CN116976948A - Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise - Google Patents
Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise Download PDFInfo
- Publication number
- CN116976948A CN116976948A CN202310931061.9A CN202310931061A CN116976948A CN 116976948 A CN116976948 A CN 116976948A CN 202310931061 A CN202310931061 A CN 202310931061A CN 116976948 A CN116976948 A CN 116976948A
- Authority
- CN
- China
- Prior art keywords
- data
- value
- dynamic feedback
- enterprise
- manufacturing enterprise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 230
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000010586 diagram Methods 0.000 title claims abstract description 45
- 238000013528 artificial neural network Methods 0.000 claims abstract description 47
- 230000008569 process Effects 0.000 claims abstract description 39
- 238000007781 pre-processing Methods 0.000 claims abstract description 20
- 238000013507 mapping Methods 0.000 claims abstract description 5
- 238000007726 management method Methods 0.000 claims description 58
- 230000006399 behavior Effects 0.000 claims description 52
- 238000004064 recycling Methods 0.000 claims description 29
- 238000012545 processing Methods 0.000 claims description 24
- 239000002245 particle Substances 0.000 claims description 22
- 238000003860 storage Methods 0.000 claims description 19
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000011156 evaluation Methods 0.000 claims description 18
- 238000011084 recovery Methods 0.000 claims description 16
- 238000007405 data analysis Methods 0.000 claims description 12
- 238000013079 data visualisation Methods 0.000 claims description 11
- 230000000007 visual effect Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 9
- 238000011027 product recovery Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 7
- 238000013500 data storage Methods 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000012098 association analyses Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000013480 data collection Methods 0.000 claims description 5
- 238000013523 data management Methods 0.000 claims description 5
- 238000003032 molecular docking Methods 0.000 claims description 4
- 230000002441 reversible effect Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000010187 selection method Methods 0.000 claims description 3
- 239000002994 raw material Substances 0.000 abstract description 7
- 238000005457 optimization Methods 0.000 abstract description 5
- 239000000047 product Substances 0.000 description 99
- 238000009826 distribution Methods 0.000 description 11
- 230000008859 change Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 239000000463 material Substances 0.000 description 9
- 238000012384 transportation and delivery Methods 0.000 description 9
- 230000035945 sensitivity Effects 0.000 description 7
- 238000004140 cleaning Methods 0.000 description 5
- 230000001364 causal effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000013439 planning Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 244000309464 bull Species 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 210000003899 penis Anatomy 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- -1 inventory Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012367 process mapping Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Finance (AREA)
- Tourism & Hospitality (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Manufacturing & Machinery (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a method and a system for generating a dynamic feedback flow graph of a full value chain of a manufacturing enterprise, wherein the method comprises the steps of obtaining a value added data source from a database of an enterprise intelligent system; extracting data information from the value-added data source, and preprocessing the data; defining the data information after data preprocessing to obtain abstract description of the data information; the data information of abstract description is configured in a dynamic feedback link of a full value chain of a manufacturing enterprise, matched with corresponding indexes, subjected to data association and predicted through a neural network; and performing value chain mapping on the value added data source to generate a dynamic feedback flow diagram of the full value chain of the manufacturing enterprise, and storing the dynamic feedback flow diagram in a database. The invention generates the dynamic feedback flow diagram of the full value chain of the manufacturing enterprise, can reasonably plan the supply/production time and period, improves the horizontal efficiency of the supply chain, reduces the raw material redundancy, realizes the stock optimization, reduces the production cost, and also provides a new solution idea for the business process link optimization of the enterprise.
Description
Technical Field
The invention belongs to the technical field of intelligent management of full value chain links of manufacturing enterprises, and particularly relates to a method and a system for generating dynamic feedback flow diagrams of full value chains of manufacturing enterprises based on neural network prediction and inventory level control.
Background
With the rapid development of internet technology and the wide application of intelligent equipment, a novel supply chain mode taking 'digitalization and intelligent' as marks is rapidly emerging, and a lean supply chain is an organic component part of production activities of manufacturing enterprises, and coordinates and integrates all activities in the processes of planning, transportation, production, storage, distribution and the like by planning unified information and information sharing; in the networked sales age, customer demands are used as a starting point, and a supply chain network is used as a consumption support, so that the supply chain in the network mode has timeliness and quick response compared with the traditional supply chain.
Compared with the traditional supply chain structure, the network supply chain structure has a plurality of network members in the whole, so that the service awareness of the network members needs to be improved, and the coordination relationship among resources is improved; the sales enterprises can predict the behaviors of clients by means of big data in a production environment of a supply chain, analyze the client demands according to prediction results and perform corresponding production allocation, suppliers supply goods in time, and data acquisition is performed on factors influencing supply, production and sales energy in the process of each value chain link, so that the sales enterprises can be used for predicting demand orders; the information of each link is mutually available in the whole supply chain process, personalized service can be penetrated in the circulation process from part production and product production to sales, and meanwhile, the whole process of supply, production and sales activities can be ensured to run safely and stably.
Disclosure of Invention
The invention aims to provide a method and a system for generating a dynamic feedback flow graph of a full value chain of a manufacturing enterprise, aiming at a series of activities generated by enterprise supply, production, marketing, product recovery, service links and the like, the influence of bull penis effect on enterprise inventory level fluctuation of each link is weakened through neural network prediction and inventory level adjustment, so that the supply, production time and period can be reasonably planned, the supply chain level efficiency is improved, the redundancy of raw materials is reduced, the inventory optimization is realized, the production cost is reduced, and a new solution idea is provided for enterprise business process link optimization.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the invention provides a method for generating a dynamic feedback flow diagram of a full value chain of a manufacturing enterprise, which comprises the following steps:
s1: collecting value-added data from a database of an enterprise intelligent system to form a value-added data source, and storing the data;
s2: extracting data information from the value-added data source, and preprocessing the extracted data information;
s3: defining the data information after data preprocessing to obtain abstract description of the data information;
S4: the data information of the abstract description is configured in a dynamic feedback link of the full value chain of the manufacturing enterprise, is matched with corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise, and is subjected to data association, so that prediction is performed through a neural network;
s5: and performing value chain mapping on the value added data source to generate a dynamic feedback flow diagram of the full value chain of the manufacturing enterprise, and storing the dynamic feedback flow diagram in a database.
Preferably, the enterprise intelligence system includes: customer management system, marketing service management system, enterprise resource management system, warehouse logistics management system, enterprise financial management system, production process control system, production process execution management system, product recovery management system, product after-sale service system, and product detection management system.
Preferably, the collecting the value-added data includes: various text data information, customer behavior information, mail information, customer webpage data information, communication data information, production process parameter information and product detection information are collected from each management system of the value added data through the data collection equipment.
Preferably, the data preprocessing includes: and carrying out data type division on the data in the value-added data source, comparing the data in each divided type with the data in the value-added data source, finding out the missing numerical value, and deleting the abnormal numerical value.
Preferably, in the step S3, defining the data information after the data preprocessing to obtain an abstract description of the data information includes:
respectively defining the data information after data preprocessing as: basic data of a supply production system related to supply, basic data of a production system related to production, basic data of a marketing service system related to marketing, basic data of a recycling service system related to recycling service;
and establishing a unified identifiable standardized mode of the data information in a plurality of systems, storing the supply index, the production index, the marketing index and the recovery service index in a key value form, distinguishing the data dimension by a key, storing the value of the value by the value, and realizing abstract description of the data information.
