Omnichannel order management system based on block chain

Abstract

The invention relates to the technical field of information, in particular to an omni-channel order management system based on a block chain. Comprising an order management and digital twinning module, an inventory management and cloud network cooperation module, a payment and intelligent contract module, a logistics and edge computing module, a customer service and after-sales management module, an authority management and security module, a data analysis and prediction module and a compliance and audit module. The order management and digital twinning module creates a digital model for an order through a digital twinning technology, and displays the change of an order state in real time; and the inventory management and cloud network cooperation module synchronizes inventory information of each channel in real time and dynamically allocates resources by using a cloud network fusion technology. Through digital twinning, an enterprise can dynamically monitor each link of an order in real time, when a complex supply chain is processed, problems can be predicted in advance, delay or order loss is avoided, and more efficient and more transparent order fulfillment is realized.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an omni-channel order management system based on blockchain. Background Art

[0002] With the development of e-commerce and mobile Internet, the mobile 3C digital product market is experiencing explosive growth. Mobile 3C digital products refer to 3C electronic consumer products including mobile phones. 3C refers to computers, communications, and consumer electronic products. Products mainly include mobile phones, computers, tablets, smart watches, cameras, headphones, speakers, game consoles and other digital devices, which are used in people's daily lives. Due to the rapid update of 3C digital products and frequent changes in market demand, the order management system is facing challenges. The traditional order management system has the following problems:

[0003] The data between supply chain nodes cannot be shared in real time, resulting in information islands and affecting the order processing speed. When facing a peak in orders, the traditional system cannot respond quickly, which can easily lead to delayed or lost order fulfillment.

[0004] Inventory management of mobile phone 3C digital products requires precision and flexibility. Excessive inventory will occupy funds, while insufficient inventory will lead to the inability to deliver orders in time, resulting in unreasonable resource allocation.

[0005] Each node in the supply chain needs to work together, but the traditional order management system cannot share supply chain data in real time, resulting in low supply chain collaboration efficiency and delayed resource scheduling.

[0006] Therefore, those skilled in the art provide an omni-channel order management system based on blockchain to solve the problems raised in the above background technology. Summary of the invention

[0007] In view of the shortcomings of the prior art, the present invention provides an omni-channel order management system based on blockchain to solve the problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a blockchain-based omni-channel order management system, including order management and digital twin modules, inventory management and cloud network collaboration modules, payment and smart contract modules, logistics and edge computing modules, customer service and after-sales management modules, authority management and security modules, data analysis and prediction modules, and compliance and audit modules;

[0009] The order management and digital twin module creates a digital model for the order through digital twin technology and displays the changes in order status in real time;

[0010] The inventory management and cloud-network collaboration module uses cloud-network integration technology to synchronize inventory information of various channels in real time and dynamically allocate resources;

[0011] The payment and smart contract module automates the payment, settlement and clearing processes through blockchain smart contract technology, and automatically triggers payment operations according to different nodes of the order;

[0012] The logistics and edge computing module uses edge computing to synchronize the processing of logistics data to the blockchain in real time, so as to track the logistics status in real time and optimize the delivery route;

[0013] The customer service and after-sales management module manages the after-sales service process of returns, exchanges and repairs, records each link of after-sales service, and combines with digital twins, so that customers and enterprises can view the after-sales status of products in real time;

[0014] The authority management and security module provides identity authentication and authority control based on blockchain technology, so that system users can access and operate the orders or data they are authorized to manage;

[0015] The data analysis and forecasting module provides market demand forecasts, order trend analysis and inventory optimization suggestions through data analysis, machine learning models and historical order data;

[0016] The compliance and audit module uses the immutability of blockchain to automatically record order, payment and inventory operation logs in the system to generate auditable records, making it easier for enterprises and auditing agencies to conduct compliance checks on the system.

[0017] Preferably, the order management and digital twin module includes an order creation unit, an order status tracking unit, a digital twin creation unit, and an order lifecycle management unit; the order creation unit generates orders from different channels; the order status tracking unit tracks the status of each order in real time and records the update of each status on the blockchain; the digital twin creation unit generates a digital twin for each order, displays the real-time status of the order in a visual manner, and simulates it in combination with a prediction model; the order lifecycle management unit is used to manage the entire process from order creation to delivery;

[0018] The inventory management and cloud network collaboration module includes a real-time inventory synchronization unit, an inventory optimization unit, an inventory transfer unit, and an inventory visualization unit; the real-time inventory synchronization unit synchronizes inventory data between different sales channels and storage locations in real time; the inventory optimization unit optimizes inventory distribution and scheduling through cloud network collaboration technology; the inventory transfer unit automatically allocates and transfers according to order requirements and inventory status; the inventory visualization unit provides a real-time visualization interface for inventory, displaying the inventory status of each channel and each warehouse;

[0019] The payment and smart contract module includes a payment processing unit, a smart contract execution unit, a clearing unit, and a payment status tracking unit; the payment processing unit integrates multiple payment methods; the smart contract execution unit automatically executes payment, settlement and refund operations; after the order is completed, the clearing unit automatically executes the settlement and clearing process and records the results on the blockchain; the payment status tracking unit tracks the various states of the payment process and synchronizes the payment progress to the order management module in real time.

[0020] Preferably, the logistics and edge computing module includes a logistics status tracking unit, an edge computing processing unit, a logistics partner management unit, and a logistics optimization unit; the logistics status tracking unit tracks each node of logistics distribution in real time and records changes in logistics status on the blockchain; the edge computing processing unit is used to process logistics data and optimize distribution routes; the logistics partner management unit is integrated with multiple logistics service providers to uniformly manage logistics partners and update logistics status; the logistics optimization unit dynamically adjusts the distribution route based on real-time traffic, order priority, and storage location;

[0021] The data analysis and prediction module includes an order trend analysis unit, an inventory prediction unit, a sales report generation unit, and an intelligent scheduling unit; the order trend analysis unit analyzes historical order data and real-time order trends to generate sales forecasts to help enterprises optimize production and inventory; the inventory prediction unit predicts inventory demand based on order data, market trends and inventory status; the sales report generation unit automatically generates sales reports; the intelligent scheduling unit automatically adjusts inventory scheduling and logistics strategies using forecast data;

[0022] The compliance and audit module includes a compliance recording unit, an audit report generation unit, a compliance monitoring unit, and a tax and regulatory management unit; the compliance recording unit automatically records logs of orders, payments, and inventory operations; the audit report generation unit automatically generates a compliance audit report based on the recorded logs; the compliance monitoring unit monitors operations in the system in real time, detects potential non-compliant behaviors, and automatically issues alerts; the tax and regulatory management unit automatically generates tax records and reports.

