A visual quality monitoring method and system for the entire production order process based on QLA
By introducing the Liyapunov optimization algorithm in large-scale personalized customized production, dynamic scheduling and QLA evaluation of manufacturers' production resources is solved, and the problems of information sharing and cost trade-offs in the production link are improved, and production efficiency and cooperation benefits are improved.
Patent Information
- Application Number
- CN202210499467.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-09
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-05-09
AI Technical Summary
During the large-scale personalized customized production process, it is difficult to adapt and timely sharing information between production links in multiple dimensions, resulting in difficult to guarantee product quality (QLA) and it is difficult to achieve QLA exchange and cost trade-offs between buyers and manufacturers. .
The online scheduling algorithm based on Lyapunov optimization is adopted to monitor the manufacturer's production situation in real time, dynamic scheduling of manufacturing service resources is realized, and a QLA evaluation and feedback mechanism is established to promote effective communication and cost trade-offs between buyers and manufacturers.
It realizes the rational allocation of manufacturers' production resources and minimizes costs, meets the QLA needs of buyers, and improves the cooperation efficiency and total production benefits between buyers and manufacturers.
Smart Images

Figure CN114792167B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of online resource monitoring, and in particular to a method and system for visually monitoring the quality of the entire process of a production order based on QLA. Background Art
[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.
[0003] In recent years, the Industrial Internet of Things has become increasingly popular. Driven by the Internet of Things, cloud computing and other fields, it has achieved tremendous development and progress. The Industrial Internet of Things provides unprecedented timeliness, flexibility, precision and efficiency for industrial manufacturing processes, promoting social and economic benefits. In the Industrial Internet of Things, mass customization is an innovative customized service model that uses mass production to meet the unique needs of buyers. Its main advantages are low cost, high efficiency and high user satisfaction. In mass customization, the intelligent manufacturing system is mainly used to meet the different product customization needs of buyers, and timely interactive feedback is used to meet the QLA needs of buyers and the cost optimization of manufacturers.
[0004] The business process of large-scale personalized customization includes demand interaction, design and development, material procurement, production planning, flexible manufacturing, logistics distribution, product satisfaction evaluation and feedback, and after-sales service. In the various production links of large-scale customized production network collaborative manufacturing, the order requirements of product buyers, the production resources of products, the quality evaluation and feedback of products, the manufacturer's product production plan, product production quality, product production response and completion time, the number of production equipment on the production line and real-time status data information are all very important data in the whole process. By fully analyzing the above collected data, the digital level of the industrial production system can be improved as a whole, and the total social production benefits can be improved.
[0005] However, in actual production, there are information blind spots between various production links. It is difficult to achieve adaptive, multi-dimensional, timely and effective flow and sharing of data obtained from various production links throughout the product life cycle, and it is also impossible to effectively guarantee the quality of the product (QLA) (product completion time and product qualification rate). It is also difficult to effectively achieve timely and accurate product satisfaction communication and two-way selection between buyers and manufacturers.
[0006] Chinese invention patent CN112862312A - Dynamic scheduling method and system for manufacturing service resources based on random online algorithm, although it can also realize the dynamic scheduling of manufacturing service resources, it does not take into account the impact of QLA in the production process, nor does it realize the trade-off between QLA and cost for manufacturers and the interaction between buyers and manufacturers in terms of QLA. Summary of the invention
[0007] In order to address the deficiencies of the prior art, the present invention provides a method and system for visually monitoring the quality of the entire production process of production orders based on QLA; based on an online scheduling algorithm optimized by Lyapunov, dynamic scheduling of manufacturing service resources is achieved by real-time monitoring of the actual production situation of each manufacturer.
[0008] In a first aspect, the present invention provides a method for visually monitoring the quality of the entire production order process based on QLA;
[0009] The QLA-based visual quality monitoring method for the entire production order process is applied to the server side, including:
[0010] The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest ranking;
[0011] After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server;
[0012] The buyer client conducts a quality assessment on the product data uploaded by the manufacturer client and feeds back the assessment results to the manufacturer client through the server.