Preferably, in the step S4, the corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise include: a supply index related to supply, a production index related to production, a marketing index related to marketing, and a recycling service index related to recycling;
in the step S4, the method for configuring the data information of the abstract description in the dynamic feedback link of the full value chain of the manufacturing enterprise and matching with the corresponding index in the dynamic feedback link of the full value chain of the manufacturing enterprise includes:
Establishing a data management system in a database, indexing and storing the data information after abstract description in the database through corresponding indexes in a dynamic feedback link of the full-value chain of the manufacturing enterprise, and completing acquisition of dynamic feedback data of the full-value chain of the manufacturing enterprise;
in the step S4, the data association includes: performing association analysis on basic operation behaviors of clients and user demands, and then associating client behavior information in a value-added data source to construct an association characteristic data set;
in the step S4, the prediction performed by the neural network includes: predicting the purchase behavior of potential customers through a neural network based on the associated characteristic data set, comprehensively considering reverse prediction of order production and sales based on the historical sales data of retailers, the historical shipping data of manufacturers and the historical demand data of orders of the manufacturers and factors influencing sales, shipping or the demand data of orders in a supply chain stage, and simultaneously combining the inventory reconciliation rate of supply chain enterprises to jointly serve as a balance variable for selection, and determining the expected supply order quantity in a supply link, the expected production quantity in a production link and the expected order quantity in a sales link.
Preferably, the prediction of the purchasing behavior of the potential customer through the neural network comprises the following steps:
(1) Customer operation data preprocessing: after the value added data source is subjected to data domain processing, feature selection is further performed, the basic operation flow and different operation behaviors of a client in the client behavior information are associated with the client requirements in the client behavior information and analyzed, a feature data set is further constructed, and an optimal feature subset is selected by adopting a correlation coefficient method;
(2) Definition of fitness: adopting mean square error of predicted value of LSTM neural network as adaptive value of particlesfit, wherein ：
；
wherein ,yto be a true value of the value,is the desired output value;
(3) Taking the position information of the particles as parameters of an LSTM neural network, and constructing the LSTM neural network;
(4) Training LSTM neural networks: updating the individual extremum and the population extremum of the particles according to the adaptive value of each particle;
(5) Iteratively updating the speed and position information of the particles by adopting nonlinear inertia weight values according to the individual extremum and the population extremum of the particles;
(6) When the speed and position information of the particles meet the conditions or reach the maximum iteration number, entering the next step to obtain optimized parameters, otherwise, returning to the step (3);
(7) After the optimized parameters are obtained, the iteration times are increased, and the LSTM neural network is retrained to obtain a trained LSTM neural network;
(8) Based on the optimal feature subset, customer purchasing behavior is predicted through the trained LSTM neural network.
Preferably, the selection method of the optimal feature subset is as follows:
firstly, taking a pearson coefficient as a selection index of a feature subset, selecting a parameter a as a threshold value in the feature subset selection process, finding out feature pairs with absolute values of correlation smaller than the parameter a based on value-added data, respectively comparing the correlation of the features to the feature data set on the basis of the selected feature pairs, and removing the feature pairs with small correlation to obtain a new feature subset;
evaluating the generated feature subsets by adopting an evaluation function to obtain an evaluation result, comparing indexes evaluated by the feature subsets with the evaluation result, updating the evaluation result and the optimal feature subset if the indexes evaluated by the feature subsets are larger than the evaluation result, otherwise, stopping the algorithm, and outputting the current optimal feature subset;
optimizing LSTM neural network model parameters based on the optimal feature subset;
and predicting the demands of customers on different products at the same time interval based on corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise.
The invention also provides a dynamic feedback flow chart of the full value chain of the manufacturing type enterprise, which is generated and obtained by the method, wherein the dynamic feedback flow chart of the full value chain of the manufacturing type enterprise comprises the following components: dynamic feedback flow diagram visual editor and configuration page, dynamic feedback flow diagram variable type and connection model, database system and real-time interface.
The invention also provides a data sharing integrated management mode platform of the dynamic feedback flow graph of the full value chain of the manufacturing enterprise, which comprises the following steps:
the system comprises a supply chain enterprise business collaboration service platform, a supply chain enterprise business collaboration platform server, a manufacturing enterprise A supplier group, a supplier enterprise internal server, a manufacturing enterprise group, a manufacturing enterprise internal server, a manufacturing enterprise B dealer group and a recycling dealer group;
the supply chain enterprise business collaboration service platform is in communication connection with a supply chain enterprise business collaboration platform server;
the provider enterprise internal server is arranged inside the manufacturer enterprise A provider group;
the manufacturing enterprise internal server is arranged inside the manufacturing enterprise group;
the manufacturer a provider group may access a recycler group, the manufacturer B dealer group may access a recycler group and a recycler group, the recycler group and the manufacturer B dealer group may only access business data from the supply chain enterprise business collaboration service platform, and all requests for access have specific access rights.
The invention also provides an analysis method of the dynamic feedback flow graph of the full value chain of the manufacturing type enterprise, which is based on the dynamic feedback flow graph of the full value chain of the manufacturing type enterprise and comprises the following steps: data acquisition and classification, data processing, data query and storage, and data visualization and analysis;
the data acquisition means that direct data are obtained from each data system, are uniformly received and integrated into a value-added data source, and data information of different types are respectively input into a uniform data storage space;
the data processing refers to definition description of data obtained in the acquisition process;
the data query and storage means that the data after data processing is stored in a data space and an index table structure is established;
the data visualization and analysis means that the data after the dynamic feedback result transmission are imaged and displayed to form a dynamic feedback value chain flow diagram, then the data in a data space are reasonably selected to form a historical data warehouse in the supply chain process, and according to the direct data obtained in each link system and the historical data stored in the data, the data association is determined through the process index and the data docking and the function association, the data association is displayed, and then the data association is used for calculating the historical data, and the data association is used for analyzing the whole value chain process.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
the invention establishes data association based on the manufacturing enterprise full value chain dynamic feedback flow graph by establishing the operation behavior of the customer and the demand of the customer, predicts the purchase intention of the potential customer by training the LSTM neural network, takes the predicted result as the yield of the manufacturer in the next period and the order of the supplier in the next period, visualizes the dynamic data change in the full value chain business link, displays the key indexes such as the inventory level, market demand and the like of the supplier, the manufacturer, the distributor and the recoverer, knows the running condition of each operation activity in real time, and keeps the management of the inventory level of the supply chain; the dynamic feedback flow graph of the full-value chain of the manufacturing enterprise can flexibly reflect the resource flow and the value direction of each link of supply, production, marketing service and recovery service by selecting important indexes, monitor the business data of each business link and visually display the dynamic feedback flow graph of the full-value chain through the terminal. At any time, the dynamic feedback flow diagram can be utilized to analyze the value chain links, and the problem is targeted and found, so that management staff can conveniently adjust strategies of supply, production and marketing services.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
The method for generating the dynamic feedback flow diagram of the full value chain of the manufacturing enterprise is further described below with reference to the accompanying drawings;
FIG. 1 is a schematic structural diagram of a data sharing integrated management platform of a full value chain dynamic feedback flow diagram of a manufacturing enterprise;
FIG. 2 is a flow chart of a method for analyzing a data supply chain of a dynamic feedback flow diagram of a full value chain of a manufacturing enterprise according to the invention;
FIG. 3 is a schematic diagram of a method for generating a dynamic feedback flow diagram of a full value chain of a manufacturing enterprise;
FIG. 4 is a schematic diagram of a prediction flow of the full value chain dynamic feedback of the manufacturing enterprise based on a trained LSTM neural network;
FIG. 5 is a causal relationship diagram of a dynamic feedback flow graph of a full value chain of a manufacturing enterprise according to the present invention;
FIG. 6 is a schematic diagram of a shared integrated management platform application interface of a dynamic feedback flow graph of a full value chain of a manufacturing enterprise.
Detailed Description
The following describes in further detail the embodiments of the present invention with reference to the drawings and examples. The following examples are illustrative of the invention and are not intended to limit the scope of the invention.
