quantitative analysis of demand uncertainty, lack of scientific mathematical model support for collaborative center location and supply and demand matching planning, failure to consider special goods loss in inventory control leading to cost accounting deviation, and the inability to achieve collaborative optimization by independently designing vehicle routes and loading plans. Based on the probability distribution characteristics of demand uncertainty parameters, this solution accurately quantifies the expected value of order volume, adopts mixed integer programming method, relies on the economic order quantity model, combines green computing energy management requirements, integrates all sub-models to construct a multi-objective optimization model with the minimization of collaboration, cost, vehicle sub-objectives and the maximization of delivery timeliness as the core, realizes the model integration of the entire process of supply chain collaborative delivery from design to operation, accurately solves the problem of order volume quantification under demand uncertainty, scientifically realizes the optimal planning of collaborative center location and supply and demand matching, refines the control of inventory costs and reduces the economic losses caused by special goods loss, and achieves collaborative optimization of delivery routes and loading plans.

[0027] (2) In traditional supply chain collaborative decision-making, the population initialization method is simple, and there is no differentiated selection strategy designed for multi-objective optimization. The crossover and mutation methods lack specificity and cannot adapt to the characteristics of different decision-making links in the supply chain. They are prone to getting trapped in local optima, which ultimately results in slow algorithm convergence speed, insufficient optimality and diversity of multi-objective decision solutions, and inability to adapt to the complex decision-making scenarios of intelligent supply chain collaboration. This solution integrates genetic algorithm and simulated annealing algorithm to design a multi-objective hybrid selection method based on parallel selection strategy. It adopts roulette wheel selection method to screen individuals with high fitness, and designs differentiated mutation and introduces non-uniform mutation mechanism. At the same time, the intermediate population is used as the initial solution of simulated annealing algorithm for local optimization. This realizes the efficient integration of the global search advantage of genetic algorithm and the local optimization advantage of simulated annealing algorithm, effectively avoids the algorithm getting trapped in local optima, and accelerates the convergence speed of the algorithm. It can accurately adapt to the decision-making characteristics of different links such as supply chain system design, inventory control, and vehicle route optimization, and significantly improves the rationality and timeliness of supply chain collaborative decision-making. It ensures the efficient collaborative operation of all participants in supply chain manufacturing, warehousing, and distribution from the decision-making level. Attached Figure Description

[0028] Figure 1 A module connection diagram of a supply chain visualization intelligent collaborative operation system provided by the present invention;

[0029] Figure 2 The flowchart illustrates the steps of the multi-objective hybrid selection method based on a parallel selection strategy provided by this invention.

[0030] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0031] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] Example 1: See Figure 1 This embodiment provides a supply chain visualization intelligent collaborative operation system, which includes a supply chain basic parameter construction module, a multi-objective collaborative distribution digital model construction module, a multi-objective hybrid selection module, and a visualization display module.

[0033] The supply chain basic parameter construction module constructs supply chain basic parameters, which include node attribute parameters, transportation and loading parameters, demand uncertainty parameters, inventory control parameters, and collaborative operation parameters.

[0034] The multi-objective collaborative delivery digital model construction module, based on the basic parameters of the supply chain, constructs a multi-objective optimization overall model through an integrated multi-objective collaborative delivery optimization mathematical model method;

[0035] The multi-objective hybrid selection module, based on the above-mentioned multi-objective optimization overall model, integrates genetic algorithms and simulated return algorithms to design a multi-objective hybrid selection method based on a parallel selection strategy, and makes decisions on intelligent collaboration in the supply chain.

[0036] The visualization module displays key supply chain information, including inventory levels, order status, and logistics transportation routes, in the form of intuitive charts, maps, and reports.

[0037] Example 2: See Figure 1 This embodiment, based on the above embodiment, includes the following steps in the multi-objective collaborative delivery digital model construction module: The integrated multi-objective collaborative delivery optimization mathematical model method:

[0038] Step A1: Construct a retailer order volume calculation model. Based on the probability distribution characteristics of the demand uncertainty parameters, calculate the retailer's expected order volume within the cycle period, which serves as the basis for subsequent cost calculation and optimization decisions.

