Large model production and operation management system and method of ERP (Enterprise Resource Planning) system

Abstract

The invention discloses a large model production and operation management system and method of an ERP system, and relates to the technical field of operation management. A production module collects processing data and dynamic data and synchronizes the data to the ERP system in real time; the financial module judges the matching between a budget and a plan by using the processing data and the dynamic data provided by the production module, and identifies problems in resource allocation; meanwhile, combining a random forest model to predict a cost fluctuation trend and a linear regression model to analyze the relevance between production efficiency and cost, and generating an optimized resource allocation and production scheduling scheme; the scheme is applied to a financial module and a production module, the financial module optimizes purchase budget and capital flow arrangement according to the adjusted resource allocation scheme, the production module readjusts task priority and a material use plan based on the optimized scheduling scheme, and all adjustment results are fed back to an ERP system in real time to form closed-loop management. And the production efficiency and the resource utilization rate are synchronously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of business management, and in particular to a large-model production and business management system and method of an ERP system. Background Art

[0002] In the production and operation management of large models, enterprises need to coordinate production processes and market demands at the same time. The isolation between financial data and production data has become an important bottleneck restricting the efficient management of enterprises. Financial data mainly focuses on budget preparation, cost control and cash flow management, reflecting the efficiency and economic benefits of resource utilization, while production data focuses on material requirements, equipment operating status and production plan execution, reflecting the actual dynamics of the production process. The separation of the two types of data seriously weakens the information flow between systems. The current large model management methods mostly use separate systems to manage financial and production information separately. There is a lack of real-time interaction and unified view between data. Financial decisions are difficult to reflect the actual situation on the production line in a timely manner, and production plans tend to ignore financial constraints, such as inventory costs and cash flow pressure, which can easily lead to resource waste or operational risks.

[0003] Traditional solutions to this fragmentation problem generally involve regular manual reporting, with companies assigning dedicated personnel to extract data from different systems and manually integrate financial information with production information. However, this method is inefficient and cannot meet the demand for real-time data in a dynamic environment. There is also a time lag problem, which results in the inability to quickly form a closed loop for feedback between production and finance, and misses the best time to allocate resources. This has a more serious impact on companies with diversified orders, high-frequency changes, and greater supply chain pressure. Therefore, there is an urgent need for a large-model production and operation management solution combined with an ERP system to solve such problems. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] The present invention provides a large-model production and operation management system and method of an ERP system to solve the problem that the current large-model management methods mostly use separate systems to manage financial and production information respectively, lack real-time interaction and unified views between data, financial decisions are difficult to reflect the actual situation of the production line in a timely manner, and production plans tend to ignore financial restrictions.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a large-scale production and operation management system of an ERP system, which includes:

[0008] The production module is used to collect processing data and collect dynamic data at the production site based on IoT devices and sensor technology. The production module combines static processing data with dynamic data to form production information, which is synchronized to the ERP system.

[0009] The production module is also used to receive the adjusted production schedule and material usage plan;

[0010] The financial module is used to receive processing data and dynamic data from the ERP system to determine whether the budget is consistent with the plan;

[0011] The financial module adjusts the procurement budget and capital flow arrangement according to the generated resource allocation adjustment plan, and feeds back the dynamically adjusted budget and capital allocation plan to the ERP system and production module;

[0012] The ERP system integrates the input and output of the financial module and the production module, and is used to update and store processing data, dynamic data, financial analysis results, resource allocation, and production scheduling adjustment plans.

[0013] As a preferred solution for the large-model production and operation management system of an ERP system described in the present invention, the processing data includes order requirements, material usage, production scheduling and preliminary resource allocation plan; the dynamic data includes equipment operating status, production progress, actual material consumption and inventory changes.

[0014] In a second aspect, the present invention provides a large-scale model production and operation management method of an ERP system, comprising:

[0015] Step S1, the production module collects processing data, including order requirements, material usage, production schedule and preliminary resource allocation plan;

[0016] Processing data is updated to the ERP system and synchronized with the financial module as the basis for preliminary financial analysis to determine whether the budget is consistent with the plan;

[0017] Step S2: At the production site, dynamic data is collected based on IoT devices and sensor technology, including equipment operation status, production progress, actual material consumption, and inventory changes. The dynamic data serves as a real-time supplement to the processing data, and together with the processing data collected by the production module, it constitutes complete production information, and the production information is updated to the ERP system.