[0023] Preferably, the order creation unit uses a multi-channel order priority algorithm. In the case of multi-channel orders, the system needs to dynamically determine the processing order according to the priorities of different channels and order attributes. The formula is as follows: P(i) = α1·C i +α2·V i +α3·T i ,

[0024] Among them, P(i) represents the priority of order i, C i represents the channel weight of the order, V iIndicates the value of the order, T i represents the time urgency of the order, α1, α2, and α3 are weighting coefficients;

[0025] The order status tracking unit predicts and tracks the change of order status through the Markov chain prediction model to describe the transfer of order status at different time nodes. The formula is as follows: P(X t+1 =x j |X t =x i )=p ij ,

[0026] Among them, X t is the state of the order at time t, p ij Is the order by state x i Transfer to state x j probability.

[0027] Preferably, in the digital twin creation unit, the partial differential equation digital twin updated by the dynamic twin needs to be dynamically updated in real time according to the status of the order and external factors. Assuming that the order status is a function of time and external factors, the partial differential equation is used to describe its changes:

[0028]

[0029] Among them, S(t,x) represents the distribution of order status at time t and spatial location x. is a function, u(t,x) is an external input variable;

[0030] In the order lifecycle management unit, the optimal order lifecycle is dynamically planned from order creation to delivery. The system needs to determine the optimal path among multiple operations to complete the order in the shortest time or at the least cost. The formula is as follows:

[0031]

[0032] Among them, V t (s t ) means that at time t, the order is in state s t The maximum value under a t is the decision made by the system at time t, R(s t ,a t ) is the order in state s t Execute decision a t The generated immediate return, γ is the discount factor.

[0033] Preferably, the inventory synchronization algorithm of the real-time inventory synchronization unit based on the distributed consistency protocol is to synchronize the inventory status between multiple nodes in real time, using the inventory synchronization mechanism based on the distributed consistency protocol, which is described by the following formula:

[0034]

[0035] Among them, I t i represents the inventory at node i at time t, represents the inventory increment transferred from node j to node i, represents the inventory quantity sent from node i to node j, It is the set of nodes participating in inventory synchronization;

[0036] The inventory optimization unit is based on the inventory optimization algorithm of linear programming. The goal of inventory optimization is to achieve the optimal inventory level. The optimization problem can be described using linear programming. The goal is to minimize the total cost, including inventory holding cost and shortage cost. The formula is as follows:

[0037]

[0038] Where Z is the total cost, h i is the inventory holding cost of the ith product, I i is the inventory of the ith product, s i is the shortage cost of the ith commodity, S i is the out-of-stock quantity of the ith product;

[0039] Constraints:

[0040] Among them, D i is the demand, T j It is the replenishment quantity from other warehouses or suppliers.

[0041] Preferably, the inventory transfer unit is based on the inventory transfer algorithm of the network flow model. Inventory transfer can be modeled as a network flow problem, the purpose of which is to transfer inventory between different warehouses to meet order requirements and minimize transfer costs. The formula is as follows:

[0042] Among them, C is the total cost of the transfer, c ij is the cost of transferring a unit of inventory from warehouse i to warehouse j, x ij is the inventory quantity transferred from warehouse i to warehouse j;

[0043] Constraints:

[0044] Among them, D j is the demand of warehouse j;

[0045] Capacity constraint: 0≤x ij ≤u ij ,

[0046] Among them, u ij is the maximum capacity that warehouse i can transfer to warehouse j;

[0047] The inventory visualization unit uses a heat map-based inventory distribution visualization algorithm to visualize the distribution of inventory in the system. The inventory distribution data is converted into a heat map, and a two-dimensional kernel density estimation is used to generate a probability density function of the inventory distribution. The formula is as follows:

[0048] in, is the density estimate of the inventory at the (x, y) position, which is used to visualize the inventory distribution. K is the kernel function. (x i ,y i ) is the coordinate position of the i-th warehouse, h x 、h y are the smoothing parameters in the x and y directions.

[0049] Preferably, the logistics status tracking unit is based on the logistics status update model of the blockchain. In order to ensure the real-time synchronization and non-tamperability of the logistics status between different nodes, a blockchain-based status update mechanism is usually adopted. Each logistics status change will be updated in the blockchain. The following hash chain formula is used to describe the logistics status tracking:

[0050] H(S t+1 )=H(H(S t )∥H(ΔS t )),

[0051] Among them, H(S t+1 ) is the state hash value of the logistics at time t+1, H(S t ) is the state hash value of the logistics at time t, ΔS t represents the change of logistics status at time t, ∥ represents serial operation;

[0052] In the edge computing processing unit, the load balancing algorithm for edge node task allocation is used. In the edge computing environment, logistics data processing tasks need to be efficiently allocated among multiple edge nodes to ensure load balancing. A load balancing algorithm based on uniform distribution is adopted, and the formula is as follows:

[0053]

[0054] Among them, L i is the workload assigned to the ith edge node, W iis the processing capacity of edge node i, T is the total task volume;

[0055] Edge node response time formula:

[0056] Among them, R i is the response time of edge node i, C i is the computational effort of the node, N i is the network delay of the node, B i is the bandwidth of the node.

[0057] Preferably, the logistics partner management unit is based on a logistics partner collaboration model based on game theory. Multiple logistics partners need to work together to complete complex transportation tasks. The game theory model is used to optimize the cooperation strategy. It is assumed that the interests and costs of the logistics partners are related, and the following Nash equilibrium model is used to describe it:

[0058]

[0059] Among them, U i is the utility function of partner i, R j is the income of partner j, C j (x j ) is the cost of partner j, which depends on its input x j ;

[0060] Partner allocation optimization formula:

[0061]

[0062] Among them, C i (x i ) is the transportation cost of partner i, T is the total task volume;

[0063] The logistics optimization unit is based on the shortest path logistics optimization of the Dijkstra algorithm. In order to optimize the logistics path, the classic Dijkstra shortest path algorithm is used to calculate the optimal transportation path from the logistics starting point to the end point. The algorithm is used to find the shortest path in a weighted graph. The formula is as follows: d(v)=min(d(u)+w(u,v)).

[0064] Where d(v) represents the shortest path distance from the starting point to node v, d(u) represents the shortest path distance from the starting point to the current node u, and w(u,v) represents the path weight from node u to node v.

[0065] Logistics cost optimization formula:

[0066] Among them, C totalis the total cost of logistics transportation, d ij is the shortest path distance from node i to node j, V ij is the cargo transportation volume between nodes i and j.