[0013] In a second aspect, the present invention provides a visual monitoring system for the quality of the entire production order process based on QLA;
[0014] The whole process quality visual monitoring system of production orders based on QLA includes: server side, manufacturer client side and buyer client side;
[0015] The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest ranking;
[0016] After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server;
[0017] The buyer client conducts a quality assessment on the product data uploaded by the manufacturer client and feeds back the assessment results to the manufacturer client through the server.
[0018] Compared with the prior art, the present invention has the following beneficial effects:
[0019] In order to effectively reduce the production costs of each manufacturer in the process of large-scale network collaborative manufacturing, the present invention introduces an online scheduling algorithm based on Lyapunov optimization, realizes dynamic scheduling of manufacturing service resources by real-time monitoring of the actual production situation of each manufacturer, and makes reasonable judgments on whether effective connections need to be established between buyers and manufacturers on the server side. For manufacturers that have established contact with buyers, they can use the scheduling algorithm based on Lyapunov optimization to achieve a trade-off between product production costs and providing QLA for buyers, and return the products to buyers after each production is completed; buyers evaluate the manufacturer's QLA, identify the manufacturer's QLA level, and then feedback to the manufacturer to promote the manufacturer's improvement of its production quality.
[0020] Based on the panoramic data model of the data center, with orders as the core, the data of products that need to be produced is extracted, and visual management and control services and applications for production orders are established. Through the full-process tracking of production order data and product parameters, the purchaser's customization needs for product parameters and the control needs of the production process are met. In addition, the system can realize QLA communication and two-way selection between purchasers and manufacturers, thereby meeting the purchaser's demand for QLA, the manufacturer's trade-off between cost and QLA, and ultimately achieving a win-win situation for purchasers and manufacturers, and improving the total social production benefits.
[0021] During the operation of the present invention, from the perspective of resource allocation, workload, etc. of each manufacturer, the production status and QLA data of each manufacturer can be monitored in real time. According to the actual production situation of each manufacturer, by introducing an online scheduling algorithm based on Lyapunov optimization, the reasonable allocation of production tasks of manufacturers and the dynamic scheduling of service resources can be effectively achieved, the average production cost of manufacturers can be minimized, the QLA requirements of buyers can be met, and effective judgment can be made on whether a cooperative relationship should be established between manufacturers and buyers.
[0022] By setting up a resource pool, it is possible to avoid repeatedly collecting the production target and the production resources from the purchaser client, which can reduce the CPU operating pressure of the purchaser client and improve the CPU utilization and system throughput.
[0023] By traversing the production status data submitted by all manufacturer clients corresponding to each purchaser and storing the production status data of each manufacturer in a queue, a complete traversal of all manufacturer client data can be achieved to avoid data omission.
[0024] Through the manufacturer's feedback on QLA and the buyer's evaluation of the QLA, the manufacturer can achieve a trade-off between production cost and QLA optimization, and can improve its algorithm parameters based on the corresponding buyer feedback to enhance its competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0026] Figure 1 is a flow chart of the method of embodiment 1;
[0027] Figure 2 This is a schematic diagram of the internal terminal connection of the system of Example 1. DETAILED DESCRIPTION
[0028] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0029] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0030] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0031] In this embodiment, all data is obtained in compliance with laws and regulations and based on the user's consent, and is used legally.
[0032] Terminology explanation:
[0033] QLA: Quality level agreement, which refers to the weighted value of the average completion time of the product and the product qualification rate:
[0034]
[0035] In QLA, w1 and w2 are weight parameters set by the manufacturer, which are set according to the manufacturer's goals. is the average waiting time for product completion, and p is the qualified rate of the product. Since the qualified rate of the product is related to the manufacturer's process, equipment, etc., the present invention does not involve the control of the qualified rate of the product, but only considers the average waiting time for product completion as a parameter of QLA.