[0018] Step S3: Based on the real-time production information in ERP, a random forest model is used to predict the cost fluctuation trend, and a linear regression model is used to analyze the correlation between production efficiency and cost; based on the cost fluctuation trend and correlation, an adjustment plan for resource allocation and production scheduling is determined;

[0019] Step S4, applying the adjustment plan generated in step S3 to the finance and production modules.

[0020] As a preferred solution of the large-scale model production and operation management method of the ERP system described in the present invention, the step of judging whether the budget is consistent with the plan is as follows:

[0021] Obtain processing data from the production module and synchronize it to the financial module;

[0022] Eliminate outliers, fill in missing values, calculate the actual cost of each production according to the process unit price and consumption, compare the actual cost with the budget plan item by item, and analyze the deviation;

[0023] Output deviation analysis results and mark abnormal data that exceeds expectations.

[0024] As a preferred solution of the large-scale model production and operation management method of the ERP system described in the present invention, the processing data collected by the production module together constitute complete production information, and the step of updating the production information to the ERP system is as follows:

[0025] Collect dynamic data and define the dynamic data set as:

[0026] D d ={d d1 ,d d2 ,…,d dm},

[0027] Among them, D d Represents a dynamic data set, d dj Indicates the jth dynamic data, j indicates the sequence number of the dynamic data, and its value range is 1≤j≤m, and m indicates the total number of dynamic data;

[0028] The acquisition function for collecting dynamic data is refined into a filter function based on weighted moving average, and the formula is:

[0029]

[0030] Among them, H(S) represents the dynamic data acquisition function, S represents the real-time monitoring data flow, and s t represents the real-time data at time t, w t represents the weight at time t, and T represents the length of the time window covered by the weight;

[0031] Perform processing data cleaning and define the processing data set as:

[0032] D p ={d p1 ,d p2 ,…,d pn},

[0033] Among them, D p represents the processing data set, d pk Indicates the kth processed data, k indicates the sequence number of the processed data, and its value range is 1≤k≤n, and n indicates the total number of processed data.

[0034] The processed data is cleaned and refined into a threshold function to remove outliers:

[0035] C(D p )={d pk ∣μ-3σ≤d pk ≤μ+3σ},

[0036] Among them, C(D p ) represents the processed data set after cleaning, μ represents the mean of the processed data, σ represents the standard deviation of the processed data, d pk Represents the kth processed data.

[0037] As a preferred solution of the large-scale model production and operation management method of an ERP system described in the present invention, the processing data collected by the production module together constitute complete production information, and the step of updating the production information to the ERP system also includes:

[0038] Fusion of dynamic data and processing data, dynamic data D d And the processing data after cleaning D c Fusion is performed by timestamp. The fusion function is refined into matching logic based on time windows. The fusion formula is:

[0039] F(D c ,D d )={(d c ,d d )∣|t c -t d |≤Δt},

[0040] Among them, F(D c ,D d ) represents the fused production information set, d c Represents the processing data after cleaning, d d Indicates dynamic data, t c and t d They represent the timestamps of the corresponding data, and Δt represents the allowed time window error;

[0041] The fused production information set F(D c ,D d ) is updated to the ERP system.

[0042] As a preferred solution of the large-model production and operation management method of the ERP system described in the present invention, the step of using the random forest model to predict the cost fluctuation trend based on the real-time production information in the ERP is as follows:

[0043] Extract real-time production information from the ERP system, including cost data and production efficiency data, and define the extracted data set as:

[0044]

[0045] Where D represents the extracted data set, c k represents the actual cost of the kth record, e k represents the production efficiency of the kth record, k represents the record number, and n represents the total number of data records;

[0046] The cost data in the data set D is divided into a training set and a test set according to the time series. The training goal of the random forest model is defined as predicting the cost fluctuation trend at future moments. The formula of the random forest prediction model is:

[0047]

[0048] in, represents the prediction cost of the k+1th record, f RF represents the random forest model, x k The feature vector representing the kth record, including the current cost c k , production efficiency k and external environment variables,

[0049] The loss function of model training is defined as the mean square error, and the formula is:

[0050]

[0051] Among them, L represents the loss function value, N represents the number of training samples, and c k and Represent the actual cost and predicted cost respectively.