[0067] Preferably, the inventory forecasting unit is based on a Bayesian inventory demand forecasting model. The inventory forecast is dynamically adjusted by combining historical orders and current market demand through Bayesian updating. The Bayesian model calculates the posterior probability by taking current information as a priori probability and combining it with future inventory demand. The Bayesian formula is as follows:

[0068] Where P(D|I) is the posterior probability of demand D given information I, P(I|D) is the likelihood function of information I when demand D is known, P(D) is the prior probability of demand, and P(I) is the marginal probability of information;

[0069] The formula for calculating the expected value of inventory demand is: E[D] = ∫D·P(D|I)dD,

[0070] Among them, E[D] is the expected value of future inventory demand, which is used to guide inventory management;

[0071] The sales report generation unit is based on the K-Means model in the clustering algorithm. Sales data usually needs to be classified by customer, product category or time period to generate targeted sales reports. The formula is as follows:

[0072] Among them, J is the objective function of clustering, C i is the i-th cluster, μ i is the center of the ith cluster, x j Belongs to cluster C i data points;

[0073] Sales trend clustering formula:

[0074] Among them, C t is the sales trend clustering index for a certain period of time, S i is the sales volume for that period, is the mean of sales;

[0075] The intelligent scheduling unit is based on the genetic algorithm to optimize the scheduling model. In intelligent scheduling, the scheduling system needs to dynamically adjust the scheduling plan according to factors such as order priority, inventory location, and transportation capacity, and use genetic algorithms to optimize the scheduling strategy. The genetic algorithm optimization formula is:

[0076]

[0077] Among them, f(x) is the objective function of the scheduling, xi is a solution in the solution space, w i is the same as the scheduling scheme x i The relevant weights;

[0078] Fitness function in genetic algorithm:

[0079] Among them, F(x) is the fitness function of the scheduling scheme, and C(x) is the cost of the scheduling scheme;

[0080] Crossover and mutation operations:

[0081] Crossover: Generate new solutions by exchanging some of the gene sequences of the solutions:

[0082] x′=(x1[1:k],x2[k+1:n]),

[0083] Mutation: Change some genes of the solution to explore new solutions:

[0084] x i ′=x i +∈,

[0085] Through multiple generations of iterative optimization, the genetic algorithm will tend to the optimal scheduling solution.

[0086] The present invention provides an omni-channel order management system based on blockchain. It has the following beneficial effects:

[0087] 1. By introducing digital twin technology, the present invention can create a digital twin model for each order and display the order status, inventory and logistics status in real time. The digital twin model is static and can predict potential problems and provide solutions, making the entire life cycle of the order completely transparent. Through digital twins, enterprises can dynamically monitor each link of the order in real time. When dealing with complex supply chains, they can predict problems in advance, avoid delays or order losses, and achieve more efficient and transparent order fulfillment.

[0088] 2. The present invention is based on cloud-network integration and cloud-network collaboration technology. The system realizes supply chain collaboration across multiple platforms and regions, so that inventory, logistics and production data can be shared and collaborated in real time between the cloud and edge computing nodes. Changes in order demand can be immediately reflected in all links of the supply chain, achieving rapid response and resource optimization. At the same time, using cloud-network collaboration technology, each node in the supply chain can collaborate in real time, breaking information silos, realizing intelligent inventory management and supply chain scheduling, shortening order processing time, and improving fulfillment efficiency.

[0089] 3. Through digital twins and machine learning models, the present invention can intelligently analyze order history data, market demand and inventory changes, predict future order trends, and automatically schedule inventory and logistics resources based on the prediction results. Cloud-network collaboration ensures the dynamic allocation of processing capacity so that orders during peak periods can be processed quickly. The system uses advanced prediction algorithms and data analysis to enable enterprises to adjust resource allocation in advance, avoid problems such as excessive or insufficient inventory, and significantly improve the accuracy and timeliness of order processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 is a system diagram of the present invention;

[0091] Figure 2 It is a schematic diagram of the order management and digital twin module of the present invention;

[0092] Figure 3 It is a schematic diagram of the inventory management and cloud network collaboration module of the present invention;

[0093] Figure 4 This is a schematic diagram of the payment and smart contract module of the present invention;

[0094] Figure 5 It is a schematic diagram of the logistics and edge computing module of the present invention;

[0095] Figure 6 It is a schematic diagram of the data analysis and prediction module of the present invention;

[0096] Figure 7 It is a schematic diagram of the compliance and audit module of the present invention. DETAILED DESCRIPTION

[0097] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0098] The present invention is described in detail below in conjunction with the accompanying drawings:

[0099] Example:

[0100] Please refer to the attached Figure 1 -Attached Figure 7, an embodiment of the present invention provides an omni-channel order management system based on blockchain, including order management and digital twin module, inventory management and cloud network collaboration module, payment and smart contract module, logistics and edge computing module, customer service and after-sales management module, authority management and security module, data analysis and prediction module, compliance and audit module; the order management and digital twin module creates a digital model for the order through digital twin technology, and displays the changes in order status in real time; the inventory management and cloud network collaboration module uses cloud network integration technology to synchronize the inventory information of various channels in real time and dynamically allocate resources; the payment and smart contract module automates the payment, settlement and clearing processes through blockchain smart contract technology, and automatically triggers payment operations according to different nodes of the order; the logistics and edge computing module uses edge computing, logistics Data processing can be synchronized to the blockchain in real time to track logistics status in real time and optimize delivery routes; the customer service and after-sales management module manages the after-sales service process of returns, exchanges and repairs, records each aspect of after-sales service, and combined with digital twins, customers and companies can view the after-sales status of products in real time; the authority management and security module provides identity authentication and authority control based on blockchain technology, and system users can access and operate orders or data they are authorized to manage; the data analysis and prediction module provides market demand forecasts, order trend analysis and inventory optimization suggestions through data analysis, machine learning models and historical order data; the compliance and audit module uses the immutability of blockchain to automatically record order, payment and inventory operation logs in the system to generate auditable records, which is convenient for companies and audit agencies to conduct compliance checks on the system.

[0101] The benefits of the order management and digital twin module are to achieve transparent management of the entire life cycle, avoid order delays or losses in advance, improve fulfillment efficiency, and shorten processing time; the benefits of the inventory management and cloud network collaboration module are to avoid inventory backlogs or shortages, improve the response speed of the supply chain, achieve cross-platform and cross-regional collaboration, and improve inventory coordination and overall management efficiency among multiple locations and warehouses; the benefits of the payment and smart contract module are to reduce manual operations, improve efficiency and reduce human errors, and effectively prevent payment fraud and disputes; the benefits of the logistics and edge computing module are to achieve full traceability of orders and real-time tracking of logistics status, improve logistics efficiency, reduce delivery delays, and enhance customer experience; the benefits of the customer service and after-sales management module are to improve transparency, reduce communication problems and disputes caused by information opacity, and improve customer satisfaction; the benefits of the authority management and security module are to reduce the risk of data leakage and facilitate auditing; the benefits of the data analysis and prediction module are to avoid supply chain mismatches and delays, and improve the accuracy and response speed of order processing; the benefits of the compliance and audit module are to reduce manual intervention in compliance management, reduce compliance costs, avoid human modification or data tampering, and improve the credibility of auditing and supervision.