[0036] Embodiment 1
[0037] This embodiment provides a visual quality monitoring method for the entire production order process based on QLA;
[0038] like Figure 1 As shown in the figure, the visual quality monitoring method for the whole process of production orders based on QLA is applied on the server side, including:
[0039] S101: The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest sorting;
[0040] S102: After receiving the production resources, the manufacturer client uses an optimized scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client through the server;
[0041] S103: The buyer client performs a quality assessment on the product data uploaded by the manufacturer client, and feeds back the assessment result to the manufacturer client through the server.
[0042] Furthermore, before the step S101, the following steps are also included:
[0043] S100-1: The server stores the serial number of the manufacturer client in a container; wherein the geographical location distance between the stored manufacturer client and the buyer client is within a set distance range;
[0044] S100-2: The server receives the production order data submitted by the purchaser client in real time, including: production product parameters, production targets and production quantities;
[0045] S100-3: The server stores the received production order data in its resource pool;
[0046] S100-4: The server pushes the production order message and receives the production status data submitted by the manufacturer client connected to it and willing to produce;
[0047] S100-5: The server traverses the production status data submitted by these manufacturer clients; and stores the production status data of all manufacturers into a queue.
[0048] Furthermore, a production order refers to the number of products of a certain type that need to be produced, as well as the product parameters of the product.
[0049] Further, S102: after receiving the production order, the manufacturer client adopts an optimization scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client; wherein the optimization scheduling algorithm adopts the Lyapunov optimization algorithm.
[0050] Furthermore, the method also includes: the manufacturer client receives new production information and feeds back the decision of whether to produce to the server, the manufacturer client divides the production time into time slots of equal length, and turns the newly arrived product production orders into a queue waiting, it starts to make production decisions in each time slot, and when production is completed, returns the completed products to the corresponding purchaser.
[0051] Further, S102: after receiving the production resources, the manufacturer client uses an optimized scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client through the server; specifically, it includes:
[0052] After receiving the production order, the manufacturer client adopts the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the purchaser client; the optimization scheduling algorithm observes the length of the product queue waiting for production in the manufacturer client at the beginning of each time slot, selects the number of devices to minimize the drift penalty formula in the Lyapunov optimization, and realizes the trade-off between QLA and cost by taking different values of the parameter V.
[0053] Further, S102: after receiving the production resources, the manufacturer client uses an optimized scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client through the server; specifically, it includes:
[0054] Assume that the length of the product queue waiting for production in the manufacturer at time slot t+1 is Q(t+1), and its value depends on the number of products arriving and processed in time slot t:
[0055] Q(t+1)=max[Q(t)-s(t),0]+a(t); (1)
[0056] Where s(t) is the number of products produced in time slot t, and a(t) is the number of new products to be produced received by the manufacturer in time slot t;
[0057] Assume that the manufacturer's production cost in time slot t is cost(t):
[0058] cost(t)=N(t)*C1+M(t)*C2; (2)
[0059] Where N(t) is the number of devices that the manufacturer already has, and M(t) is the number of temporary devices that the manufacturer wants to expand in time slot t, and its value range is 0≤M(t)≤Mmax , C1 is the cost of each device owned, C2 is the cost of a single device for temporary extension of the time slot, and generally C2>C1, so it costs more for manufacturers to use temporary equipment.
[0060] For temporarily extended equipment, it is released when the production of the occupied products is completed. Then the number of equipment owned by the manufacturer in time slot t is s(t):
[0061] s(t)=N(t)+M(t); (3)
[0062] Using the online scheduling algorithm based on Lyapunov optimization, first set the metric for the waiting queue length:
[0063] Set the Lyapunov function L(Q(t)) to:
[0064]
[0065] The smaller the Lyapunov function is, the smaller the waiting queue length is.