[0052] As a preferred solution of the large-scale production and operation management method of an ERP system described in the present invention, the steps of using a linear regression model to analyze the correlation between production efficiency and cost; and determining the adjustment plan of resource allocation and production scheduling based on the cost fluctuation trend and the correlation are as follows:

[0053] Using linear regression model to analyze production efficiency k For cost c k The regression formula is:

[0054] c k=β0+β1e k +∈ k ,

[0055] Among them, β0 represents the intercept, β1 represents the regression coefficient of production efficiency on cost, ∈ k represents the regression residual,

[0056] The regression coefficient is calculated by the least squares method, and the calculation formula is:

[0057]

[0058] in, represents the mean value of production efficiency, represents the mean cost,

[0059] Combining the prediction results of the random forest model with the correlation of linear regression analysis, a resource allocation and production scheduling adjustment plan is generated. The resource allocation optimization objective function is:

[0060]

[0061] Among them, R represents the total cost of resource allocation, α i represents the unit cost of the i-th resource, r i represents the allocation amount of the i-th resource, and M represents the number of resource types;

[0062] The production schedule adjustment plan aims to improve efficiency, and the adjustment model is:

[0063]

[0064] Among them, P represents the total production efficiency, β j represents the weight of the j-th production task, p j represents the completion efficiency of the jth task, and T represents the total number of tasks.

[0065] As a preferred solution of the large-scale model production and operation management method of the ERP system described in the present invention, in which: in step S4, the financial module adjusts the procurement budget and capital flow arrangement according to the resource allocation adjustment plan:

[0066] The production module adjusts the production schedule and material usage plan according to the production schedule adjustment plan;

[0067] The applied adjustment plan continues to be fed back to the ERP system to further update resource allocation and production scheduling.

[0068] As a preferred solution of the large-scale model production and operation management method of the ERP system described in the present invention, the step of applying the adjustment plan generated in step S3 to the financial and production modules is as follows:

[0069] According to the resource allocation adjustment plan R, optimize the procurement budget set B p :

[0070] B p = {b p1 ,b p2 ,…,b pM},

[0071] Among them, B p represents the optimized procurement budget set, b pi represents the purchase budget of the i-th resource, i represents the sequence number of the resource type, M represents the total number of resource types,

[0072] The budget adjustment formula is:

[0073] b pi =α i ·r i +Δb pi ,

[0074] Among them, α i represents the unit cost of the i-th resource, r i represents the demand for the i-th resource, Δb pi Indicates budget revisions due to market fluctuations or supply chain changes;

[0075] According to the adjustment results of the procurement budget, reallocate the capital flow and define the capital flow set as follows:

[0076] F={f1,f2,…,f K},

[0077] Among them, F represents the capital flow set, f k represents the capital allocation of the kth stage, k represents the sequence number of the capital flow stage, and K represents the total number of stages.

[0078] The cash flow adjustment formula is:

[0079]

[0080] Among them, γ ki It represents the proportion of resources allocated to the i-th resource in the k-th stage;

[0081] According to the production schedule adjustment plan P, optimize the production plan set:

[0082] S p ={s p1 ,s p2 ,…,s pT},

[0083] Among them, S prepresents the optimized production schedule set, s pj represents the production plan of the jth task, j represents the task number, T represents the total number of tasks,

[0084] The production schedule adjustment formula is:

[0085]

[0086] Among them, β j represents the priority weight of the jth task, e j Indicates the execution efficiency of the current j-th task;

[0087] According to the task plan adjusted by the production schedule, optimize the material usage plan set:

[0088] M u ={m u1 ,m u2 ,…,m uT},

[0089] Among them, M u Represents the optimized material usage plan set, m uj represents the material requirement of the jth task, j represents the task number,

[0090] The material usage adjustment formula is:

[0091]

[0092] Among them, λ ji Indicates the usage ratio of the i-th resource required for the j-th task.