[0102] The order management and digital twin module includes an order creation unit, an order status tracking unit, a digital twin creation unit, and an order lifecycle management unit; the order creation unit generates orders from different channels; the order status tracking unit tracks the status of each order in real time and records the update of each status on the blockchain; the digital twin creation unit generates a digital twin for each order, displays the real-time status of the order in a visual way, and simulates it in combination with the prediction model; the order lifecycle management unit is used to manage the entire process from order creation to delivery; the inventory management and cloud network collaboration module includes a real-time inventory synchronization unit, an inventory optimization unit, an inventory transfer unit, and an inventory visualization unit; the real-time inventory synchronization unit synchronizes orders between different sales channels and storage locations in real time. Synchronize inventory data; the inventory optimization unit optimizes inventory distribution and scheduling through cloud network collaborative technology; the inventory transfer unit automatically allocates and transfers according to order requirements and inventory status; the inventory visualization unit provides a real-time visualization interface for inventory, displaying the inventory status of each channel and warehouse; the payment and smart contract module includes a payment processing unit, a smart contract execution unit, a clearing unit, and a payment status tracking unit; the payment processing unit integrates multiple payment methods; the smart contract execution unit automates payment, settlement, and refund operations; after the order is completed, the clearing unit automatically executes the settlement and clearing process and records the results on the blockchain; the payment status tracking unit tracks the various states of the payment process and synchronizes the progress of the payment to the order management module in real time The logistics and edge computing module includes a logistics status tracking unit, an edge computing processing unit, a logistics partner management unit, and a logistics optimization unit. The logistics status tracking unit tracks each node of logistics distribution in real time and records changes in logistics status on the blockchain. The edge computing processing unit is used to process logistics data and optimize distribution routes. The logistics partner management unit is integrated with multiple logistics service providers to uniformly manage logistics partners and update logistics status. The logistics optimization unit dynamically adjusts the distribution route based on real-time traffic, order priority, and storage location. The data analysis and prediction module includes an order trend analysis unit, an inventory prediction unit, a sales report generation unit, and an intelligent scheduling unit. The order trend analysis unit analyzes historical order data and Real-time order trends generate sales forecasts to help companies optimize production and inventory; the inventory forecasting unit predicts inventory demand based on order data, market trends and inventory status; the sales report generation unit automatically generates sales reports; the intelligent scheduling unit uses forecast data to automatically adjust inventory scheduling and logistics strategies; the compliance and audit module includes a compliance recording unit, an audit report generation unit, a compliance monitoring unit, and a tax and regulatory management unit; the compliance recording unit automatically records logs of orders, payments and inventory operations; the audit report generation unit automatically generates compliance audit reports based on the recorded logs; the compliance monitoring unit monitors operations in the system in real time, detects potential non-compliant behaviors, and automatically issues alerts; the tax and regulatory management unit automatically generates tax records and reports.

[0103] The benefits of the order creation unit are to avoid order dispersion, improve management efficiency, reduce manual operations, and increase the speed of order processing; the benefits of the order status tracking unit are to ensure that data cannot be tampered with and ensure the credibility of order tracking; the benefits of the digital twin creation unit are to foresee potential problems in advance and improve the predictability of order management; the benefits of the order life cycle management unit are to achieve full process monitoring, reduce human intervention in the order management process, and improve the efficiency and accuracy of order processing; the benefits of the real-time inventory synchronization unit are to avoid overselling or inventory shortages and improve the transparency of inventory management; the benefits of the inventory optimization unit are to optimize inventory distribution and scheduling, improve inventory turnover, and reduce inventory holding costs; the benefits of the inventory transfer unit are to improve inventory utilization efficiency, reduce order fulfillment time, and reduce human intervention; the benefits of the inventory visualization unit are to reduce management blind spots caused by opaque inventory information; the benefits of the payment processing unit are to improve payment flexibility and user experience, reduce manual operations, and improve payment efficiency and security; the benefits of the smart contract execution unit are to reduce human intervention and errors, and reduce fraud risks; the benefits of the clearing unit are to improve settlement efficiency, reduce errors and Delay, improve transaction security; the benefits of the payment status tracking unit are to improve the efficiency of fund management; the benefits of the logistics status tracking unit are to avoid delivery delays or errors and improve customer experience; the benefits of the edge computing processing unit are to reduce network delays and effectively reduce the burden of centralized processing; the benefits of the logistics partner management unit are to improve logistics coordination capabilities and management efficiency, and reduce confusion and delays in logistics management; the benefits of the logistics optimization unit are to reduce delivery time and cost, and reduce corporate logistics operating costs; the benefits of the order trend analysis unit are to improve the accuracy of order processing; the benefits of the inventory forecasting unit are to avoid inventory backlogs or out-of-stocks, and improve the foresight and flexibility of inventory management; the benefits of the sales report generation unit are to provide valuable data support; the benefits of the intelligent scheduling unit are to reduce manual intervention and improve the optimization of resource allocation; the benefits of the compliance record unit are to reduce compliance risks; the benefits of the audit report generation unit are to reduce the workload of manual audits and ensure that the audit process is efficient and accurate; the benefits of the compliance monitoring unit are to reduce the burden of manual monitoring and reduce the compliance risks of enterprises; the benefits of the tax and regulatory management unit are to reduce the errors of enterprises in tax processing and reduce compliance risks.

[0104] The order creation unit uses a multi-channel order priority algorithm. In the case of multi-channel orders, the system needs to dynamically determine the processing order based on the priorities of different channels and order attributes. The formula is as follows: P(i) = α1·C i +α2·V i +α3·T i ,

[0105] Among them, P(i) represents the priority of order i, C i represents the channel weight of the order, V i Indicates the value of the order, T irepresents the time urgency of the order, α1, α2, and α3 are weighting coefficients;

[0106] The order status tracking unit predicts and tracks the changes in order status through the Markov chain prediction model to describe the transfer of order status at different time nodes. The formula is as follows: P(X t+1 =x j |X t =x i )=p ij ,

[0107] Among them, X t is the state of the order at time t, p ij Is the order by state x i Transfer to state x j probability.

[0108] The multi-channel order priority algorithm can dynamically adjust the processing order of each order according to the channel weight, order value and time urgency of the order, ensuring that high-priority orders are processed first. In the case of a large number of orders, system resources are effectively allocated to avoid bottlenecks and delays in order processing;

[0109] The Markov chain prediction model helps the system predict the future status and processing progress of an order by calculating the transition probability between different order states. It can help companies better control the order processing rhythm, predict order delays or anomalies in advance, and make corresponding adjustments, thereby improving the accuracy of order fulfillment and customer satisfaction.