[0066] According to Little's Law, the average waiting queue length is proportional to the average waiting time of the product, so the shorter the average waiting queue length, the shorter the average waiting time of the product, which means an increase in QLA.
[0067] Then, the Lyapunov drift technique is used to calculate the change of the Lyapunov function in a time slot. The smaller the drift value, the shorter the queue length. Then, the conditional expected Lyapunov drift value Δ(Q(t)) is set as:
[0068]
[0069] For manufacturers, the arrival of production tasks is independent and identically distributed, so its expected value Set to:
[0070]
[0071] Then formula (5) becomes:
[0072]
[0073] The manufacturer minimizes the right side of the above formula in each time slot t to reduce the drift value as much as possible, thereby obtaining a smaller average waiting queue length.
[0074] However, for manufacturers, only considering QLA may make the cost unnecessarily large. In order to consider both QLA and cost, the optimization target adds the cost part, so it becomes to minimize the drift plus penalty expression of formula (8):
[0075]
[0076] Where V≥ is the adjustment parameter for minimizing the average waiting queue length and minimizing the average cost. and is finite, then there exists a constant B that satisfies the relation:
[0077]
[0078] Then expand the calculation:
[0079]
[0080] In order to minimize the right side of formula (10), the following terms in formula (10) are minimized:
[0081]
[0082] That is, in each time slot t, minimize the value of formula (12):
[0083] Value=Vcost(t)-Q(t)s(t); (12)
[0084] That is: Value=V(N(t)*C1+M(t)*C2)-Q(t)*(N(t)+M(t)); (13)
[0085] That is: Value = M(t)*(V*C2-Q(t))+N(t)*(V*C1-Q(t)); (14)
[0086] Where V is a regulation parameter greater than 0. Since N(t) is the number of devices available to the manufacturer in the time slot, it is considered as a known value at time slot t. V, Q(t), C1, and C2 are also known values at the beginning of time slot t. Then for formula (14), N(t)*(V*C1-Q(t)) is an uncontrollable known value;
[0087] When V*C2-Q(t)<0, in order to minimize the Value value, the maximum value of M(t) is M max , but the actual number of temporary devices to be expanded is related to the number of product backlogs in the current queue;
[0088] When V*C2-Q(t)≥0, the manufacturer will not expand the temporary equipment, but only use the equipment it owns for production, and try to keep the equipment in production state to avoid equipment idleness.
[0089] In addition, manufacturers adjust the size of parameter V according to their goals (optimal QLA or minimum cost) to achieve a trade-off between QLA and cost. The setting of the V value is related to their historical experience and system parameters. Each manufacturer sets the parameter V value based on the specific environment, which is generally related to the specific production environment and production goals.
[0090] If the V value set by the manufacturer is larger, it means that the manufacturer pays more attention to cost;
[0091] If the manufacturer sets a smaller V value, more products will be scheduled for production in each time slot, which means that QLA (average waiting time for products) is more important. the smaller).
[0092] Furthermore, the method further comprises:
[0093] S104: The buyer client continuously matches with the manufacturer client. If the match is satisfied, that is, a manufacturer with high QLA and willingness to produce is found, production resources are pushed; if not, all corresponding manufacturer clients are continuously re-matched until a manufacturer that can meet the requirements is found.
[0094] The step of pushing the production resources to one or more manufacturer clients ranked first specifically includes:
[0095] Add the current production resources to the transfer system and push the production resources through the transfer system.
[0096] Furthermore, the method further comprises:
[0097] S105: Constantly traverse the production resource list, manufacturer list and buyer list, delete invalid and pushed resources, and remove the cancelled manufacturer client number and invalid buyer client number from the server.
[0098] Embodiment 2
[0099] This embodiment provides a visual monitoring system for the quality of the entire production order process based on QLA;
[0100] like Figure 2 As shown in the figure, the visual quality monitoring system for the whole process of production orders based on QLA includes: server side, manufacturer client side and buyer client side;
[0101] The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest ranking;
[0102] After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server;
[0103] The buyer client conducts a quality assessment on the product data uploaded by the manufacturer client and feeds back the assessment results to the manufacturer client through the server.