FIG. 1 is a schematic diagram of a data sharing and integration management method of a dynamic feedback value chain of a manufacturing enterprise, which aims at a clustered alliance form of the manufacturing enterprise, wherein the enterprise can interact with a database of a platform through a data exchange program by an internal database of the enterprise, so as to communicate data with an inter-enterprise business collaborative service platform.
FIG. 2 is a flow chart of a method for analyzing a dynamic feedback data supply chain of a manufacturing enterprise full value chain, which comprises data acquisition, data query and storage, and data visualization and analysis.
As shown in fig. 3, the embodiment of the invention discloses a method for generating a dynamic feedback flow graph of a full value chain of a manufacturing enterprise, which comprises the following steps:
s1: collecting value-added data from a database of an enterprise intelligent system to form a value-added data source, and storing the data;
s2: extracting data information from the value-added data source, and preprocessing the extracted data information;
s3: defining the data information after data preprocessing to obtain abstract description of the data information;
S4: the data information of the abstract description is configured in a dynamic feedback link of the full value chain of the manufacturing enterprise, is matched with corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise, and is subjected to data association, so that prediction is performed through a neural network;
s5: and performing value chain mapping on the value added data source to generate a dynamic feedback flow diagram of the full value chain of the manufacturing enterprise, and storing the dynamic feedback flow diagram in a database.
In step S1, the enterprise intelligent system includes: customer management system, marketing service management system, enterprise resource management system, warehouse logistics management system, enterprise financial management system, production process control system, production process execution management system, product recovery management system, product after-sale service system, and product detection management system.
The collecting of the value added data comprises the following steps: various text data information, customer behavior information, mail information, customer webpage data information, communication data information, production process parameter information and product detection information are collected from each management system of the value added data through the data collection equipment.
The data preprocessing comprises the following steps: and carrying out data type division on the data in the value-added data source, comparing the data in each divided type with the data in the value-added data source, finding out the missing numerical value, and deleting the abnormal numerical value.
In the step S3, defining the data information after the data preprocessing to obtain an abstract description of the data information, including:
respectively defining the data information after data preprocessing as: basic data of a supply production system related to supply, basic data of a production system related to production, basic data of a marketing service system related to marketing, basic data of a recycling service system related to recycling service;
and establishing a unified identifiable standardized mode of the data information in a plurality of systems, storing the supply index, the production index, the marketing index and the recovery service index in a key value form, distinguishing the data dimension by a key, storing the value of the value by the value, and realizing abstract description of the data information.
In order to achieve the technical scheme, the data acquisition is an interface for acquiring data from the outside of the system and inputting the data into the system, and the data acquisition system integrates signals, sensors, exciters, data acquisition equipment and application software; the method comprises the steps of obtaining direct data from each data system in a network collaborative manufacturing platform, and integrating the direct data into a value-added data source storage medium using a plurality of information; the data sources were initially obtained from various systems and modules in the network operations front-end page, production plant, financial accounting, supply recovery:
The data obtained from the systems are uniformly received and integrated into a data source of the system, and the data source is used for driving the visual display of information data; for a standardized system, a built-in value-added data source and an event target source are provided, and the data in a set time period can be directly acquired by adapting. For non-standardized systems, support expansion is required in the form of specifications.
The storage of the data source needs to have definite identification setting; the identification comprises the supply capacity, the sales attribute, the material production period, the equipment utilization rate, the standard production takt, the equipment change time, the personnel, the shift working time, the equipment name and the number, the customer basic information, the distribution condition, the product market sales quantity, the market sales fluctuation, the customer purchasing power, the product discount sensitivity and the distribution duration tolerance; the supply capacity represents the output efficiency of raw material production and processing of the supplier enterprise; the sales attribute refers to options and dimensions of the commodity for consumers to select, such as color, package, size and other settings of the product;
the material production cycle refers to the time spent producing each material; the equipment utilization rate refers to the ratio of the service time of equipment to the theoretical production time; the standard takt refers to the theoretical rate of product processing production; the equipment change time refers to the time consumed by equipment itself when the equipment is switched from producing one part to producing other parts;
The personnel refers to the number of the personnel actually participated in the work in one period of duty; the shift working time refers to the production time of one factory period specified by an enterprise, namely the interval between two adjacent shifts, and the shift working time can be arranged according to actual conditions according to the common practice of a production line, namely between 8 and 12 hours;
the device name and number refer to the identifying information of the device; the customer basic information comprises records of basic information such as addresses, names, ages or sexes of customers; the delivery status refers to delivery information such as the time and place of the product;
the product marketing quantity refers to the total quantity of product network marketing and off-line marketing; market sales fluctuation refers to the influence of natural factors or other objective factors on product sales in a period of time; customer purchasing power refers to the purchase price that a customer can accept;
the product discount sensitivity refers to the attractive force of the discount strength of the enterprise commodity to the customer; the distribution duration tolerance refers to the acceptance degree of the time when the commodity is delivered to the hand of a customer;
the data processing is carried out, the acquired data are defined after being acquired from a data source, the data are subjected to operations such as cleaning, and the standardized format of the data is maintained;
Finally, abstracting data sources, and initially designating basic data of a supply production system related to supply, basic data of a production system related to production, basic data of a marketing service system related to marketing service and basic data of a recovery service system related to recovery service.
Describing supply indexes, production indexes, marketing service indexes, recovery service indexes and the like by establishing data specifications and standards, and configuring a data source in a dynamic feedback flow diagram of a full value chain of a manufacturing enterprise; and realizing a data-driven real-time dynamic feedback flow graph analysis scheme by utilizing the data sources. The data query and storage refers to the storage of data in a data storage medium as historical data for the retention of dynamic feedback processes on a time scale;
the data query refers to establishing an index table and storing processed data in a data medium. The data visualization and analysis means that initial data is visualized in the form of a chart and the like, then the data after the dynamic feedback result transmission is graphically displayed in a same window comparison visualization manner, and the change of index parameters of each link is dynamically analyzed in real time to form a dynamic feedback flow chart of a full value chain link.
As shown in fig. 5 and fig. 6, in order to optimize the above technical solution, the basic data of the supply production system related to supply in the present invention includes: purchasing (price polling, contract, order placement, bidding), vendor business scope, raw material basis information, inventory type, supply capacity, sales attributes, direct shipping costs, indirect shipping costs, quality data files, service levels; the basic data from the production system related to production includes: the type of equipment, start, operation, downtime, personnel count, efficiency, facilities, equipment, tools, materials, consumables, direct operation cost, indirect operation cost, production equipment yield, process conditions and quality detection data files.
The basic data from the marketing service system related to the marketing service includes marketing discounts, inventory, sales volume, market fluctuations, product price predictions, delivery, product performance, customer base information, shopping behavior, web product browsing, commodity evaluation sensitivity, customer purchasing power, product promotion sensitivity, delivery efficiency receptivity.
The basic data from the reclamation service system related to the reclamation service includes: market recycling level, product durability, product reworkable technology level, metal content ratio required for recycling product, inventory, product inspection and maintenance, disassembly, product recycling type, quantity, recycling cost, operation cost, and business qualification.
The abstract is to establish a unified identifiable standardized mode of data in a plurality of systems, store the supply index, the production index, the marketing service and the recovery index in the form of key values, distinguish data dimensions by keys, and store the value of the key values.
The supply index includes inventory level, expected production quantity, product production order, part production rate, and supply capacity.
The production indexes comprise production quantity, inventory quantity, delay delivery, product sales rate prediction, product delivery rate, expected productivity, actual production capacity and product qualification rate. Marketing service indicators include brand promotion, degree of collaboration compatibility, asset and liability, marketing ability, product services, customer services, reputation and reputation, and sense of social responsibility.