[0093] The beneficial effects of the present invention are as follows: the present invention significantly improves the collaborative efficiency of production and financial management.

[0094] In the present invention, the production module collects processing data and dynamic data, and combines IoT devices and sensor technology to obtain equipment operation status, production progress, actual material consumption and inventory change data in real time to form complete production information, and synchronizes it to the ERP system in real time; the financial module uses the processing data and dynamic data provided by the production module to determine the matching of the budget and the plan, and accurately identifies problems in resource allocation by eliminating outliers, filling missing values ​​and performing deviation analysis; at the same time, the random forest model is combined to predict the cost fluctuation trend and the linear regression model is combined to analyze the correlation between production efficiency and cost, and generate optimized resource allocation and production scheduling plans; these plans are applied to the financial module and the production module, the financial module optimizes the procurement budget and capital flow arrangement according to the adjusted resource allocation plan, and the production module readjusts the task priority and material use plan based on the optimized scheduling plan, and all adjustment results are fed back to the ERP system in real time to form a closed-loop management, so that production efficiency and resource utilization are improved simultaneously, and budget management is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0096] Figure 1 It is a structural schematic diagram of the large model production and operation management system of the present invention.

[0097] Figure 2 It is a schematic diagram of the process of the large model production, operation and management method of the present invention. DETAILED DESCRIPTION

[0098] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0099] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0100] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0101] Example 1, reference Figure 1 and Figure 2 This embodiment provides a large-scale production and operation management system of an ERP system, including:

[0102] The production module is used to collect processing data and collect dynamic data at the production site based on IoT devices and sensor technology. The production module combines static processing data with dynamic data to form production information, which is synchronized to the ERP system.

[0103] The production module is also used to receive the adjusted production schedule and material usage plan;

[0104] Processing data, including order requirements, material usage, production schedules, and preliminary resource allocation plans;

[0105] Dynamic data, including equipment operating status, production progress, actual material consumption and inventory changes;

[0106] The financial module is used to receive processing data and dynamic data from the ERP system to determine whether the budget is consistent with the plan;

[0107] The financial module adjusts the procurement budget and capital flow arrangement according to the generated resource allocation adjustment plan, and feeds back the dynamically adjusted budget and capital allocation plan to the ERP system and production module;

[0108] The ERP system integrates the input and output of the financial module and the production module, and is used to update and store processing data, dynamic data, financial analysis results, resource allocation, and production scheduling adjustment plans.

[0109] This embodiment also provides a large model production and operation management method of an ERP system, including:

[0110] Step S1, the production module collects processing data, including order requirements, material usage, production schedule and preliminary resource allocation plan;

[0111] Processing data is updated to the ERP system and synchronized with the financial module as the basis for preliminary financial analysis to determine whether the budget is consistent with the plan;

[0112] The steps to determine whether the budget is consistent with the plan are:

[0113] Obtain processing data from the production module and synchronize it to the financial module;

[0114] Eliminate outliers, fill in missing values, calculate the actual cost of each production according to the process unit price and consumption, compare the actual cost with the budget plan item by item, and analyze the deviation;

[0115] Output deviation analysis results and mark abnormal data that exceeds expectations;

[0116] Specifically, through cost calculation and budget comparison, deviation analysis results are generated for preliminary verification by staff.

[0117] Step S2: At the production site, dynamic data is collected based on IoT devices and sensor technology, including equipment operation status, production progress, actual material consumption, and inventory changes. The dynamic data serves as a real-time supplement to the processing data, and together with the processing data collected by the production module, it constitutes complete production information, and the production information is updated to the ERP system.

[0118] Together with the processing data collected by the production module, it constitutes complete production information, and the steps to update the production information to the ERP system are as follows:

[0119] Collect dynamic data and define the dynamic data set as:

[0120] D d ={d d1 ,d d2 ,…,d dm},

[0121] Among them, D d Represents a dynamic data set, d dj Indicates the jth dynamic data, j indicates the sequence number of the dynamic data, and its value range is 1≤j≤m, and m indicates the total number of dynamic data;

[0122] The acquisition function for collecting dynamic data is refined into a filter function based on weighted moving average, and the formula is:

[0123]

[0124] Among them, H(S) represents the dynamic data acquisition function, S represents the real-time monitoring data flow, and s t represents the real-time data at time t, w t represents the weight at time t, and T represents the length of the time window covered by the weight;

[0125] Perform processing data cleaning and define the processing data set as:

[0126] D p ={d p1 ,d p2 ,…,d pn},

[0127] Among them, D p represents the processing data set, d pk Indicates the kth processed data, k indicates the sequence number of the processed data, and its value range is 1≤k≤n, and n indicates the total number of processed data.