[0039] Step A2: Constructing the collaborative optimization sub-model. Using a 0-1 mixed-integer programming method, optimize the location scheme of the collaborative center and the supply-demand matching relationship between the manufacturing plant and the collaborative center. The sub-objective is to minimize the total cost in the collaborative system design phase. The formula used for the collaborative sub-objective function is as follows: ;

[0040] In the formula, It is a collaborative sub-objective function. It is the number of candidate points for the collaboration center. Yes Traversal index, It refers to the number of manufacturing plants. Yes Traversal index, It is a 0-1 decision variable; a value of 1 indicates that a candidate node for the collaboration center is selected, and a value of 0 indicates that no node is selected. It is a candidate point for the collaboration center. Fixed construction costs, It is a manufacturing plant To Collaboration Center Candidate Points Supply volume It is a manufacturing plant The unit production cost It is a manufacturing plant To Collaboration Center Candidate Points The unit transportation cost;

[0041] Step A3: Construction of the operational inventory control sub-model. Based on the economic order quantity model, calculate the inventory costs of retailers and collaboration centers, while also considering the special goods loss costs, and construct the operational inventory control sub-model. The inventory costs of retailers and collaboration centers include the retailer's economic order quantity, the retailer's inventory holding cost, the retailer's order processing cost, the special goods loss cost, and the collaboration center's inventory holding cost.

[0042] Step A4: Design and construct an integrated inventory control model, fusing the collaborative optimization sub-model and the operational inventory control sub-model as the core sub-model for multi-objective optimization. Simultaneously, supplement the uniqueness constraint of retailer orders and integrate the cost objective function. The formula used for the cost objective function is as follows: ;

[0043] In the formula, It is the cost objective function. It is the number of retailers. Yes Traversal index, , and Retailers Inventory holding costs, order processing costs, and special goods loss costs, It is a candidate point for the collaboration center. Inventory holding costs;

[0044] Step A5: Construction of joint vehicle route optimization model. Combining green computing and energy management requirements, a joint vehicle route optimization model is constructed to achieve coordinated optimization of delivery route planning and loading planning. The vehicle sub-objectives are minimizing delivery operating costs and minimizing total carbon emissions, while also satisfying constraints on vehicle load, time window, and loading / unloading time.

[0045] Step A6: Construct a multi-objective optimization model. Integrate the above-mentioned retailer order volume calculation model, collaborative optimization sub-model, operational inventory control sub-model, design and operational inventory control integrated model, and vehicle route joint optimization model to construct a multi-objective optimization model with the objectives of minimizing collaborative sub-objectives, minimizing cost objectives, minimizing vehicle sub-objectives, and maximizing delivery on-time rate.

[0046] Example 3: See Figure 1 This embodiment, based on the above embodiment, describes a multi-objective hybrid selection method based on a parallel selection strategy within the multi-objective hybrid selection module, specifically including the following steps:

[0047] Step B1: Population initialization. Based on the decision variable characteristics of collaborative delivery, a chromosome structure is designed using a combination of real number encoding and 0-1 encoding. The chromosome is divided into three core segments: system design segment, inventory control segment, and vehicle route joint optimization segment. An initial population is generated using a combination of random initialization and heuristic initialization. In the initial population, heuristically initialized individuals account for 30%, while randomly initialized individuals account for 70%, generated based on the shortest distance principle and supply-demand balance principle.

[0048] Step B2: Parallel selection operation. The initial population is randomly divided into subpopulations. Each subpopulation corresponds to one of the four optimization objectives: minimizing the collaborative sub-objective, minimizing the cost objective, minimizing the vehicle sub-objective, and maximizing the on-time delivery rate. For each subpopulation, the fitness value of an individual is calculated based on its corresponding optimization objective. The fitness value of the cost objective is the reciprocal of the objective function value, and the fitness value of the on-time delivery rate objective is itself. The roulette wheel selection method is used for each subpopulation to select the top 50% of individuals with high fitness values ​​to form a candidate individual set. The candidate individual sets of the subpopulations are merged to obtain the parent population.