[0110] Through the priority algorithm, the system can intelligently sort orders, and the Markov chain model further provides accurate prediction and monitoring of order status changes. Combining the two, the system achieves more intelligent order management. Order processing changes from dynamic allocation of priority to real-time monitoring of status, and the overall process becomes more efficient, transparent and controllable.

[0111] In the digital twin creation unit, the partial differential equation for dynamic twin update The digital twin needs to be dynamically updated in real time according to the order status and external factors. Assuming that the order status is a function of time and external factors, the partial differential equation is used to describe its changes:

[0112]

[0113] Among them, S(t,x) represents the distribution of order status at time t and spatial location x. is a function, u(t,x) is an external input variable;

[0114] In the order lifecycle management unit, the optimal order lifecycle is dynamically planned from order creation to delivery. The system needs to determine the optimal path among multiple operations to complete the order in the shortest time or at the lowest cost. The formula is as follows:

[0115]

[0116] Among them, V t (s t ) means that at time t, the order is in state s t The maximum value under a t is the decision made by the system at time t, R(s t ,a t ) is the order in state s t Execute decision a t The generated immediate return, γ is the discount factor.

[0117] The partial differential equation describes the changes in order status in time and space in real time. Combined with external input variables, the system can dynamically update the digital twin of the order and provide a global view of the order status. In this way, the system can ensure that the order status is adjusted by changes in external conditions and ensure that the order is in the best state.

[0118] The dynamic programming algorithm calculates the status value and return of the order to help the system make the best decision at key points in the order processing process. In terms of selecting the delivery route, inventory transfer and other important operations, the system can complete the order processing in the shortest time or at the lowest cost.

[0119] The combination of the two enables the system to dynamically monitor changes in order status and to intelligently manage orders based on optimal lifecycle planning. The integration solution ensures that the order management system is always in the optimal state, and realizes intelligent and automated optimization in order creation, processing, logistics scheduling, inventory allocation and other links, thereby improving the system's processing efficiency and responsiveness.

[0120] The inventory synchronization algorithm based on the distributed consistency protocol of the real-time inventory synchronization unit is to synchronize the inventory status between multiple nodes in real time. The inventory synchronization mechanism based on the distributed consistency protocol is described by the following formula:

[0121]

[0122] in, represents the inventory at node i at time t, represents the inventory increment transferred from node j to node i, represents the inventory quantity sent from node i to node j, It is the set of nodes participating in inventory synchronization;

[0123] Inventory Optimization Unit Inventory Optimization Algorithm Based on Linear Programming The goal of inventory optimization is to achieve the optimal inventory level. The optimization problem can be described using linear programming. The goal is to minimize the total cost, including inventory holding cost and shortage cost. The formula is as follows:

[0124]

[0125] Where Z is the total cost, h i is the inventory holding cost of the ith product, I i is the inventory of the ith product, s i is the shortage cost of the ith commodity, S i is the out-of-stock quantity of the ith product;

[0126] Constraints:

[0127] Among them, D i is the demand, T j It is the replenishment quantity from other warehouses or suppliers.

[0128] Through the inventory synchronization algorithm of the distributed consistency protocol, the system can ensure that inventory data is synchronized and consistent in real time between multiple nodes or storage locations. Multiple sales channels and storage nodes can share accurate inventory information to avoid inventory transfer problems caused by inconsistent data. It is used in inventory management scenarios across multiple regions or channels to effectively solve the problem of isolated inventory information islands.

[0129] Based on linear programming, the system can optimize the balance between inventory holding and shortages, ensuring that while meeting market demand, inventory holding costs and shortage costs are minimized. Combined with inventory synchronization, enterprises can ensure real-time understanding of the inventory status of each node, dynamically adjust inventory levels based on the optimal inventory allocation plan, and optimize resource utilization;

[0130] The distributed inventory synchronization algorithm ensures the timeliness and accuracy of inventory information, and linear programming further optimizes the configuration on this basis. Through the combination of algorithms, the system can realize intelligent inventory allocation and resource optimization among multiple nodes, thereby improving inventory turnover, reducing costs and enhancing the responsiveness of the supply chain.

[0131] Inventory transfer unit Inventory transfer algorithm based on network flow model Inventory transfer can be modeled as a network flow problem. The purpose is to transfer inventory between different warehouses to meet order requirements and minimize transfer costs. The formula is as follows:

[0132] Among them, C is the total cost of the transfer, c ij is the cost of transferring a unit of inventory from warehouse i to warehouse j, x ijis the inventory quantity transferred from warehouse i to warehouse j;

[0133] Constraints:

[0134] Among them, D j is the demand of warehouse j;

[0135] Capacity constraint: 0≤x ij ≤u ij ,

[0136] Among them, u ij is the maximum capacity that warehouse i can transfer to warehouse j;

[0137] The inventory visualization unit uses a heat map-based inventory distribution visualization algorithm to visualize the distribution of inventory in the system. The inventory distribution data is converted into a heat map, and the probability density function of the inventory distribution is generated using a two-dimensional kernel density estimate. The formula is as follows:

[0138] in, is the density estimate of the inventory at the (x, y) position, which is used to visualize the inventory distribution. K is the kernel function. (x i ,y i ) is the coordinate position of the i-th warehouse, h x 、h y are the smoothing parameters in the x and y directions.

[0139] The network flow model helps the system find the optimal allocation path and strategy by minimizing the allocation cost between different warehouses. The algorithm ensures that the needs of each warehouse are met during the allocation process, while complying with capacity constraints to ensure the rationality and efficiency of inventory scheduling, which helps to reduce transportation and allocation costs, ensure the rational allocation of resources, and improve supply chain efficiency.

[0140] By visualizing inventory distribution in the form of heat maps, managers can clearly understand whether inventory is concentrated or sparse in a region. Two-dimensional kernel density estimation makes the density information of inventory distribution clearer, helping managers to promptly identify areas with excess or shortage inventory. The intuitive inventory distribution display helps to optimize inventory layout and improve the agility of the supply chain.

[0141] Combined with algorithms, the system can intelligently optimize inventory transfers based on demand and inventory capacity. It can help managers fully understand inventory distribution through visualization. Heat maps provide inventory distribution density information, making transfer decisions more data-supported. Managers can make quick decisions based on real-time inventory distribution and further optimize resource allocation in the supply chain.