[0128] The processed data is cleaned and refined into a threshold function to remove outliers:

[0129] C(D p )={d pk ∣μ-3σ≤d pk ≤μ+3σ},

[0130] Among them, C(D p ) represents the processed data set after cleaning, μ represents the mean of the processed data, σ represents the standard deviation of the processed data, d pk represents the kth processed data;

[0131] Together with the processing data collected by the production module, the complete production information is formed, and the steps of updating the production information to the ERP system also include:

[0132] Fusion of dynamic data and processing data, dynamic data D d And the processing data after cleaning D c Fusion is performed by timestamp. The fusion function is refined into matching logic based on time windows. The fusion formula is:

[0133] F(D c ,D d )={(d c ,d d )∣|t c -t d |≤Δt},

[0134] Among them, F(D c ,D d ) represents the fused production information set, d c Represents the processing data after cleaning, d d Indicates dynamic data, t c and t d They represent the timestamps of the corresponding data, and Δt represents the allowed time window error;

[0135] The fused production information set F(D c ,D d ) Update to ERP system;

[0136] Specifically, the dynamic data is collected smoothly through the weighted moving average filter function, and the abnormal values ​​are effectively eliminated by combining the cleaning threshold method of the processed data. Then, the dynamic data and the processed data are fused through timestamp matching to generate complete production information, which is updated to the ERP system in real time to ensure the accuracy of data fusion and enhance the responsiveness of the ERP system to the dynamic production environment.

[0137] Step S3: Based on the real-time production information in ERP, a random forest model is used to predict the cost fluctuation trend, and a linear regression model is used to analyze the correlation between production efficiency and cost; based on the cost fluctuation trend and correlation, an adjustment plan for resource allocation and production scheduling is determined;

[0138] Based on the real-time production information in ERP, the steps of using the random forest model to predict the cost fluctuation trend are as follows:

[0139] Extract real-time production information from the ERP system, including cost data and production efficiency data, and define the extracted data set as:

[0140]

[0141] Where D represents the extracted data set, c k represents the actual cost of the kth record, e k represents the production efficiency of the kth record, k represents the record number, and n represents the total number of data records;

[0142] The cost data in the data set D is divided into a training set and a test set according to the time series. The training goal of the random forest model is defined as predicting the cost fluctuation trend at future moments. The formula of the random forest prediction model is:

[0143]

[0144] in, represents the prediction cost of the k+1th record, f RF represents the random forest model, x k The feature vector representing the kth record, including the current cost c k , production efficiency k and external environment variables,

[0145] The loss function of model training is defined as the mean square error, and the formula is:

[0146]

[0147] Among them, L represents the loss function value, N represents the number of training samples, and c k and denote the actual cost and the predicted cost respectively;

[0148] The linear regression model is used to analyze the relationship between production efficiency and cost. Based on the cost fluctuation trend and the relationship, the steps to determine the adjustment plan of resource allocation and production scheduling are as follows:

[0149] Using linear regression model to analyze production efficiency k For cost c k The regression formula is:

[0150] c k =β0+β1e k +∈ k ,

[0151] Among them, β0 represents the intercept, β1 represents the regression coefficient of production efficiency on cost, ∈ k represents the regression residual,

[0152] The regression coefficient is calculated by the least squares method, and the calculation formula is:

[0153]

[0154] in, represents the mean value of production efficiency, represents the mean cost,

[0155] Combining the prediction results of the random forest model with the correlation of linear regression analysis, a resource allocation and production scheduling adjustment plan is generated. The resource allocation optimization objective function is:

[0156]

[0157] Among them, R represents the total cost of resource allocation, α i represents the unit cost of the i-th resource, r i represents the allocation amount of the i-th resource, and M represents the number of resource types;