The operation flow of the customer on the product comprises the following steps: browse products, collect products, add products to shopping carts, submit orders, pay orders. Performing association analysis on basic operation behaviors of clients and client requirements, and then associating data of the client behaviors in a value-added data source to construct an association characteristic data set;
finally predicting the purchase behavior of the potential customer through the LSTM neural network; in the stage of a supply chain, comprehensively considering reverse prediction order production and sales based on historical sales data of retailers, shipping history data of manufacturers, order demand history data of manufacturers and other factors influencing sales, shipping or order demand data in the supply chain, and simultaneously determining expected supply order quantity in a supply link, expected production quantity of a production link and expected order quantity of a sales link by combining inventory adjustment rates of supply chain enterprises as selection balance variables;
When the influence of the change of the demand of the market clients in the production process on the time factor needs to be analyzed, calling a client-related characteristic data set from a data warehouse; mathematical models of the most relevant data of indexes such as extraction, supply, production, sales, product recovery and the like are used for analyzing and predicting the stock level of each link in the value chain, so that the smooth operation of the links of the supply chain is guided.
In the step S4, the corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise include: a supply index related to supply, a production index related to production, a marketing index related to marketing, and a recycling service index related to recycling;
in the step S4, the method for configuring the data information of the abstract description in the dynamic feedback link of the full value chain of the manufacturing enterprise and matching with the corresponding index in the dynamic feedback link of the full value chain of the manufacturing enterprise includes:
establishing a data management system in a database, indexing and storing the data information after abstract description in the database through corresponding indexes in a dynamic feedback link of the full-value chain of the manufacturing enterprise, and completing acquisition of dynamic feedback data of the full-value chain of the manufacturing enterprise;
In the step S4, the data association includes: performing association analysis on basic operation behaviors of clients and user demands, and then associating client behavior information in a value-added data source to construct an association characteristic data set;
in the step S4, the prediction performed by the neural network includes: predicting the purchase behavior of potential customers through a neural network based on the associated characteristic data set, comprehensively considering reverse prediction of order production and sales based on the historical sales data of retailers, the historical shipping data of manufacturers and the historical demand data of orders of the manufacturers and factors influencing sales, shipping or the demand data of orders in a supply chain stage, and simultaneously combining the inventory reconciliation rate of supply chain enterprises to jointly serve as a balance variable for selection, and determining the expected supply order quantity in a supply link, the expected production quantity in a production link and the expected order quantity in a sales link.
In step 4, the matching method is to establish a data management system, and according to a database, establish data storage and index by using the database, and complete the acquisition of the dynamic feedback data of the full value chain of the manufacturing enterprise. The operation flow of the marketing service terminal customer on the product comprises the following steps: browse products, collect products, add products to shopping carts, submit orders, pay orders. And constructing an associated characteristic data set by carrying out association analysis on the basic operation behaviors of the clients and the user demands and then associating the data of the client behaviors in the value-added data source. And finally predicting the purchasing behavior of the potential customer through the neural network.
As shown in fig. 2, the above technical solution is further optimized, and the analysis flow steps of the supplier supply order prediction based on the trained LSTM neural network to operate on the customer and the expected order quantity of the dealer, the expected productivity of the manufacturer, include:
step one: obtaining data; associated with the forecast are mainly product attribute information and some customer behavior data; the product attribute information includes: the selling date, the product brand, the selling season, the belonging classification, the price, the original price, the product possession quantity, the selling days and the like; the operation behaviors of clients are mainly divided into five types: browsing products, collecting products, adding the products into a shopping cart, submitting orders and paying the orders. Wherein, the collection operation is an important feature for demand prediction, and has strong correlation with the demand of the final product in general; all the operation behaviors are recorded in the digital persistence device to form a complete client operation behavior data set which can be used for predicting different product requirements; the operation data of the clients can record the clicking and browsing number of the products, the collection times, the times of adding shopping carts, the sharing times and the sales amount of the products.
Step two: selecting characteristics; analyzing the basic operation flow of the client and the relation between different operation behaviors and client demands in the flow, namely the main characteristics related to the demands are the operations executed by the client, and then analyzing the client behavior data through a data set to obtain the conversion relation between the different operation behaviors of the client and the client demands;
the conversion rate is the number of product sales/the execution times of each operation, and the linear relation between simple requirements and customer operation behaviors can be obtained through analysis of the conversion rate, and whether the customer operation behaviors are important characteristics affecting the requirements or not can be deduced, so that the basic characteristic selection process can be completed.
Step three: data processing; many missing values and unreasonable data exist in the original data, so that operations such as cleaning, noise reduction and the like are required to be performed on the data set. For the phenomenon of missing values of the data, the continuity of the data can be ensured by adopting a zero padding or adjacent time period average value taking mode, and meanwhile, the condition that a large number of data samples cannot be used is avoided. For unreasonable data, such as malicious forms, the sales volume is high but the click volume of a customer is high, shopping carts are added, and the collection volume is small; or a data type mismatch and data values outside of normal expected ranges. The data elimination mode is adopted for the phenomena.
Step four: constructing association features; the purpose of selecting the optimal feature subset is to remove redundant features irrelevant to a predicted target, reduce the dimension of a data set and improve the training efficiency of a model; and constructing more associated features by adopting a correlation method on the basis of the features of the original data set.
For example, combining the original features of the original dataset or using statistical methods, a new feature of value is constructed: for example, the purchasing operation can reflect sales of the product, can respectively count the maximum value, the minimum value and the standard deviation of the sales of the product, and can objectively reflect sales and discrete degree; a sliding window approach may be used, and for each product, the relevant features in the time ranges of 1, 3, 5, 7, 14 days before it may be calculated, and new features in different time ranges may be obtained by changing the window length and the sliding step size. The optimal feature subset is then selected, but the data is normalized before finding the optimal subset. Because normalization can solve the difference of the value ranges among different features, each feature is in the same order of magnitude, and the method is suitable for comprehensive comparison and evaluation; however, the maximum benefit of normalization is that the optimizing process of the optimal solution is obviously gentle, and the optimal solution is easier to converge; after normalization processing, the correlations between different features and the target variable can be calculated and analyzed, the correlations between different features can be analyzed, and redundant features can be removed. The method can adopt the Pearson correlation coefficient method to prepare for the subsequent model training.
Step five: an LSTM-based demand prediction model is selected and parameters of the network are optimized using an improved particle swarm algorithm.
Step six: model prediction. And predicting the requirements of different products with the same time interval by adopting the selected evaluation index.
In the value chain process prediction link, the sales volume prediction of the retail end is combined with historical sales data of a network platform and data such as click rate, browse rate, yield and the like of potential customers concerned to be developed to predict, and finally the sales volume prediction of the retail end is obtained; the enterprise establishes the upper limit of the expected stock according to the sales quantity forecast, sets the deadline, compares the difference between the expected stock and the actual stock with the forecast quantity for replenishment, determines the maximum quantity as the next-period ordering quantity, and performs the reservation to the upper level of the supply chain.
Predicting sales of manufacturers by combining historical data of shipping volumes, obtaining a value of expected inventory by the predicted data and setting an inventory duration period, comparing a difference between the expected inventory and an actual inventory with the expected inventory, and taking the maximum value as the expected production product quantity;
predicting sales of suppliers by combining historical production ordering data of manufacturers, obtaining a value of expected inventory by the predicted data and setting an inventory duration period, comparing a difference between the expected inventory and an actual inventory with the expected inventory, and taking the maximum value as the number of parts expected to be supplied by the suppliers;
The recycling of the recycling manufacturer comprises two parts, namely part recycling and product recycling; according to the sales rate of the sales market and the market change of the recovered products, setting the recovery ratio of the products in different periods, and supplementing the obtained recovered products into the inventory of enterprises for reproduction and utilization;
the marketing service part is changed from a service profit chain to a service value chain, and the main functions of the marketing service part are to promote the service to customers in terms of employee satisfaction, enterprise internal quality and customer satisfaction, develop potential customers, and convert the potential customers into customers who formally purchase products, thereby driving the product sales of retailers, the product production of enterprises and the supply of parts of suppliers.