[0142] The logistics status tracking unit is based on the logistics status update model of the blockchain. In order to ensure the real-time synchronization and non-tamperability of the logistics status between different nodes, a blockchain-based status update mechanism is usually adopted. Each logistics status change will be updated in the blockchain. The following hash chain formula is used to describe the logistics status tracking:

[0143] H(S t+1 )=H(H(S t )∥H(ΔS t )),

[0144] Among them, H(S t+1 ) is the state hash value of the logistics at time t+1, H(S t ) is the state hash value of the logistics at time t, ΔS t represents the change of logistics status at time t, ∥ represents serial operation;

[0145] Load balancing algorithm for edge node task allocation in edge computing processing unit In the edge computing environment, logistics data processing tasks need to be efficiently allocated among multiple edge nodes to ensure load balancing. A load balancing algorithm based on uniform distribution is adopted. The formula is as follows:

[0146]

[0147] Among them, L i is the workload assigned to the ith edge node, W i is the processing capacity of edge node i, T is the total task volume;

[0148] Edge node response time formula:

[0149] Among them, R i is the response time of edge node i, C i is the computational effort of the node, N i is the network delay of the node, B i is the bandwidth of the node.

[0150] Through the hash chain formula of blockchain, each update of logistics status will be recorded on the blockchain to ensure the transparency and immutability of logistics information, which means that the logistics status is highly reliable and transparent in the system, all status changes can be traced, and data can be prevented from being maliciously tampered with or lost;

[0151] In the process of logistics data processing, task allocation in the edge computing environment is the key. Through the load balancing algorithm based on uniform distribution, the system can dynamically allocate tasks according to the processing capacity of each edge node, ensure the workload balance of each node, improve the processing efficiency of logistics data, and avoid overload or excessive idleness of some nodes;

[0152] Combined with the response time formula of the edge node, the system can monitor the actual operating status of the node, evaluate the response speed through computing power, network delay and bandwidth, and dynamically optimize task allocation to improve the overall performance of the edge computing system and ensure the processing speed of logistics data and the real-time response capability of the system.

[0153] The logistics partner management unit is based on the game theory of logistics partner collaboration model. Multiple logistics partners need to work together to complete complex transportation tasks. The game theory model is used to optimize the cooperation strategy. Assuming that the interests and costs of logistics partners are related, the following Nash equilibrium model is used to describe it:

[0154]

[0155] Among them, U i is the utility function of partner i, R j is the income of partner j, C j (x j ) is the cost of partner j, which depends on its input x j ;

[0156] Partner allocation optimization formula:

[0157]

[0158] Among them, C i (x i ) is the transportation cost of partner i, T is the total task volume;

[0159] The logistics optimization unit is based on the shortest path logistics optimization of the Dijkstra algorithm. In order to optimize the logistics path, the classic Dijkstra shortest path algorithm is used to calculate the optimal transportation path from the logistics starting point to the end point. The algorithm is used to find the shortest path in a weighted graph. The formula is as follows: d(v)=min(d(u)+w(u,v)).

[0160] Where d(v) represents the shortest path distance from the starting point to node v, d(u) represents the shortest path distance from the starting point to the current node u, and w(u,v) represents the path weight from node u to node v.

[0161] Logistics cost optimization formula:

[0162] Among them, C total is the total cost of logistics transportation, d ij is the shortest path distance from node i to node j, V ij is the cargo transportation volume between nodes i and j.

[0163] Through the game theory model, the system can balance the interests and costs of different logistics partners and the best cooperation strategy of all parties in transportation tasks. The Nash equilibrium model ensures that partners obtain the best benefits under current conditions, and the partner allocation optimization formula minimizes transportation costs and realizes the reasonable allocation of logistics tasks, making multi-party collaboration more efficient and effectively improving the flexibility and cost-effectiveness of overall logistics operations.

[0164] Di jkstra algorithm is used to find the shortest path in a complex transportation network to ensure that goods can be delivered at the optimal cost and in the shortest time. At the same time, the logistics cost optimization formula combines the shortest path and cargo flow to help the system calculate the total transportation cost and propose an optimization plan, which can effectively reduce transportation time and cost in logistics operations and improve the overall efficiency of logistics.

[0165] The combination of the two enables the system to have highly intelligent decision-making capabilities when scheduling logistics tasks and selecting the best transportation routes. The game theory model provides reasonable benefit distribution and cost control for multiple partners, while the Dijkstra algorithm optimizes transportation routes. Through the two algorithms, the system can efficiently allocate resources in complex logistics networks, balance the partners' benefits and transportation tasks, and ensure the optimality of transportation routes.

[0166] The inventory forecasting unit is based on the Bayesian inventory demand forecasting model. The inventory forecast is dynamically adjusted by combining historical orders and current market demand through Bayesian updates. The Bayesian model uses current information as a priori probability and combines it with future inventory demand to calculate the posterior probability. The Bayesian formula is as follows:

[0167] Where P(D|I) is the posterior probability of demand D given information I, P(I|D) is the likelihood function of information I when demand D is known, P(D) is the prior probability of demand, and P(I) is the marginal probability of information;

[0168] The formula for calculating the expected value of inventory demand is: E[D] = ∫D·P(D|I)dD,

[0169] Among them, E[D] is the expected value of future inventory demand, which is used to guide inventory management;

[0170] The sales report generation unit is based on the K-Means model in the clustering algorithm. Sales data usually needs to be classified by customer, product category or time period to generate targeted sales reports. The formula is as follows:

[0171]

[0172] Among them, J is the objective function of clustering, C i is the i-th cluster, μi is the center of the ith cluster, x j Belongs to cluster C i data points;

[0173] Sales trend clustering formula:

[0174] Among them, C t is the sales trend clustering index for a certain period of time, S i is the sales volume for that period, is the mean of sales;

[0175] The intelligent scheduling unit optimizes the scheduling model based on genetic algorithms. In intelligent scheduling, the scheduling system needs to dynamically adjust the scheduling plan based on factors such as order priority, inventory location, and transportation capacity, and use genetic algorithms to optimize the scheduling strategy. The genetic algorithm optimization formula is:

[0176]

[0177] Among them, f(x) is the objective function of the scheduling, x i is a solution in the solution space, w i is the same as the scheduling scheme x i The relevant weights;

[0178] Fitness function in genetic algorithm:

[0179] Among them, F(x) is the fitness function of the scheduling scheme, and C(x) is the cost of the scheduling scheme;

[0180] Crossover and mutation operations:

[0181] Crossover: Generate new solutions by exchanging some of the gene sequences of the solutions:

[0182] x′=(x1[1:k],x2[k+1:n]),

[0183] Mutation: Change some genes of the solution to explore new solutions:

[0184] x i ′=x i +∈,

[0185] Through multiple generations of iterative optimization, the genetic algorithm will tend to the optimal scheduling solution.