[0158] The production schedule adjustment plan aims to improve efficiency, and the adjustment model is:

[0159]

[0160] Among them, P represents the total production efficiency, β j represents the weight of the j-th production task, p j represents the completion efficiency of the jth task, and T represents the total number of tasks;

[0161] Specifically, the random forest model is used to predict cost fluctuation trends, helping companies make accurate judgments on future cost changes. Linear regression analysis is used to analyze the relationship between production efficiency and cost, providing theoretical support for resource optimization and production adjustments. Combining the prediction and analysis results, resource allocation and production scheduling adjustment plans are generated, which can effectively improve resource utilization and production efficiency, while reducing operating costs, achieving the dual goals of cost control and production optimization.

[0162] Step S4, applying the adjustment plan generated in step S3 to the finance and production modules;

[0163] In step S4, the financial module adjusts the procurement budget and capital flow arrangement according to the resource allocation adjustment plan:

[0164] The production module adjusts the production schedule and material usage plan according to the production schedule adjustment plan;

[0165] The applied adjustment plan continues to be fed back to the ERP system to further update resource allocation and production scheduling;

[0166] The steps of applying the adjustment plan generated in step S3 to the financial and production modules are:

[0167] According to the resource allocation adjustment plan R, optimize the procurement budget set B p :

[0168] B p = {b p1 ,b p2 ,…,b pM},

[0169] Among them, B p represents the optimized procurement budget set, b pi represents the purchase budget of the i-th resource, i represents the sequence number of the resource type, M represents the total number of resource types,

[0170] The budget adjustment formula is:

[0171] b pi =α i ·r i +Δb pi ,

[0172] Among them, α i represents the unit cost of the i-th resource, r i represents the demand for the i-th resource, Δb pi Indicates budget revisions due to market fluctuations or supply chain changes;

[0173] According to the adjustment results of the procurement budget, reallocate the capital flow and define the capital flow set as follows:

[0174] F={f1,f2,…,f K},

[0175] Among them, F represents the capital flow set, f k represents the capital allocation of the kth stage, k represents the sequence number of the capital flow stage, and K represents the total number of stages.

[0176] The cash flow adjustment formula is:

[0177]

[0178] Among them, γki It represents the proportion of resources allocated to the i-th resource in the k-th stage;

[0179] According to the production schedule adjustment plan P, optimize the production plan set:

[0180] S p ={s p1 ,s p2 ,…,s pT},

[0181] Among them, S p represents the optimized production schedule set, s pj represents the production plan of the jth task, j represents the task number, T represents the total number of tasks,

[0182] The production schedule adjustment formula is:

[0183]

[0184] Among them, β j represents the priority weight of the jth task, e j Indicates the execution efficiency of the current j-th task;

[0185] According to the task plan adjusted by the production schedule, optimize the material usage plan set:

[0186] M u ={m u1 ,m u2 ,…,m uT},

[0187] Among them, M u Represents the optimized material usage plan set, m uj represents the material requirement of the jth task, j represents the task number,

[0188] The material usage adjustment formula is:

[0189]

[0190] Among them, λ ji Indicates the usage ratio of the i-th resource required for the j-th task;

[0191] Specifically, this step adjusts the procurement budget and cash flow arrangement according to the resource allocation plan to ensure that the resource allocation of the financial module can support the optimized production schedule. At the same time, the production module reallocates task priorities and optimizes material usage plans based on the adjusted scheduling plan. The adjusted plan improves the efficiency of capital and resource utilization, while ensuring the smooth operation of the production process, providing a guarantee for the company to achieve cost savings and improve production efficiency.

[0192] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A large-scale production and operation management system of an ERP system, comprising a production module, a financial module and an ERP system, characterized in that: The production module is used to collect processing data and collect dynamic data at the production site based on IoT devices and sensor technology. The production module combines static processing data with dynamic data to form production information, which is synchronized to the ERP system. The production module is also used to receive the adjusted production schedule and material usage plan; The financial module is used to receive processing data and dynamic data from the ERP system to determine whether the budget is consistent with the plan; The financial module adjusts the procurement budget and capital flow arrangement according to the generated resource allocation adjustment plan, and feeds back the dynamically adjusted budget and capital allocation plan to the ERP system and production module; The ERP system integrates the input and output of the financial module and the production module, and is used to update and store processing data, dynamic data, financial analysis results, resource allocation, and production scheduling adjustment plans.