As shown in fig. 4, predicting potential customer purchase behavior through a neural network includes the steps of:
(1) Customer operation data preprocessing: after the value added data source is subjected to data domain processing, feature selection is further performed, the basic operation flow and different operation behaviors of a client in the client behavior information are associated with the client requirements in the client behavior information and analyzed, a feature data set is further constructed, and an optimal feature subset is selected by adopting a correlation coefficient method;
(2) Definition of fitness: adopting mean square error of predicted value of LSTM neural network as adaptive value of particlesfit, wherein ：
；
wherein ,yto be a true value of the value,is the desired output value;
(3) Taking the position information of the particles as parameters of an LSTM neural network, and constructing the LSTM neural network;
(4) Training LSTM neural networks: updating the individual extremum and the population extremum of the particles according to the adaptive value of each particle;
(5) Iteratively updating the speed and position information of the particles by adopting nonlinear inertia weight values according to the individual extremum and the population extremum of the particles;
(6) When the speed and position information of the particles meet the conditions or reach the maximum iteration number, entering the next step to obtain optimized parameters, otherwise, returning to the step (3);
(7) After the optimized parameters are obtained, the iteration times are increased, and the LSTM neural network is retrained to obtain a trained LSTM neural network;
(8) Based on the optimal feature subset, customer purchasing behavior is predicted through the trained LSTM neural network.
Specifically, the selection method of the optimal feature subset includes:
firstly, taking a pearson coefficient as a selection index of a feature subset, selecting a parameter a as a threshold value in the feature subset selection process, finding out feature pairs with absolute values of correlation smaller than the parameter a based on value-added data, respectively comparing the correlation of the features to the feature data set on the basis of the selected feature pairs, and removing the feature pairs with small correlation to obtain a new feature subset;
Evaluating the generated feature subsets by adopting an evaluation function to obtain an evaluation result, comparing indexes evaluated by the feature subsets with the evaluation result, updating the evaluation result and the optimal feature subset if the indexes evaluated by the feature subsets are larger than the evaluation result, otherwise, stopping the algorithm, and outputting the current optimal feature subset;
optimizing LSTM neural network model parameters based on the optimal feature subset;
and predicting the demands of customers on different products at the same time interval based on corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise.
The LSTM neural network prediction link further comprises: manufacturer, vendor;
predicting the number of products expected to be produced by the manufacturer through the trained LSTM neural network: predicting through LSTM neural network by combining with historical data of shipping amount of the manufacturer, obtaining expected inventory quantity by predicting data and setting inventory duration period, comparing difference between expected inventory quantity and actual inventory quantity with expected inventory quantity, and taking maximum value as expected product quantity of the manufacturer;
Predicting the number of suppliers expected to supply via a trained LSTM neural network: the historical production ordering data of the vendor's sales chain combination manufacturer is predicted by the LSTM neural network, the expected stock quantity is obtained by predicting the data and setting the stock duration period, the difference between the expected stock quantity and the actual stock quantity is compared with the expected stock quantity, and the maximum value is used as the expected supply quantity of the vendor.
The invention also provides a method for generating the dynamic feedback flow diagram of the full value chain of the manufacturing type enterprise, which comprises the following steps: dynamic feedback flow diagram visual editor and configuration page, dynamic feedback flow diagram variable type and connection model, database system and real-time interface.
The visual editor and the configuration page of the full value chain dynamic feedback flow graph of the manufacturing enterprise are used for the professional analyst to perform self definition on the element indexes, page layout, connection mode, icon style and visual parameter display of the full value chain dynamic feedback. And the full value chain dynamic feedback flow graph data is combined with actual supply, production and marketing services, the data output by the dynamic feedback of the service link is recovered, the data in the actual link is associated with event indexes, and the corresponding process is bound with the function of the full value chain dynamic feedback flow graph. The full value chain dynamic feedback real-time interface is used for displaying the dynamic change condition of the index on the real-time interface and supporting the operation and the relevance analysis of the data.
The invention also provides a method for analyzing the dynamic feedback flow graph of the full value chain of the manufacturing type enterprise, which is generated and obtained by the method, and the method is based on the dynamic feedback flow graph of the full value chain of the manufacturing type enterprise and comprises the following steps: data acquisition and classification, data processing, data query and storage, and data visualization and analysis.
The data acquisition means that direct data are obtained from each data system, are uniformly received and integrated into a value-added data source, and data information of different types are respectively input into a uniform data storage space;
the data processing refers to definition description and standardization of data obtained in the acquisition process;
the data query and storage means that the data after data processing is stored in a data space and an index table structure is established;
the data visualization and analysis means that the data after the dynamic feedback result transmission is imaged and displayed to form a dynamic feedback value chain flow graph. And then rationally selecting the data in the data space to form a historical data warehouse in the supply chain process, determining the data association through the process index and the data docking and function association according to the direct data obtained in each link system and the historical data stored in the data, displaying the data association, and then calculating the historical data for analyzing the whole value chain process.
Specifically, the data acquisition is an interface for acquiring data from the outside of the system and inputting the data into the system, and the data acquisition system integrates signals, sensors, exciters, data acquisition equipment and application software. The data is obtained directly from each data system in the network collaborative manufacturing platform, and is integrated into a value-added data source storage medium using a plurality of information. The data sources are initially obtained from various systems and modules in the network operations front-end page, production plant, financial accounting, supply recovery. For example: customer management system, marketing service management system, enterprise resource management system, warehouse logistics management system, enterprise financial management system, production process control system, product recovery management system, after-sales service system, and product detection management system. The data obtained from these systems are received and integrated in a unified way into the data sources of the system according to the invention, which are used to drive the visual display of information data. For a standardized system, a built-in value-added data source and an event target source are provided, and the data in a set time period can be directly acquired by adapting. For non-standardized systems, support expansion is required in the form of specifications. The storage of the data source requires explicit identification settings. The identification comprises the supply capacity, the sales attribute, the material production period, the equipment utilization rate, the standard production takt, the equipment change time, the personnel, the shift working time, the equipment name and the number, the customer basic information, the distribution condition, the product market sales quantity, the market sales fluctuation, the customer purchasing power, the product discount sensitivity and the distribution duration tolerance; the supply capacity represents the output efficiency of raw material production and processing of the supplier enterprise; the sales attribute refers to options and dimensions of the commodity for consumers to select, such as color, package, size and other settings of the product; the material production cycle refers to the time spent producing each material; the equipment utilization rate refers to the ratio of the service time of equipment to the theoretical production time; the standard takt refers to the theoretical rate of product processing production; the equipment change time refers to the time consumed by equipment itself when the equipment is switched from producing one part to producing other parts; the personnel refers to the number of the personnel actually participated in the work in one period of duty; the shift working time refers to the production time of one factory period specified by an enterprise, namely the interval between two adjacent shifts, and the shift working time can be arranged according to actual conditions according to common practice of a production line, namely between 8 hours and 12 hours. The device name and number refer to the identifying information of the device; the customer basic information comprises records of basic information such as addresses, names, ages or sexes of customers; the delivery status refers to delivery information such as the time and place of the product; the product marketing quantity refers to the total quantity of product network marketing and off-line marketing; market sales fluctuation refers to the influence of natural factors or other objective factors on product sales in a period of time; customer purchasing power refers to the purchase price that a customer can accept; the product discount sensitivity refers to the attractive force of the discount strength of the enterprise commodity to the customer; the distribution duration tolerance refers to the acceptance degree of the time when the commodity is delivered to the hand of a customer;
And the data processing is carried out, the acquired data are defined after being acquired from a data source, the data are subjected to operations such as cleaning, and the standardized format of the data is maintained. Finally, abstracting data sources, and initially designating basic data of a supply production system related to supply, basic data of a production system related to production, basic data of a marketing service system related to marketing service and basic data of a recovery service system related to recovery service. By establishing data specifications and standards, supply indexes, production indexes, marketing service indexes, recovery service indexes and the like are described, and data sources are configured in a dynamic feedback flow diagram of a full value chain of a manufacturing enterprise. And realizing a data-driven real-time dynamic feedback flow graph analysis scheme by utilizing the data sources. The data query and storage refers to the storage of data in a data storage medium as historical data for the retention of dynamic feedback processes on a time scale; the data query refers to establishing an index table and storing processed data in a data medium. The data visualization and analysis means that initial data is visualized in the form of a chart and the like, then the data after the dynamic feedback result transmission is graphically displayed in a same window comparison visualization manner, and the change of index parameters of each link is dynamically analyzed in real time to form a dynamic feedback flow chart of a full value chain link. The data in the data storage medium is required to be reasonably selected to form a historical data warehouse in the whole value chain process, and according to the direct data obtained in each link system and the historical data stored in the data, the data association can be determined through the process index and the data docking and the function association, and then the historical data association can be used for calculating the historical data and analyzing the whole value chain process.