[0104] Furthermore, after receiving the production order, the manufacturer client adopts the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client; the optimization scheduling algorithm adopts the Lyapunov optimization algorithm.
[0105] Furthermore, the method also includes: the manufacturer client receives new production information and feeds back the decision of whether to produce to the server, the manufacturer client divides the production time into time slots of equal length, and turns the newly arrived product production orders into a queue waiting, it starts to make production decisions in each time slot, and when production is completed, returns the completed products to the corresponding purchaser.
[0106] Furthermore, after receiving the production resources, the manufacturer client uses an optimized scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client through the server; specifically, it includes:
[0107] After receiving the production order, the manufacturer client adopts the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the purchaser client; the optimization scheduling algorithm observes the length of the product queue waiting for production in the manufacturer client at the beginning of each time slot, selects the number of devices to minimize the drift penalty formula in the Lyapunov optimization, and realizes the trade-off between QLA and cost by taking different values of the parameter V.
[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. The visual quality monitoring method for the entire production order process based on QLA is characterized by: Applied to the server side, including: The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest ranking; After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server; The buyer client conducts quality assessment on the product data uploaded by the manufacturer client and feeds back the assessment results to the manufacturer client through the server; After receiving the production resources, the manufacturer client uses an optimized scheduling algorithm to complete the production of the product, and uploads the completed product data to the buyer client through the server; specifically including: Assume that the length of the product queue waiting for production in the manufacturer at time slot t+1 is Q(t+1), and its value depends on the number of products arriving and processed in time slot t: Q(t+1)=max[Q(t)-s(t),0]+a(t); (1) Where s(t) is the number of products produced in time slot t, and a(t) is the number of new products to be produced received by the manufacturer in time slot t; Assume that the manufacturer's production cost in time slot t is cost(t): cost(t)=N(t)*C1+M(t)*C2; (2) Where N(t) is the number of devices that the manufacturer already has, and M(t) is the number of temporary devices that the manufacturer wants to expand in time slot t, and its value range is 0≤M(t)≤M max , C1 is the cost of each device owned, and C2 is the cost of a single device temporarily extended for that time slot; For temporarily extended equipment, it is released when the production of the occupied products is completed. Then the number of equipment owned by the manufacturer in time slot t is s(t): s(t)=N(t)+M(t); (3) Using the online scheduling algorithm based on Lyapunov optimization, first set the metric for the waiting queue length: Set the Lyapunov function L(Q(t)) to: The smaller the Lyapunov function, the smaller the waiting queue length; Then, the Lyapunov drift technique is used to calculate the change of the Lyapunov function in a time slot. The smaller the drift value, the shorter the queue length. Then, the conditional expected Lyapunov drift value Δ(Q(t)) is set as: For manufacturers, the arrival of production tasks is independent and identically distributed, so its expected value Set to: Then formula (5) becomes: The manufacturer minimizes the right side of the above formula in each time slot t to reduce the drift value as much as possible, thereby obtaining a smaller average waiting queue length; However, in order to consider both QLA and cost, the optimization objective adds the cost part, so it becomes to minimize the drift plus penalty expression of formula (8): Where V ≥ 0, which is the adjustment parameter for minimizing the average waiting queue length and minimizing the average cost; setting the parameter second-order moment and is finite, then there exists a constant B that satisfies the relation: Then expand the calculation: In order to minimize the right side of formula (10), the following terms in formula (10) are minimized: That is, in each time slot t, minimize the value of formula (12): Value=Vcost(t)-Q(t)s(t); (12) That is: Value=V(N(t)*C1+M(t)*C2)-Q(t)*(N(t)+M(t)); (13) That is: Value = M(t)*(V*C2-Q(t))+N(t)*(V*C1-Q(t)); (14) Where V is a regulation parameter greater than 0; because N(t) is the number of devices available to the manufacturer in the time slot, it is considered as a known value at time slot t, and V, Q(t), C1, and C2 are also known values at the beginning of time slot t. Then, for formula (14), N(t)*(V*C1-Q(t)) is an uncontrollable known value; When V*C2-Q(t)<0, in order to minimize the Value value, the maximum value of M(t) is M max , but the actual number of temporary devices to be expanded is related to the number of product backlogs in the current queue; When V*C2-Q(t)≥0, the manufacturer does not expand temporary equipment, but only uses the equipment it owns for production, and tries to keep the equipment in production status to avoid equipment idleness; In addition, manufacturers adjust the value of parameter V according to their goals to achieve a trade-off between QLA and cost. The setting of the V value is related to their historical experience and system parameters. Each manufacturer sets the value of parameter V according to the specific environment, which is generally related to the specific production environment and production goals. If the V value set by the manufacturer is larger, it means that the manufacturer pays more attention to cost; If the manufacturer sets a smaller V value, more products will be scheduled for production in each time slot, which means that QLA is more important; The QLA is the average waiting time for product completion.