2. The large-scale production and operation management system of an ERP system according to claim 1, characterized in that: The processing data includes order requirements, material usage, production schedule and preliminary resource allocation plan; the dynamic data includes equipment operating status, production progress, actual material consumption and inventory changes.

3. A large model production and operation management method of an ERP system, based on a large model production and operation management system of an ERP system according to any one of claims 1 to 2, characterized in that: include: Step S1, the production module collects processing data, including order requirements, material usage, production schedule and preliminary resource allocation plan; Processing data is updated to the ERP system and synchronized with the financial module as the basis for preliminary financial analysis to determine whether the budget is consistent with the plan; Step S2: At the production site, dynamic data is collected based on IoT devices and sensor technology, including equipment operation status, production progress, actual material consumption, and inventory changes. The dynamic data serves as a real-time supplement to the processing data, and together with the processing data collected by the production module, it constitutes complete production information, and the production information is updated to the ERP system. Step S3, based on the real-time production information in ERP, a random forest model is used to predict the cost fluctuation trend, and a linear regression model is used to analyze the correlation between production efficiency and cost; Determine resource allocation and production schedule adjustment plans based on cost fluctuation trends and correlations; Step S4, applying the adjustment plan generated in step S3 to the finance and production modules.

4. A large-scale model production and operation management method of an ERP system as claimed in claim 3, characterized in that: The steps for determining whether the budget is consistent with the plan are: Obtain processing data from the production module and synchronize it to the financial module; Eliminate outliers, fill in missing values, calculate the actual cost of each production according to the process unit price and consumption, compare the actual cost with the budget plan item by item, and analyze the deviation; Output deviation analysis results and mark abnormal data that exceeds expectations.

5. A large-scale model production and operation management method of an ERP system as claimed in claim 4, characterized in that: The processing data collected by the production module together constitute complete production information, and the steps of updating the production information to the ERP system are: Collect dynamic data and define the dynamic data set as: D d ={d d1 ,d d2 ,...,d dm }, Among them, D d Represents a dynamic data set, d dj Indicates the jth dynamic data, j indicates the sequence number of the dynamic data, and its value range is 1≤j≤m, and m indicates the total number of dynamic data; The acquisition function for collecting dynamic data is refined into a filter function based on weighted moving average, and the formula is: Among them, H(S) represents the dynamic data acquisition function, S represents the real-time monitoring data flow, and s t represents the real-time data at time t, w t represents the weight at time t, and T represents the length of the time window covered by the weight; Perform processing data cleaning and define the processing data set as: D p ={d p1 ,d p2 ,...,d pn }, Among them, D p represents the processing data set, d pk Indicates the kth processed data, k indicates the sequence number of the processed data, and its value range is 1≤k≤n, and n indicates the total number of processed data. The processed data is cleaned and refined into a threshold function to remove outliers: C(D p )={d pk |μ-3σ≤d pk ≤μ+3σ}, Among them, C(D p ) represents the processed data set after cleaning, μ represents the mean of the processed data, σ represents the standard deviation of the processed data, d pk Represents the kth processed data.

6. A large-scale model production and operation management method of an ERP system as claimed in claim 5, characterized in that: The processing data collected by the production module together constitute complete production information, and the step of updating the production information to the ERP system also includes: Fusion of dynamic data and processing data, dynamic data D d And the processing data after cleaning D c Fusion is performed by timestamp. The fusion function is refined into matching logic based on time windows. The fusion formula is: F(D c ,D d )={(d c ,d d )||t c -t d |≤Δt}, Among them, F(D c , D d ) represents the fused production information set, d c Represents the processing data after cleaning, d d Indicates dynamic data, t c and t d They represent the timestamps of the corresponding data, and Δt represents the allowed time window error; The fused production information set F(D c , D d ) is updated to the ERP system.