The invention also provides a system for realizing the analysis method, which comprises a data acquisition module, a data processing module, a data query and storage module and a data visualization and analysis module; the data acquisition module is used for acquiring data from each system and integrating the data into the data acquisition system, and is used for integrating and uploading the data from each acquisition system; the data processing module is used for cleaning and standardizing acquired data; the data query and storage module is used for storing the processed data in a database and establishing an index table; the data visualization and analysis module is used for visualizing the data of each link, visualizing the data in the form of a full value chain dynamic feedback flow graph, and being capable of using the data for process calculation analysis of the whole value chain.
As shown in FIG. 1, the business collaboration service platform for the supply chain enterprise comprises three parts, namely a supply collaboration platform subsystem, a production collaboration platform subsystem and a marketing service collaboration platform subsystem, which adopt a data sharing integrated management mode. Wherein the manufacturing/supply enterprise recycling group can access the supply cooperative platform subsystem and inventory management, part management and distribution management in the production cooperative platform subsystem, and cooperate with the replenishment of remanufactured products/parts of the manufacturing/supply enterprise; the system modules of the analysis method are embedded into the platform, so that data can be more conveniently called on the platform.
The invention also provides a data sharing integrated management mode platform of the dynamic feedback flow graph of the full value chain of the manufacturing enterprise, which comprises a supply chain enterprise business collaborative service platform, a supply chain enterprise business collaborative platform server, a manufacturing enterprise A supplier group, a supplier enterprise internal server, a manufacturing enterprise group, a manufacturing enterprise internal server, a manufacturing enterprise B dealer group and a recycling dealer group;
the supply chain enterprise business collaboration service platform is in communication connection with a supply chain enterprise business collaboration platform server;
the provider enterprise internal server is arranged inside the manufacturer enterprise A provider group;
the manufacturing enterprise internal server is arranged inside the manufacturing enterprise group;
the manufacturer A provider group can access the recoverer group, the manufacturer B dealer group and the recoverer group can access the manufacturer B dealer group, and the recoverer group and the manufacturer B dealer group can only access service data from the service collaboration service platform of the supply chain enterprise, and can only have specific access rights. The service collaboration service platform for the supply chain enterprise comprises three parts, namely a supply collaboration platform subsystem, a production collaboration platform subsystem and a marketing service collaboration platform subsystem. Wherein the manufacturing/supply enterprise recycling group has access to the supply co-platform subsystem and inventory management, parts management, distribution management in the production co-platform subsystem, in coordination with the replenishment of remanufactured products/components by the manufacturing/supply enterprise. The system modules of the analysis method are embedded into the platform, so that data can be more conveniently called on the platform.
The data analysis in the method for analyzing the dynamic feedback flow graph of the full value chain of the manufacturing enterprise comprises the steps of analyzing and learning the association in the data extraction information, determining frequent item sets by utilizing a data mining algorithm, and reasonably selecting and deleting historical data to form a data warehouse capable of reflecting the state history of the supply chain so as to perform data modeling. When it is desired to analyze the impact of changes in the demand of market clients on time factors during production, a client-associated feature dataset is invoked from a data warehouse. Mathematical models of the most relevant data of indexes such as extraction, supply, production, sales, product recovery and the like are used for analyzing and predicting the stock level of each link in the value chain, so that the smooth operation of the links of the supply chain is guided.
The invention also provides a display information editing unit which is used for generating information display, editing the process through the interactive interface, and displaying the full value chain dynamic feedback flow diagram of the manufacturing enterprise by releasing the display information.
The dynamic feedback flow diagram of the full value chain of the manufacturing enterprise takes an upstream supplier, a midstream manufacturer, a downstream distributor and a recovery service provider into consideration, can flexibly monitor the business data of each link under the full value chain, and displays the dynamic feedback flow diagram of the full value chain of the manufacturing enterprise through a visual terminal. Can be used to determine the order rate of the next periodic dealer, the expected production rate of the manufacturer, the order reservation rate of the supplier, reduce the planning uncertainty between enterprises, stabilize the inventory levels of the suppliers, manufacturers, dealers by adjusting the part production rate, part shipment rate, product production rate, product shipment rate, product sales rate, inventory adjustment rate, etc. By analyzing the operation behaviors of the clients in advance, the client groups with potential labor saving in the market are mined, and decision guarantee is provided for planning uncertainty among enterprises. And determining that no value-added process is generated in the product marketing process, and providing a maximized marketing value-added service mode for management personnel.
Examples
The method is used for realizing the dynamic feedback flow graph analysis of the full value chain of the manufacturing enterprise, and specifically comprises the following steps:
and 1, collecting information about supply, production, marketing service and recovery service links, and establishing sound data source storage. The data from the supply system related to the supply link comprises stock quantity in the current period of raw material purchase, main processing equipment and labels, quantity, product production period, daily productivity, types of main supplied products, forms of enterprise delivery, purchasing discount strength, supplier transportation time and distance, material specification of re-purchase, information of common specification including weight, material, size and the like, number of determining supply manufacturer, purchase unit price and the like; the data from the production system related to the manufacturing link comprises the working time parameters of the production equipment, the production process conditions and quality data, the direct operation cost and the indirect production cost, and the production data closely related to the time parameters, including the previous and subsequent relation of the production equipment, the number of the equipment and the reliability parameters of a workbench which are effectively operated, the production yield ratio, the qualified product ratio and the like; the data from the marketing system related to the marketing services link is closely related to the connection of the marketing effect and the selected marketing pattern. The method comprises the steps of product price prediction, product performance promotion, efficacy during distribution, sales volume, competitive products, basic information of clients, client relationship networks, client purchasing power, promotion sensitivity and tolerance of distribution duration; the data from the recycler system related to the recycling link includes enterprise recycling capability, recycling channel, recycled product remanufacturability, recycled inventory quantity, manufacturer and supplier buyback price, market recyclability fluctuation amount, etc.;
Step 2, collecting and transmitting the data, wherein the data comprise collecting input of hardware and interface communication between software data systems;
step 3, combining a customer management system, a marketing service system, an enterprise resource management system, a warehouse logistics management system, an enterprise financial management system, a production process control system, a production process execution management system, a product recycling management system, a product after-sales service system and a product detection management system to acquire dynamic data of different continuous time periods, and performing mutual matching in an actual link to serve as basic data in a data source;
and 4, selecting operation behavior data of a client, cleaning and standardizing the data, associating the characteristics, selecting an optimal characteristic subset through a correlation coefficient method, reducing the complexity of the data, and taking the data as a training set and a testing set of the LSTM. And obtaining optimal parameters through training to predict the purchasing behavior of the customer.