2. The QLA-based visual quality monitoring method for the entire production order process as claimed in claim 1, characterized in that: The method also includes: the manufacturer client receives new production information and feeds back the decision of whether to produce to the server, the manufacturer client divides the production time into time slots of equal length, and turns the newly arrived product production orders into a queue waiting, it starts to make production decisions in each time slot, and when production is completed, returns the completed products to the corresponding purchaser.
3. The QLA-based visual quality monitoring method for the entire production order process as claimed in claim 1, characterized in that: The method further comprises: The buyer client continuously matches with the manufacturer client. If the match is satisfied, that is, a manufacturer with high QLA and willingness to produce is found, production resources will be pushed; if not, all corresponding manufacturer clients will be continuously re-matched until a manufacturer that can meet the requirements is found.
4. The QLA-based visual quality monitoring method for the entire production order process as claimed in claim 1, characterized in that: The method further comprises: Always traverse the production resource list, manufacturer list and buyer list, delete invalid and pushed resources, and remove the cancelled manufacturer client number and invalid buyer client number from the server.
5. A visual quality monitoring system for the entire production order process based on QLA, based on a visual quality monitoring method for the entire production order process based on QLA as claimed in any one of claims 1 to 4, characterized in that: include: Server side, manufacturer client and buyer client; The server traverses the production status data reported by different manufacturer clients, sorts the manufacturer clients in descending order according to QLA, and pushes the production resources and order data to one or more manufacturer clients with the highest ranking; After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server; The buyer client conducts a quality assessment on the product data uploaded by the manufacturer client and feeds back the assessment results to the manufacturer client through the server.
6. The QLA-based production order full-process quality visual monitoring system as claimed in claim 5, characterized in that: After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server; The optimization scheduling algorithm adopts Lyapunov optimization algorithm.
7. The QLA-based production order full-process quality visual monitoring system as claimed in claim 5, characterized in that: The method also includes: the manufacturer client receives new production information and feeds back the decision of whether to produce to the server, the manufacturer client divides the production time into time slots of equal length, and turns the newly arrived product production orders into a queue waiting, it starts to make production decisions in each time slot, and when production is completed, returns the completed products to the corresponding purchaser.
8. The QLA-based production order full-process quality visual monitoring system as claimed in claim 5, characterized in that: After receiving the production resources, the manufacturer client uses the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the buyer client through the server; Specifically include: After receiving the production order, the manufacturer client adopts the optimization scheduling algorithm to complete the production of the product and uploads the completed product data to the purchaser client; the optimization scheduling algorithm observes the length of the product queue waiting for production in the manufacturer client at the beginning of each time slot, selects the number of devices to minimize the drift penalty formula in the Lyapunov optimization, and realizes the trade-off between QLA and cost by taking different values of the parameter V.