7. A large-scale model production and operation management method of an ERP system as claimed in claim 6, characterized in that: The steps of using the random forest model to predict the cost fluctuation trend based on the real-time production information in ERP are as follows: Extract real-time production information from the ERP system, including cost data and production efficiency data, and define the extracted data set as: Where D represents the extracted data set, c k represents the actual cost of the kth record, e k represents the production efficiency of the kth record, k represents the record number, and n represents the total number of data records; The cost data in the data set D is divided into a training set and a test set according to the time series. The training goal of the random forest model is defined as predicting the cost fluctuation trend at future moments. The formula of the random forest prediction model is: in, represents the prediction cost of the k+1th record, f RF represents the random forest model, x k The feature vector representing the kth record, including the current cost c k , production efficiency k and external environment variables, The loss function of model training is defined as the mean square error, and the formula is: Among them, L represents the loss function value, N represents the number of training samples, and c k and Represent the actual cost and predicted cost respectively.

8. A large-scale model production and operation management method of an ERP system as claimed in claim 7, characterized in that: The linear regression model is used to analyze the correlation between production efficiency and cost; based on the cost fluctuation trend and the correlation, the steps of determining the adjustment plan of resource allocation and production scheduling are as follows: Using linear regression model to analyze production efficiency k For cost c k The regression formula is: c k =β0+β1e k +∈ k , Among them, β0 represents the intercept, β1 represents the regression coefficient of production efficiency on cost, ∈ k represents the regression residual, The regression coefficient is calculated by the least squares method, and the calculation formula is: in, represents the mean value of production efficiency, represents the mean cost, Combining the prediction results of the random forest model with the correlation of linear regression analysis, a resource allocation and production scheduling adjustment plan is generated. The resource allocation optimization objective function is: Among them, R represents the total cost of resource allocation, α i represents the unit cost of the i-th resource, r i represents the allocation amount of the i-th resource, and M represents the number of resource types; The production schedule adjustment plan aims to improve efficiency, and the adjustment model is: Among them, P represents the total production efficiency, β j represents the weight of the j-th production task, p j represents the completion efficiency of the jth task, and T represents the total number of tasks.

9. A large-scale model production and operation management method of an ERP system as claimed in claim 8, characterized in that: In step S4, the financial module adjusts the procurement budget and capital flow arrangement according to the resource allocation adjustment plan: The production module adjusts the production schedule and material usage plan according to the production schedule adjustment plan; The applied adjustment plan continues to be fed back to the ERP system to further update resource allocation and production scheduling.

10. A large-scale model production and operation management method of an ERP system as claimed in claim 9, characterized in that: The step of applying the adjustment plan generated in step S3 to the financial and production modules is: According to the resource allocation adjustment plan R, optimize the procurement budget set B p : B p ={b p1 ,b p2 ,...,b pM }, Among them, B p represents the optimized procurement budget set, b pi represents the purchase budget of the i-th resource, i represents the sequence number of the resource type, M represents the total number of resource types, The budget adjustment formula is: b pi =a i ·r i +Δb pi , Among them, α i represents the unit cost of the i-th resource, r i represents the demand for the i-th resource, Δb pi Indicates budget revisions due to market fluctuations or supply chain changes; According to the adjustment results of the procurement budget, reallocate the capital flow and define the capital flow set as follows: F={f1,f2,...,f K }, Among them, F represents the capital flow set, f k represents the capital allocation of the kth stage, k represents the sequence number of the capital flow stage, and K represents the total number of stages. The cash flow adjustment formula is: Among them, γ ki It represents the proportion of resources allocated to the i-th resource in the k-th stage; According to the production schedule adjustment plan P, optimize the production plan set: S p ={S p1 ,s p2 ,...,s pT }, Among them, S p represents the optimized production schedule set, S pj represents the production plan of the jth task, j represents the task number, T represents the total number of tasks, The production schedule adjustment formula is: Among them, β j represents the priority weight of the jth task, e j Indicates the execution efficiency of the current j-th task; According to the task plan adjusted by the production schedule, optimize the material usage plan set: M u ={m u1 ,m u2 ,...,m uT }, Among them, M u Represents the optimized material usage plan set, m uj represents the material requirement of the jth task, j represents the task number, The material usage adjustment formula is: Among them, λ ji Indicates the usage ratio of the i-th resource required for the j-th task.

Images