[0186] The Bayesian inventory demand forecasting model generates more accurate demand forecasts based on the latest market information by dynamically adjusting inventory forecasts, helping companies optimize inventory levels and avoid excessive or insufficient inventory. The inventory demand expected value calculation formula provides strong data support for inventory management;

[0187] The K-Means clustering algorithm helps companies cluster sales data by customer, product category, time period, and other dimensions to generate more targeted sales reports. The sales trend clustering formula can detect fluctuations or trends in sales and help companies adjust sales strategies in a timely manner.

[0188] The genetic algorithm optimizes order scheduling through iteration, combining factors such as order priority, inventory location and transportation capacity to find the optimal scheduling plan. Crossover and mutation operations ensure the diversity of scheduling plans, enabling the system to achieve optimal resource scheduling in a complex order and logistics environment.

[0189] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A blockchain-based omni-channel order management system, characterized in that: Including order management and digital twin module, inventory management and cloud network collaboration module, payment and smart contract module, logistics and edge computing module, customer service and after-sales management module, authority management and security module, data analysis and prediction module, compliance and audit module; The order management and digital twin module creates a digital model for the order through digital twin technology and displays the changes in order status in real time; The inventory management and cloud-network collaboration module uses cloud-network integration technology to synchronize inventory information of various channels in real time and dynamically allocate resources; The payment and smart contract module automates the payment, settlement and clearing processes through blockchain smart contract technology, and automatically triggers payment operations according to different nodes of the order; The logistics and edge computing module uses edge computing to synchronize the processing of logistics data to the blockchain in real time, so as to track the logistics status in real time and optimize the delivery route; The customer service and after-sales management module manages the after-sales service process of returns, exchanges and repairs, records each link of after-sales service, and combines with digital twins, so that customers and enterprises can view the after-sales status of products in real time; The authority management and security module provides identity authentication and authority control based on blockchain technology, so that system users can access and operate the orders or data they are authorized to manage; The data analysis and forecasting module provides market demand forecasts, order trend analysis and inventory optimization suggestions through data analysis, machine learning models and historical order data; The compliance and audit module uses the immutability of blockchain to automatically record order, payment and inventory operation logs in the system to generate auditable records, making it easier for enterprises and auditing agencies to conduct compliance checks on the system.

2. According to claim 1, a blockchain-based omni-channel order management system is characterized in that: The order management and digital twin module includes an order creation unit, an order status tracking unit, a digital twin creation unit, and an order lifecycle management unit; the order creation unit generates orders from different channels; The order status tracking unit tracks each status of the order in real time and records the update of each status on the blockchain; The digital twin creation unit generates a digital twin for each order, displays the real-time status of the order in a visual manner, and performs simulation in combination with the prediction model; the order lifecycle management unit is used to manage the entire process from order creation to delivery; The inventory management and cloud network collaboration module includes a real-time inventory synchronization unit, an inventory optimization unit, an inventory transfer unit, and an inventory visualization unit; The real-time inventory synchronization unit synchronizes inventory data in real time between different sales channels and storage locations; the inventory optimization unit optimizes inventory distribution and scheduling through cloud network collaboration technology; The inventory transfer unit automatically allocates and transfers according to order requirements and inventory status; The inventory visualization unit provides a real-time visualization interface of inventory, showing the inventory status of each channel and each warehouse; The payment and smart contract module includes a payment processing unit, a smart contract execution unit, a clearing unit, and a payment status tracking unit; the payment processing unit integrates multiple payment methods; the smart contract execution unit automatically executes payment, settlement and refund operations; after the order is completed, the clearing unit automatically executes the settlement and clearing process and records the results on the blockchain; the payment status tracking unit tracks the various states of the payment process and synchronizes the payment progress to the order management module in real time.

3. According to the blockchain-based omni-channel order management system of claim 1, it is characterized in that: The logistics and edge computing module includes a logistics status tracking unit, an edge computing processing unit, a logistics partner management unit, and a logistics optimization unit; the logistics status tracking unit tracks each node of logistics distribution in real time and records changes in logistics status on the blockchain; the edge computing processing unit is used to process logistics data and optimize distribution routes; the logistics partner management unit is integrated with multiple logistics service providers to uniformly manage logistics partners and update logistics status; the logistics optimization unit dynamically adjusts the distribution route based on real-time traffic, order priority, and storage location; The data analysis and prediction module includes an order trend analysis unit, an inventory prediction unit, a sales report generation unit, and an intelligent scheduling unit; The order trend analysis unit analyzes historical order data and real-time order trends to generate sales forecasts to help enterprises optimize production and inventory; the inventory forecast unit predicts inventory demand based on order data, market trends and inventory status; the sales report generation unit automatically generates sales reports; the intelligent scheduling unit automatically adjusts inventory scheduling and logistics strategies using forecast data; The compliance and audit module includes a compliance record unit, an audit report generation unit, a compliance monitoring unit, and a tax and regulation management unit; The compliance recording unit automatically records logs of orders, payments and inventory operations; the audit report generation unit automatically generates a compliance audit report based on the recorded logs; the compliance monitoring unit monitors operations in the system in real time, detects potential non-compliant behaviors, and automatically issues alerts; the tax and regulatory management unit automatically generates tax records and reports.

4. According to claim 2, the omni-channel order management system based on blockchain is characterized in that: The order creation unit uses a multi-channel order priority algorithm. In the case of multi-channel orders, the system needs to dynamically determine the processing order based on the priorities of different channels and order attributes. The formula is as follows: P(i) = α1·C i +α2·V i +α3·T i , Among them, P(i) represents the priority of order i, C i represents the channel weight of the order, V i Indicates the value of the order, T i represents the time urgency of the order, α1, α2, and α3 are weighting coefficients; The order status tracking unit predicts and tracks the change of order status through the Markov chain prediction model to describe the transfer of order status at different time nodes. The formula is as follows: P(X t+1 =x j |X t =x i )=p ij , Among them, X t is the state of the order at time t, p ij Is the order by state x i Transfer to state x j probability.

5. The blockchain-based omni-channel order management system according to claim 2, characterized in that: In the digital twin creation unit, the partial differential equation digital twin updated by the dynamic twin needs to be dynamically updated in real time according to the order status and external factors. Assuming that the order status is a function of time and external factors, the partial differential equation is used to describe its changes: Among them, S(t,x) represents the distribution of order status at time t and spatial location x. is a function, u(t,x) is an external input variable; In the order lifecycle management unit, the optimal order lifecycle is dynamically planned from order creation to delivery. The system needs to determine the optimal path among multiple operations to complete the order in the shortest time or at the least cost. The formula is as follows: Among them, V t (s t ) means that at time t, the order is in state s t The maximum value under a t is the decision made by the system at time t, R(s t ,a t ) is the order in state s t Execute decision a t The generated immediate return, γ is the discount factor.