[0049] Step B3: Crossover operation. A segmented crossover strategy is adopted, with different crossover methods for different segments of the chromosome. The system design segment uses a single-point crossover method with crossover points randomly selected. The inventory control segment uses an arithmetic crossover method, where the gene value of the offspring is a weighted average of the gene values ​​of the parent individuals, with weights randomly generated. The vehicle path joint optimization segment uses an ordered crossover method to generate offspring individuals, which are then merged with the parent population to obtain an intermediate population.

[0050] Step B4: Mutation operation. Design a differentiated mutation strategy and introduce a non-uniform mutation mechanism. The system design section uses the bit flip mutation method to flip the 0-1 variables. The inventory control section uses the non-uniform mutation method. The vehicle route joint optimization section uses the route exchange mutation method to randomly exchange the delivery vehicle or route order of two retailers.

[0051] Step B5: Simulated annealing local optimization. Using the intermediate population as the initial solution of the simulated annealing algorithm, perform local optimization on each individual to obtain the optimized population;

[0052] Step B6: Population update. Based on the non-dominance of Pareto optimal solutions, the optimized population is screened, non-dominant individuals are retained, and new randomly generated individuals are added. The maximum number of evolutionary iterations is preset, and steps B2 to B6 are repeated until the maximum number of evolutionary iterations is reached.

[0053] Example 4 is based on the above examples. In Example 1, the node attribute parameters include the number of manufacturers, the number of candidate nodes for the collaboration center, the number of retailers, the manufacturing capacity, the unit production cost, the fixed construction cost of the collaboration center, the operating cost coefficient, the geographical coordinates of the retailers, the demand time window, and the unit time stockout cost. The transportation and loading parameters include the unit transportation cost and distance from the manufacturer to the collaboration center, the unit transportation cost and distance from the collaboration center to the retailers, the rated load of the delivery vehicles, the energy consumption rate per unit mileage, the carbon emission coefficient, the fixed vehicle call cost, the unit loading and unloading time cost, the unit loss rate of fresh goods, and the unit loss cost. The demand uncertainty parameters include the demand of retailers per unit time that follows a normal distribution and the number of orders of retailers per period that follows a Poisson distribution. The inventory control parameters include the unit inventory holding cost of retailers, the order processing cost, the economic order quantity coefficient, the safety stock level of the collaboration center, and the inventory turnover cycle. The collaborative operation parameters include the population size, crossover probability, mutation probability, and maximum number of evolution iterations of the genetic algorithm, and the initial temperature, cooling coefficient, termination temperature, and Markov chain length of the simulated annealing algorithm.

[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0055] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0056] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A supply chain visualized intelligent collaborative working system, characterized in that It includes a supply chain basic parameter construction module, a multi-objective collaborative delivery digital model construction module, a multi-objective hybrid selection module, and a visualization display module; The supply chain basic parameter construction module constructs supply chain basic parameters, which include node attribute parameters, transportation and loading parameters, demand uncertainty parameters, inventory control parameters, and collaborative operation parameters. The multi-objective collaborative delivery digital model construction module, based on the basic parameters of the supply chain, constructs a multi-objective optimization overall model through an integrated multi-objective collaborative delivery optimization mathematical model method; The multi-objective hybrid selection module, based on the above-mentioned multi-objective optimization overall model, integrates genetic algorithms and simulated return algorithms to design a multi-objective hybrid selection method based on a parallel selection strategy, and makes decisions on intelligent collaboration in the supply chain. The visualization module displays key supply chain information, including inventory levels, order status, and logistics transportation routes, in the form of intuitive charts, maps, and reports.