And 5, establishing a causal relation graph of dynamic feedback according to key links of the dynamic feedback of the full-value chain of the manufacturing enterprise to be concerned, establishing a dynamic feedback flow graph of the full-value of the manufacturing enterprise according to the mapping relation, loading the regular data in the steps 1 to 4 into the flow graph of the dynamic feedback of the full-value chain of the manufacturing enterprise, establishing a data management system, and displaying the data obtained by each system in real time in the form of the dynamic feedback flow graph after the key indexes are selected after the data obtained by each system are processed along with the progress of the business links.
The design of the method is a exposable mode after data collection, the inherent logic and implementation mode are integrated and intercommunicated by advanced modern Internet technology, the data are selected, screened, standardized and the like on the basis of a plurality of basic data, and related indexes are selected to establish the mutual association, so that a dynamic feedback logic flow diagram of an integral closed loop is formed, and visual analysis is carried out on the value links in the dynamic feedback logic flow diagram.
Such as for the number of records of customer behavior operations: the operation times of the customer behaviors are influenced by the quality of products, the appearance of the products, the functions of the products and the like, and the influences are also influenced by a series of factors such as enthusiasm of staff of enterprises, the excitation of staff by salary setting of the enterprises, the proficiency of staff skills, the importance of the staff on the quality of the products and the like; thus, a series of back elements can be concentrated and presented on one index, and the influence of the index on other important indexes is deeply mined when the index is changed; and simultaneously, a causal relation graph is established, the relation among all important indexes is clarified, a dynamic feedback flow graph is formed, and the dynamic feedback process analysis of the complete value chain link is realized.
In summary, the invention discloses a method for generating dynamic feedback flow graph analysis of a full value chain of a manufacturing enterprise. The method comprises the steps of collecting value-added data from an enterprise intelligent system; processing the obtained value-added data source, storing the processed value-added data source and carrying out association representation with the corresponding service link; selecting proper indexes to carry out mathematical modeling operation on the data source; performing process mapping on the value added data source to generate a dynamic feedback value chain flow graph; and carrying out output analysis on the generated dynamic feedback value chain flow diagram of the manufacturing enterprise, and storing the result into a historical database. The invention aims at a series of activities generated by enterprise supply/production/marketing/product recovery/service links and the like, and reduces the influence of bull penis effect on enterprise stock level fluctuation of each link through neural network prediction and stock level adjustment. The method not only can reasonably plan the supply/production time and period, improve the horizontal efficiency of a supply chain and reduce the redundancy of raw materials, but also can realize stock optimization and production cost, and also provides a new solution for optimizing the business process links of enterprises.
The design of the method is a exposable mode after data collection, the back of the internal logic and implementation mode utilizes the integration and intercommunication of advanced modern Internet technology, the data are subjected to selection, screening, standardization and other treatments on the basis of having a plurality of basic data, and the related indexes are selected to establish the mutual association, so that a dynamic feedback logic flow diagram of an integral closed loop is formed, and the visual analysis is carried out aiming at the value links in the dynamic feedback logic flow diagram.
Such as for the number of records of customer behavior operations: the number of customer behavior operations is affected by the quality of the product, the appearance of the product, the function of the product, etc., which are in turn affected by a series of factors such as the enthusiasm of the staff of the enterprise, the motivation of the staff by the salary setting of the enterprise, the proficiency of the staff skills, the importance of the staff on the quality of the product, etc. Thus, a series of back elements can be concentrated and presented on one index, and the influence of the index on other important indexes when the index changes is deeply mined. And simultaneously, a causal relation graph is established, the relation among all important indexes is clarified, a dynamic feedback flow graph is formed, and the dynamic feedback process analysis of the complete value chain link is realized.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The method for generating the dynamic feedback flow graph of the full value chain of the manufacturing enterprise is characterized by comprising the following steps of:
s1: collecting value-added data from a database of an enterprise intelligent system to form a value-added data source, and storing the data;
s2: extracting data information from the value-added data source, and preprocessing the extracted data information;
s3: defining the data information after data preprocessing to obtain abstract description of the data information;
s4: the data information of the abstract description is configured in a dynamic feedback link of the full value chain of the manufacturing enterprise, is matched with corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise, carries out data association, and predicts through a neural network;
s5: and performing value chain mapping on the value added data source to generate a dynamic feedback flow diagram of the full value chain of the manufacturing enterprise, and storing the dynamic feedback flow diagram in a database.
2. The method for generating a dynamic feedback flow graph of a manufacturing enterprise full value chain according to claim 1, wherein in step S1, the enterprise intelligent system comprises: customer management system, marketing service management system, enterprise resource management system, warehouse logistics management system, enterprise financial management system, production process control system, production process execution management system, product recovery management system, product after-sales service system, and product detection management system;
In the step S1, the collecting the value-added data includes: various text data information, customer behavior information, mail information, customer webpage data information, communication data information, production process parameter information and product detection information are collected from each management system of the value added data through the data collection equipment.
3. The method for generating a dynamic feedback flow graph of a manufacturing enterprise full value chain according to claim 1, wherein the data preprocessing comprises: and carrying out data type division on the data in the value-added data source, comparing the data in each divided type with the data in the value-added data source, finding out the missing numerical value, and deleting the abnormal numerical value.
4. The method for generating the dynamic feedback flow graph of the full value chain of the manufacturing enterprise according to claim 1, wherein in the step S3, the data information after the data preprocessing is defined to obtain the abstract description of the data information, and the method comprises the following steps:
respectively defining the data information after data preprocessing as: basic data of a supply production system related to supply, basic data of a production system related to production, basic data of a marketing service system related to marketing, basic data of a recycling service system related to recycling service;
And establishing a unified identifiable standardized mode of the data information in a plurality of systems, storing the supply index, the production index, the marketing index and the recovery service index in a key value form, distinguishing the data dimension by a key, storing the value of the value by the value, and realizing abstract description of the data information.
5. The method for generating the dynamic feedback flow graph of the full value chain of the manufacturing enterprise according to claim 1, wherein in the step S4, the corresponding index in the dynamic feedback link of the full value chain of the manufacturing enterprise comprises: a supply index related to supply, a production index related to production, a marketing index related to marketing, and a recycling service index related to recycling;
in the step S4, the method for configuring the data information of the abstract description in the dynamic feedback link of the full value chain of the manufacturing enterprise and matching with the corresponding index in the dynamic feedback link of the full value chain of the manufacturing enterprise includes:
establishing a data management system in a database, indexing and storing the data information after abstract description in the database through corresponding indexes in a dynamic feedback link of the full-value chain of the manufacturing enterprise, and completing acquisition of dynamic feedback data of the full-value chain of the manufacturing enterprise;
In the step S4, the data association includes: performing association analysis on basic operation behaviors of clients and user demands, and then associating client behavior information in a value-added data source to construct an association characteristic data set;
in the step S4, the prediction performed by the neural network includes: predicting the purchase behavior of potential customers through a neural network based on the associated characteristic data set, comprehensively considering reverse prediction of order production and sales based on the historical sales data of retailers, the historical shipping data of manufacturers and the historical demand data of orders of the manufacturers and factors influencing sales, shipping or the demand data of orders in a supply chain stage, and simultaneously combining the inventory reconciliation rate of supply chain enterprises to jointly serve as a balance variable for selection, and determining the expected supply order quantity in a supply link, the expected production quantity in a production link and the expected order quantity in a sales link.