6. The blockchain-based omni-channel order management system according to claim 2, characterized in that: The inventory synchronization algorithm of the real-time inventory synchronization unit based on the distributed consistency protocol is to synchronize the inventory status between multiple nodes in real time, using the inventory synchronization mechanism based on the distributed consistency protocol, which is described by the following formula: in, represents the inventory at node i at time t, represents the inventory increment transferred from node j to node i, represents the inventory quantity sent from node i to node j, It is the set of nodes participating in inventory synchronization; The inventory optimization unit is based on the inventory optimization algorithm of linear programming. The goal of inventory optimization is to achieve the optimal inventory level. The optimization problem can be described using linear programming. The goal is to minimize the total cost, including inventory holding cost and shortage cost. The formula is as follows: Where Z is the total cost, h i is the inventory holding cost of the ith product, I i is the inventory of the ith product, s i is the shortage cost of the ith commodity, S i is the out-of-stock quantity of the ith product; Constraints: Among them, D i is the demand, T j It is the replenishment quantity from other warehouses or suppliers.

7. The blockchain-based omni-channel order management system according to claim 2, characterized in that: The inventory transfer unit is based on the inventory transfer algorithm of the network flow model. Inventory transfer can be modeled as a network flow problem. The purpose is to transfer inventory between different warehouses to meet order requirements and minimize transfer costs. The formula is as follows: Among them, C is the total cost of transfer, c ij is the cost of transferring a unit of inventory from warehouse i to warehouse j, x ij is the inventory quantity transferred from warehouse i to warehouse j; Constraints: Among them, D j is the demand of warehouse j; Capacity constraint: 0≤x ij ≤u ij , Among them, u ij is the maximum capacity that warehouse i can transfer to warehouse j; The inventory visualization unit uses a heat map-based inventory distribution visualization algorithm to visualize the distribution of inventory in the system. The inventory distribution data is converted into a heat map, and a two-dimensional kernel density estimation is used to generate a probability density function of the inventory distribution. The formula is as follows: in, is the density estimate of the inventory at the (x, y) position, which is used to visualize the inventory distribution. K is the kernel function. (x i ,y i ) is the coordinate position of the i-th warehouse, h x 、h y are the smoothing parameters in the x and y directions.

8. The blockchain-based omni-channel order management system according to claim 3 is characterized in that: The logistics status tracking unit is based on the logistics status update model of the blockchain. In order to ensure the real-time synchronization and non-tamperability of the logistics status between different nodes, a blockchain-based status update mechanism is usually adopted. Each logistics status change will be updated in the blockchain. The following hash chain formula is used to describe the logistics status tracking: H(S t+1 )=H(H(S t )∥H(ΔS t )), Among them, H(S t+1 ) is the state hash value of the logistics at time t+1, H(S t ) is the state hash value of the logistics at time t, ΔS t represents the change of logistics status at time t, ∥ represents serial operation; In the edge computing processing unit, the load balancing algorithm for edge node task allocation is used. In the edge computing environment, logistics data processing tasks need to be efficiently allocated among multiple edge nodes to ensure load balancing. A load balancing algorithm based on uniform distribution is adopted, and the formula is as follows: Among them, L i is the workload assigned to the ith edge node, W i is the processing capacity of edge node i, T is the total task volume; Edge node response time formula: Among them, R i is the response time of edge node i, C i is the computational effort of the node, N i is the network delay of the node, B i is the bandwidth of the node.

9. The blockchain-based omni-channel order management system according to claim 3, characterized in that: The logistics partner management unit is based on the logistics partner collaboration model of game theory. Multiple logistics partners need to work together to complete complex transportation tasks. The game theory model is used to optimize the cooperation strategy. It is assumed that the interests and costs of logistics partners are related. The following Nash equilibrium model is used to describe it: Among them, U i is the utility function of partner i, R j is the income of partner j, C j (x j ) is the cost of partner j, which depends on its input x j ; Partner allocation optimization formula: Among them, C i (x i ) is the transportation cost of partner i, T is the total task volume; The logistics optimization unit is based on the shortest path logistics optimization of the Dijkstra algorithm. In order to optimize the logistics path, the classic Dijkstra shortest path algorithm is used to calculate the optimal transportation path from the logistics starting point to the end point. The algorithm is used to find the shortest path in a weighted graph. The formula is as follows: d(v)=min(d(u)+w(u,v)). Where d(v) represents the shortest path distance from the starting point to node v, d(u) represents the shortest path distance from the starting point to the current node u, and w(u,v) represents the path weight from node u to node v. Logistics cost optimization formula: Among them, C total is the total cost of logistics transportation, d ij is the shortest path distance from node i to node j, V ij is the cargo transportation volume between nodes i and j.

10. The blockchain-based omni-channel order management system according to claim 3, characterized in that: The inventory forecasting unit is based on the Bayesian inventory demand forecasting model. The inventory forecast is dynamically adjusted by combining historical orders and current market demand through Bayesian updating. The Bayesian model calculates the posterior probability by taking the current information as the prior probability and combining it with the future inventory demand. The Bayesian formula is as follows: Where P(D|I) is the posterior probability of demand D given information I, P(I|D) is the likelihood function of information I when demand D is known, P(D) is the prior probability of demand, and P(I) is the marginal probability of information; The formula for calculating the expected value of inventory demand is: E[D] = ∫D·P(D|I)dD, Among them, E[D] is the expected value of future inventory demand, which is used to guide inventory management; The sales report generation unit is based on the K-Means model in the clustering algorithm. Sales data usually needs to be classified by customer, product category or time period to generate targeted sales reports. The formula is as follows: Among them, J is the objective function of clustering, C i is the i-th cluster, μ i is the center of the ith cluster, x j Belongs to cluster C i data points; Sales trend clustering formula: Among them, C t is the sales trend clustering index for a certain period of time, S i is the sales volume for that period, is the mean of sales; The intelligent scheduling unit is based on the genetic algorithm to optimize the scheduling model. In intelligent scheduling, the scheduling system needs to dynamically adjust the scheduling plan according to factors such as order priority, inventory location, and transportation capacity, and use genetic algorithms to optimize the scheduling strategy. The genetic algorithm optimization formula is: Among them, f(x) is the objective function of the scheduling, x i is a solution in the solution space, w i is the same as the scheduling scheme x i The relevant weights; Fitness function in genetic algorithm: Among them, F(x) is the fitness function of the scheduling scheme, and C(x) is the cost of the scheduling scheme; Crossover and mutation operations: Crossover: Generate new solutions by exchanging some of the gene sequences of the solutions: x′=(x1[1:k],x2[k+1:n]), Mutation: Change some genes of the solution to explore new solutions: x i ′=x i +∈, Through multiple generations of iterative optimization, the genetic algorithm will tend to the optimal scheduling solution.

Images