2. The supply chain visualization intelligent collaborative operation system according to claim 1, characterized in that, The integrated multi-objective collaborative delivery optimization mathematical model method in the multi-objective collaborative delivery digital model construction module specifically includes the following steps: Step A1: Construct a retailer order volume calculation model. Based on the probability distribution characteristics of the demand uncertainty parameters, calculate the retailer's expected order volume within the cycle period. Step A2: Construction of the collaborative optimization sub-model. The 0-1 mixed integer programming method is used to optimize the location scheme of the collaborative center and the supply and demand matching relationship between the manufacturing plant and the collaborative center. The sub-objective is to minimize the total cost in the collaborative system design phase, which is denoted as the collaborative sub-objective function. Step A3: Construction of the operational inventory control sub-model. Based on the economic order quantity model, calculate the inventory costs of retailers and collaboration centers, while also considering the loss costs of special goods, and construct the operational inventory control sub-model. Step A4: Design and construct an integrated model for inventory control, integrating the collaborative optimization sub-model and the operational inventory control sub-model as the core sub-model for multi-objective optimization, while supplementing the uniqueness constraint of retailer orders and integrating the cost objective function; Step A5: Construction of joint vehicle routing optimization model. Combining green computing and energy management requirements, construct a joint vehicle routing optimization model with minimizing delivery operation costs and minimizing total carbon emissions as vehicle sub-objectives. Step A6: Construct a multi-objective optimization model. Integrate the above-mentioned retailer order volume calculation model, collaborative optimization sub-model, operational inventory control sub-model, design and operational inventory control integrated model, and vehicle route joint optimization model to construct a multi-objective optimization model with the objectives of minimizing collaborative sub-objectives, minimizing cost objectives, minimizing vehicle sub-objectives, and maximizing delivery on-time rate.

3. The supply chain visualization intelligent collaborative operation system according to claim 1, characterized in that, In the multi-objective hybrid selection module, the multi-objective hybrid selection method based on the parallel selection strategy specifically includes the following steps: Step B1: Population initialization. Based on the decision variable characteristics of collaborative delivery, a chromosome structure is designed using a combination of real number encoding and 0-1 encoding. The chromosome is divided into three core segments: system design segment, inventory control segment, and vehicle route joint optimization segment. Step B2: Parallel selection operation. The initial population is randomly divided into subpopulations. Each subpopulation corresponds to one of the four optimization objectives: minimizing the collaborative sub-objective, minimizing the cost objective, minimizing the vehicle sub-objective, and maximizing the on-time delivery rate. For each subpopulation, the fitness value of individuals is calculated based on its corresponding optimization objective. The roulette wheel selection method is used for each subpopulation to select the top 50% of individuals with high fitness values ​​to form a candidate individual set. The candidate individuals sets of the subpopulations are merged to obtain the parent population. Step B3: Crossover operation. A segmented crossover strategy is adopted, with different crossover methods for different segments of the chromosome. The system design segment uses a single-point crossover method with crossover points randomly selected. The inventory control segment uses an arithmetic crossover method, where the gene value of the offspring is a weighted average of the gene values ​​of the parent individuals, with weights randomly generated. The vehicle path joint optimization segment uses an ordered crossover method to generate offspring individuals, which are then merged with the parent population to obtain an intermediate population. Step B4: Mutation operation. Design a differentiated mutation strategy and introduce a non-uniform mutation mechanism. The system design section uses the bit flip mutation method to flip the 0-1 variables. The inventory control section uses the non-uniform mutation method. The vehicle route joint optimization section uses the route exchange mutation method to randomly exchange the delivery vehicle or route order of two retailers. Step B5: Simulated annealing local optimization. Using the intermediate population as the initial solution of the simulated annealing algorithm, perform local optimization on each individual to obtain the optimized population; Step B6: Population update. Based on the non-dominance of Pareto optimal solutions, the optimized population is screened, non-dominant individuals are retained, and new randomly generated individuals are added. The maximum number of evolutionary iterations is preset, and steps B2 to B6 are repeated until the maximum number of evolutionary iterations is reached.

Images