6. The method for generating the dynamic feedback flow graph of the full value chain of the manufacturing enterprise of claim 5, wherein the predicting of the purchasing behavior of the potential customer through the neural network comprises the following steps:
(1) Customer operation data preprocessing: after the value added data source is subjected to data domain processing, feature selection is further performed, the basic operation flow and different operation behaviors of a client in the client behavior information are associated with the client requirements in the client behavior information and analyzed, a feature data set is further constructed, and an optimal feature subset is selected by adopting a correlation coefficient method;
(2) Definition of fitness: adopting mean square error of predicted value of LSTM neural network as adaptive value of particlesfit, wherein ：
；
wherein ,yto be a true value of the value,is the desired output value;
(3) Taking the position information of the particles as parameters of an LSTM neural network, and constructing the LSTM neural network;
(4) Training LSTM neural networks: updating the individual extremum and the population extremum of the particles according to the adaptive value of each particle;
(5) Iteratively updating the speed and position information of the particles by adopting nonlinear inertia weight values according to the individual extremum and the population extremum of the particles;
(6) When the speed and position information of the particles meet the conditions or reach the maximum iteration number, entering the next step to obtain optimized parameters, otherwise, returning to the step (3);
(7) After the optimized parameters are obtained, the iteration times are increased, and the LSTM neural network is retrained to obtain a trained LSTM neural network;
(8) Based on the optimal feature subset, customer purchasing behavior is predicted through the trained LSTM neural network.
7. The method for generating the dynamic feedback flow graph of the full value chain of the manufacturing enterprise of claim 6, wherein the selection method of the optimal feature subset is as follows:
firstly, taking a pearson coefficient as a selection index of a feature subset, selecting a parameter a as a threshold value in the feature subset selection process, finding out feature pairs with absolute values of correlation smaller than the parameter a based on value-added data, respectively comparing the correlation of the features to the feature data set on the basis of the selected feature pairs, and removing the feature pairs with small correlation to obtain a new feature subset;
Evaluating the generated feature subsets by adopting an evaluation function to obtain an evaluation result, comparing indexes evaluated by the feature subsets with the evaluation result, updating the evaluation result and the optimal feature subset if the indexes evaluated by the feature subsets are larger than the evaluation result, otherwise, stopping the algorithm, and outputting the current optimal feature subset;
optimizing LSTM neural network model parameters based on the optimal feature subset;
and predicting the demands of customers on different products at the same time interval based on corresponding indexes in the dynamic feedback link of the full value chain of the manufacturing enterprise.
8. A manufactured enterprise full value chain dynamic feedback flow graph generated by the method of any one of claims 1-7, wherein the manufactured enterprise full value chain dynamic feedback flow graph comprises: dynamic feedback flow diagram visual editor and configuration page, dynamic feedback flow diagram variable type and connection model, database system and real-time interface.
9. The utility model provides a data sharing integrated management mode platform of manufacturing type enterprise full value chain dynamic feedback flow diagram which characterized in that includes:
the system comprises a supply chain enterprise business collaboration service platform, a supply chain enterprise business collaboration platform server, a manufacturing enterprise A supplier group, a supplier enterprise internal server, a manufacturing enterprise group, a manufacturing enterprise internal server, a manufacturing enterprise B dealer group and a recycling dealer group;
The supply chain enterprise business collaboration service platform is in communication connection with a supply chain enterprise business collaboration platform server;
the provider enterprise internal server is arranged inside the manufacturer enterprise A provider group;
the manufacturing enterprise internal server is arranged inside the manufacturing enterprise group;
the manufacturer a provider group may access a recycler group, the manufacturer B dealer group may access a recycler group and a recycler group, the recycler group and the manufacturer B dealer group may only access business data from the supply chain enterprise business collaboration service platform, and all requests for access have specific access rights.
10. The analysis method of the dynamic feedback flow diagram of the full value chain of the manufacturing type enterprise is characterized by comprising the following steps of: data acquisition and classification, data processing, data query and storage, and data visualization and analysis;
the data acquisition means that direct data are obtained from each data system, are uniformly received and integrated into a value-added data source, and data information of different types are respectively input into a uniform data storage space;
The data processing refers to definition description of data obtained in the acquisition process;
the data query and storage means that the data after data processing is stored in a data space and an index table structure is established;
the data visualization and analysis means that the data after the dynamic feedback result transmission are imaged and displayed to form a dynamic feedback value chain flow diagram, then the data in a data space are reasonably selected to form a historical data warehouse in the supply chain process, and according to the direct data obtained in each link system and the historical data stored in the data, the data association is determined through the process index and the data docking and the function association, the data association is displayed, and then the data association is used for calculating the historical data, and the data association is used for analyzing the whole value chain process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310931061.9A CN116976948A (en) | 2023-07-27 | 2023-07-27 | Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310931061.9A CN116976948A (en) | 2023-07-27 | 2023-07-27 | Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116976948A true CN116976948A (en) | 2023-10-31 |
Family
ID=88479016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310931061.9A Pending CN116976948A (en) | 2023-07-27 | 2023-07-27 | Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116976948A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952567A (en) * | 2024-03-25 | 2024-04-30 | 四川多联实业有限公司 | Production management method and system based on MES intelligent manufacturing |
-
2023
- 2023-07-27 CN CN202310931061.9A patent/CN116976948A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952567A (en) * | 2024-03-25 | 2024-04-30 | 四川多联实业有限公司 | Production management method and system based on MES intelligent manufacturing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tan et al. | A business process intelligence system for enterprise process performance management | |
| Ni et al. | Extended QFD and data-mining-based methods for supplier selection in mass customization | |
| Altmann | A supply chain design approach considering environmentally sensitive customers: the case of a German manufacturing SME | |
| Wang et al. | Improving inventory effectiveness in RFID-enabled global supply chain with Grey forecasting model | |
| US20030236721A1 (en) | Dynamic cost accounting | |
| CN115081868B (en) | Business process-oriented ERP (Enterprise resource planning) and MES (manufacturing execution system) management business system pushing method | |
| Rubel | Increasing the Efficiency and Effectiveness of Inventory Management by Optimizing Supply Chain through Enterprise Resource Planning Technology | |
| Krmac | Intelligent value chain networks: business intelligence and other ICT tools and technologies in supply/demand chains | |
| Zhu | Optimization and Simulation for E‐Commerce Supply Chain in the Internet of Things Environment | |
| Zhang et al. | Big data-enabled intelligent synchronisation for the complex production logistics system under the opti-state control strategy | |
| Lin et al. | [Retracted] Supply Chain Management System for Automobile Manufacturing Enterprises Based on SAP | |
| CN116976948A (en) | Method and system for generating dynamic feedback flow diagram of full value chain of manufacturing enterprise | |
| CN111679814A (en) | Data-driven data center system | |
| Katircioglu et al. | Supply chain scenario modeler: A holistic executive decision support solution | |
| US20080027834A1 (en) | Systems and methods for inventory management | |
| Chan et al. | Knowledge-based simulation and analysis of supply chain performance | |
| Tian | An effective model for consumer need prediction using big data analytics | |
| Stefanovic et al. | Application of data mining for supply chain inventory forecasting | |
| Che | Pricing strategy and reserved capacity plan based on product life cycle and production function on LCD TV manufacturer | |
| Ding | The impact of supply chain management on a company’s operation and decision based on the multidimensional data analysis of upstream and downstream industry market states | |
| Li | Optimization Design of Short Life Cycle Product Logistics Supply Chain Scheme Based on Support Vector Machine | |
| Rolf et al. | A review on unsupervised learning algorithms and applications in supply chain management | |
| Riinawati | The Development of Information Technology and Its Influence on the Field of Management Accounting | |
| Sheth et al. | A proficient process for Systematic Inventory Management | |
| Shiau | A semi-automatic planning BOM generator based on VMI forecasting |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |