Seamless pipe real-time cost accounting algorithm based on production management system

Abstract

This invention relates to the field of cost accounting and discloses a seamless real-time cost accounting algorithm based on a production management system, providing a full-chain real-time cost accounting solution. It includes constructing a raw material procurement price matrix, using a sliding window method to accurately match production order times with procurement time periods to determine raw material input costs; real-time collection of machine start-up and completion times, combined with time quotas to calculate labor costs; collection of total energy consumption readings and the operating time of each piece of equipment, constructing a system of linear equations to solve for the real-time power and energy consumption costs of individual equipment, solving the energy consumption allocation problem for equipment without independent electricity meters; and calculating cumulative losses based on the yield rate of each process, deducting the value of waste recycling. This invention achieves refined control of production costs, providing real-time and accurate data support for sales quotations.

Description

Technical Field

[0001] This invention relates to the field of cost accounting, and more particularly to a seamless real-time cost accounting algorithm based on a production management system. Background Technology

[0002] As the manufacturing industry enters an era of refined management, the pipe processing industry faces increasingly fierce market competition. Seamless steel pipes, as a fundamental raw material in industry, exhibit typical characteristics of a flow-type production model. As shown in the figure, the production process of seamless pipes typically includes a series of continuous steps such as heating round steel, piercing, rolling, cooling, straightening, cutting ends, flaw detection, chamfering, and packaging.

[0003] In this continuous production model, raw materials need to be continuously fed into the equipment for heating and processing, making it difficult to frequently start and stop the production line for single, small-batch orders. This process characteristic makes it difficult for companies to accurately determine the true production cost of a specific specification and batch of products when faced with customer inquiries, thus making it impossible to provide a precise quote that is both competitive and guarantees profit margins.

[0004] Currently, the commonly used cost accounting method in the industry mainly adopts the "allocation method," which roughly estimates the unit cost using the formula "(total sales revenue - total material cost - total management cost) / total production quantity (or total weight)." However, in actual production and operation, this extensive accounting method has the following significant technical pain points and management blind spots:

[0005] Existing technologies often overlook the cost differences that arise when different pipe types are used, at different times, or by production personnel with varying skill levels operating equipment with different energy efficiency levels. In reality, these factors can cause significant cost fluctuations between different batches of the same product specification. The amortization method masks these differences, leading to distorted pricing data.

[0006] The main raw material for seamless steel pipes, round steel, is mostly sourced from the futures market, and its purchase price fluctuates dramatically over time. Although existing ERP systems record purchase orders, in continuous production, it is difficult for companies to accurately match the current batch of round steel used in production with specific purchase prices at different points in history (i.e., it is impossible to accurately track "which batch of steel at which price was used for this pipe"). This results in material cost accounting often having to use monthly weighted average prices, failing to reflect real-time market cost fluctuations.

[0007] In seamless steel pipe production workshops, there are usually only general water, electricity, and gas meters; most factories fail to equip each piece of production equipment with a dedicated electricity or gas meter. Therefore, current technology cannot accurately calculate the actual energy consumption of a single piece of equipment during a specific order's production period; it can only simply average the total energy consumption of the workshop. This results in the cost of high-energy-consuming and low-energy-consuming processes being averaged, failing to truly reflect the energy costs of different process paths.

[0008] Therefore, we propose a seamless real-time cost accounting algorithm based on a production management system to solve the above problems. Summary of the Invention

[0009] This invention provides a seamless real-time cost accounting algorithm based on a production management system, which is used to provide a real-time cost accounting solution for the entire chain.

[0010] The first aspect of this invention provides a real-time cost accounting method for seamless tubes based on a production management system, applied to seamless tube production scenarios. The method includes: acquiring the raw material specifications and input time of a target production order; establishing a raw material purchase price matrix with time period as the horizontal axis and supplier and specification as the vertical axis; matching the corresponding purchase unit price in the raw material purchase price matrix according to the input time; obtaining the raw material cost based on the purchase unit price and the raw material input weight of the target production order; collecting the start time and completion time of operators for the target production order, calculating the difference between the two to obtain working hours, and obtaining a preset piece-rate price corresponding to the process. The system outputs labor costs; acquires the total energy consumption reading of the workshop within a preset statistical period, and simultaneously acquires the running time of each production device connected to the meter that generates the total energy consumption reading, thus obtaining energy costs; acquires the yield rate of each process node through which the target production order passes, calculates the accumulated loss weight based on the raw material input weight and the yield rate of each process node according to the process flow sequence, acquires the current scrap recycling unit price, and obtains the scrap recycling value; and obtains the unit real-time production cost of seamless tubes based on the raw material cost, the labor cost, the energy cost, the scrap recycling value, and the reported work weight of the target production order.

[0011] Optionally, in a first implementation of the first aspect of the present invention, the specific method for establishing the raw material procurement price matrix includes: obtaining purchase order data from an ERP system, wherein the purchase order data includes purchase time, supplier information, specifications and purchase price; processing the purchase time using a sliding window method, dividing the continuous time axis into several discrete time period windows; filling the purchase price within each time period window into the matrix to form the raw material procurement price matrix, so as to index the corresponding purchase unit price through the time period window into which the input time falls.

[0012] Optionally, in a second implementation of the first aspect of the present invention, a step of predicting the raw material cost of a specific supplier is further included: when the target production order specifies a specific supplier and the supplier has not yet generated a current purchase order, multiple historical quotation data of the specific supplier within a historical time period are obtained; a multinomial regression model of the quotation value with time and fluctuation factors is established; the price estimate for a future preset time period is calculated using the multinomial regression model, and the price estimate is used as the purchase unit price.

[0013] Optionally, in the third implementation of the first aspect of the present invention, the specific formula for calculating labor costs is: labor cost = (completion time - start time) × the time quota of the machine for the process; wherein, the start time and completion time are captured in real time by the production management system from the work report records at the production site.

[0014] Optionally, in the fourth implementation of the first aspect of the present invention, the specific steps for calculating the average operating power of each production device include:

[0015] Construct a linear equation with multiple variables: ;

[0016] in, This represents the total energy consumption reading per unit time. Let be the running time of the nth device per unit time. Let the average operating power of the nth device be denoted by . Collect multiple sets of total energy consumption readings per unit time and the corresponding operating time of each device, and construct a system of linear equations. Solve the system of linear equations in groups to obtain multiple sets of solutions for the average operating power of each device, and take the average value of the multiple sets of solutions to obtain the average operating power.

[0017] Optionally, in the fifth implementation of the first aspect of the present invention, the specific formula for calculating the accumulated loss weight is as follows:

[0018] ;

[0019] in, Input weight for raw materials, to This refers to the yield rate of each process node arranged in the process sequence.

[0020] Optionally, in a sixth implementation of the first aspect of the present invention, the method further includes a step of predicting the scrap rate based on a hybrid model: obtaining characteristic parameters affecting the yield, wherein the characteristic parameters include at least: the quality factor of the raw material supplier, the pressure parameter of the piercing machine, the temperature parameter of the piercing machine, and the heat treatment time parameter; establishing a functional relationship between the scrap rate and the characteristic parameters; establishing a prediction model using a hybrid algorithm of decision tree and multinomial regression, predicting the future scrap rate based on the current production process parameters, and calculating the loss weight using the predicted scrap rate.

[0021] Optionally, in the seventh implementation of the first aspect of the present invention, a rework cost accounting step is also included: when the production management system detects rework, the initial material input quantity is recorded. and additional material input for rework Obtain the average yield of this type of seamless tube. Based on the aforementioned unit real-time production cost The formula for calculating the increased cost of a single rework is as follows:

[0022] .

[0023] Beneficial effects: By accurately allocating losses to the process chain and precisely matching the current raw material prices, companies can clearly know whether a particular order is profitable, thus providing the sales department with a highly competitive and secure price floor.

[0024] There is no need to equip each piece of equipment in the workshop with a separate electricity or gas meter. The energy consumption of a single machine can be accurately allocated. The average power is calculated by solving the equation system algorithm. While ensuring that the data is sufficient to support the pricing decision, a lot of instrument and meter procurement costs and subsequent calibration and maintenance work are saved.

[0025] In a market environment with volatile raw material prices, the ability to predict cost trends and avoid losses from quoting prices can be achieved by using regression analysis of suppliers' historical quotations. This allows for high-confidence cost forecasts to be provided even before raw materials are actually purchased (or prices are uncertain), helping companies mitigate operational risks from fluctuations in the futures market.

[0026] Production managers can intuitively see how adjusting a certain process parameter will affect the final scrap cost, thereby guiding process optimization and reducing production losses. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the manufacturing process of seamless tubes.

[0028] Figure 2 A cost diagram for a unit quantity of steel pipes.

[0029] Figure 3 This is a schematic diagram of the procurement price matrix for round steel.

[0030] Figure 4 This is a schematic diagram of the matching matrix for all time periods and purchase unit prices.

[0031] Figure 5 To obtain a schematic diagram of the purchase price.

[0032] Figure 6 This is a diagram illustrating the estimated prices of raw materials.

[0033] Figure 7 This is a diagram illustrating the method of calculating working hours.

[0034] Figure 8 This is a schematic diagram of the electricity meter connection structure in the seamless tube production workshop.

[0035] Figure 9 This diagram illustrates the electricity consumption displayed on the meter and the operating time of each device, calculated on an hourly basis.

[0036] Figure 10 This is a diagram illustrating how losses are broken down into orders. Detailed Implementation

[0037] This invention provides a seamless real-time cost accounting algorithm based on a production management system, used to provide a full-link real-time cost accounting solution. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0038] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the seamless real-time cost accounting method based on a production management system in this invention includes:

[0039] 101. Calculate the raw material cost of the order; obtain the raw material specifications and input time of the target production order through the production management system; establish a raw material purchase price matrix with time period as the horizontal axis and supplier and specification as the vertical axis; match the corresponding purchase unit price in the raw material purchase price matrix according to the input time, and multiply the purchase unit price by the raw material input weight of the target production order to obtain the raw material cost.

[0040] It is understood that the executing entity of this invention can be a seamless real-time cost accounting device based on a production management system, or it can be a terminal or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0041] Specifically, the steps for forecasting raw material costs for a specific supplier are as follows:

[0042] When a target production order specifies a particular supplier, and that supplier has not yet generated a purchase order for the current period, retrieve multiple historical quotes from that particular supplier within the historical time period.

[0043] Establish a multinomial regression model of price quotes with time and volatility factors;

[0044] The estimated price for a future time period is calculated using a multinomial regression model, and this estimated price is used as the unit purchase price.

[0045] 102. Calculate the labor cost of the order; collect the start time and completion time of the operators for the target production order through the production management system, calculate the difference between the two to obtain the working hours; obtain the preset piece rate unit price corresponding to the process; multiply the working hours by the piece rate unit price to obtain the labor cost;

[0046] Specifically, the formula for calculating labor costs is as follows:

[0047] Labor cost = (completion time - start time) × time quota of the machine for this process;

[0048] The start and completion times are retrieved in real time from the work reports on the production site by the production management system.

[0049] 103. Calculate equipment energy costs; obtain the total energy consumption reading of the workshop within a preset statistical period, and simultaneously obtain the running time of each production equipment connected to the meter that generates the total energy consumption reading; based on the linear relationship between the total energy consumption reading and the running time of each production equipment, construct and solve a set of energy consumption equations to obtain the average operating power of each production equipment; calculate the energy cost based on the equipment running time and average operating power occupied by the target production order.

[0050] Specifically, the steps to calculate the average operating power of each production device include:

[0051] Construct a linear equation with multiple variables: ;

[0052] in, This represents the total energy consumption reading per unit time. Let be the running time of the nth device per unit time. Let n be the average operating power of the nth device;

[0053] Collect multiple sets of total energy consumption readings per unit time and the corresponding operating time of each device, and construct a system of linear equations;

[0054] The linear equations are solved in groups to obtain multiple solutions for the average operating power of each device. The average operating power is obtained by averaging the solutions.

[0055] 104. Calculate waste loss deduction; obtain the yield rate of each process node in the target production order; calculate the accumulated loss weight according to the process flow sequence based on the raw material input weight and the yield rate of each process node; obtain the current waste recycling unit price, multiply the loss weight by the waste recycling unit price to obtain the waste recycling value.

[0056] Specifically, the formula for calculating the accumulated weight loss is as follows:

[0057] ;

[0058] in, Input weight for raw materials, to This refers to the yield rate of each process node arranged in the process sequence.

[0059] Furthermore, it also includes the step of predicting the scrap rate based on a hybrid model:

[0060] Obtain the characteristic parameters that affect the yield rate. The characteristic parameters should include at least the following:

[0061] The quality factors of the raw material supplier, the pressure parameters of the piercing machine, the temperature parameters of the piercing machine, and the heat treatment time parameters;

[0062] Establish the functional relationship between waste rate and characteristic parameters;

[0063] A prediction model is established using a hybrid algorithm of decision tree and multinomial regression. Based on the current production process parameters, the future scrap rate is predicted, and the weight loss is calculated using the predicted scrap rate.

[0064] 105. Calculate the unit real-time cost; add the raw material cost, labor cost, and energy cost, and subtract the waste recycling value to obtain the total net cost; divide the total net cost by the reported weight of the target production order to obtain the unit real-time production cost of the seamless tube.

[0065] Furthermore, it also includes the rework cost accounting steps:

[0066] When the production management system detects rework, it records the initial material input quantity. and additional material input for rework ;

[0067] Obtain the average yield of this type of seamless tube. ;

[0068] Based on unit real-time production cost The formula for calculating the increased cost of a single rework is as follows:

[0069] .

[0070] The following describes the specific implementation schemes, and the manufacturing process of the seamless tube is as follows: Figure 1 As shown:

[0071] from Figure 1 The data shows that the production process of seamless steel pipes is a flow-type production model, meaning that round steel bars need to be continuously heated before entering the equipment for processing, making it difficult to produce according to orders. This makes it difficult to calculate the cost per unit. The most common cost accounting method in the industry is (sales revenue - material cost - management cost) / production quantity or weight per unit model, i.e., the amortization method.

[0072] However, in reality, the cost can vary greatly depending on the type of pipe, the time, the production personnel, and the production equipment. Accurate cost data is needed when quoting sales prices, so obtaining the most real-time production cost possible has become a necessary condition for refined management.

[0073] Reference Figure 2 We divide the cost per unit quantity of steel pipes into four parts: materials, labor, overhead, and waste material loss. The statistical methods for each part will be analyzed below.

[0074] Material costs for seamless steel pipes are relatively straightforward, mainly consisting of round steel bars, auxiliary materials such as pickling and phosphating processes, and hardware. The cost of auxiliary materials and hardware accounts for less than 1% of the total material cost and is not analyzed here. The cost of round steel bars primarily comes from the purchase price, which is largely sourced from the futures market due to its volatile price. We have defined a round steel purchase price matrix, as follows: Figure 3 As shown:

[0075] The horizontal axis T1-TN represents the procurement time. We use the sliding window method to define a matching matrix between all time periods and the unit price, as follows: Figure 4 ;

[0076] In this way, we can match the purchase price of a certain supplier a, a certain specification b, and a certain time t.

[0077] Next, let's clarify how each ton of finished pipe is linked to a specific purchase price. The production management system needs to retrieve purchase orders from the ERP system to obtain information such as purchase prices. Figure 5 As shown:

[0078] The above is the association algorithm that matches each order based on the raw materials that have already been purchased. For some customers who need to specify raw material suppliers, since raw material procurement occurs after the sales quotation, it is necessary to make a predictive assessment of the supplier's supply cost to ensure the accuracy of the quotation;

[0079] We assume that within a year x, suppliers y1 to yn will submit a total of n quotes, denoted as x1, x2, x3, ..., xn, with each quote corresponding to p1, p2, p3, ..., pn. Assuming a relationship between quotes and time, we assign p = f(x). We also define the fluctuation factor for each supplier as z1 to zn, resulting in p = f(t, z). We use a multinomial regression algorithm for modeling, such as... Figure 6 ;

[0080] This chart provides an estimate of raw material prices for the next 1-2 times, which can guide current pricing.

[0081] Labor costs; the labor costs involved in the production of seamless steel pipes include two parts: the labor costs of on-site workers in the workshop and the labor costs of all management personnel. Since the labor costs of management personnel are relatively fixed, they are not included in this calculation.

[0082] Workers in the workshop are paid on a piece-rate basis. Worker wages = piece-rate price (hourly rate) × hours. The piece-rate price is relatively fixed by the finance department, while the hours need to be retrieved from the production system. The calculation method is as follows: Figure 7 ;

[0083] Worker's working hours = t2 - t1, which is the completion time - start time. Suppose there are two workers, a and b, who start working on machine c at time ta1 and finish at time ta2. Then worker a's working hours = ta2 - ta1, and the wage for this piece = (worker a's working hours quota on machine c) × (ta2 - ta1).

[0084] Costs: As a seamless steel pipe manufacturing plant, the main costs come from water, electricity, gas, and some auxiliary materials. The costs of water, electricity, and gas are derived from data collected from the water, electricity, and gas meters within the production system. The cost of water is relatively low compared to electricity and gas, and is therefore negligible here. A diagram showing the meter connection architecture of a seamless steel pipe production workshop is provided below. Figure 8 ;

[0085] like Figure 8 As shown, most factories have not equipped each piece of equipment with a dedicated electricity meter, making it crucial to accurately calculate the electricity consumption of each individual device. This involves calculating the hourly electricity consumption displayed on the meters along with the operating time of each device (derived from monitoring networked devices within the production system). Figure 9 As shown;

[0086] The electricity meter readings and operating time are known. To calculate the average power of the equipment, assume the average power of equipment 1 is... The average power of device 2 is And so on, the average power of device n Then we can derive the equation. In the next n hours, there will be n similar equations, that is... And so on:

[0087] ;

[0088] ; ...

[0090] ;

[0091] Because here ~ The values ​​are floating, with a small range of fluctuation. There are three variables. By combining this system of equations, we can form... A system of three linear equations in three variables will yield the following results. The solutions are then averaged to obtain the final average power. ~ This allows you to calculate the real-time electricity cost for each device (note: there may be some discrepancies with the master bill, but it is sufficient as a basis for sales quotations).

[0092] The real-time gas cost of the natural gas meter is calculated using the logic described above.

[0093] Scrap and Losses; Seamless pipe manufacturing generates significant losses. For example, some processes require cutting both ends of the pipe to meet customer length requirements and maintain flatness—these are called pipe cuts, which are then weighed and sold as scrap steel. The key is to allocate these losses to orders. This is because most processes in seamless pipe manufacturing involve losses. Figure 10 For example;

[0094] If the order uses t tons of raw materials, the loss will be... ;

[0095] If the current selling price of scrap steel is p yuan / ton, then the cost of this order can be reduced by p. loss element;

[0096] Furthermore, during the project's R&D phase, precise control of the scrap rate is necessary to reduce material waste. The scrap rate (i.e., yield) is directly related to the quality of raw materials and process parameters. Raw material quality is primarily related to the supplier; we define the supplier quality factor as x. Among the process parameters, the pressure y and temperature z of the piercing machine, as well as the heat treatment time m, are directly related. Therefore, the scrap rate n = f(x, y, z, m), and we establish a hybrid model using a combination of decision trees and multinomial regression.

[0097] This allows us to predict the scrap rate in the next small phase of the production process and indirectly calculate the cost of scrap.

[0098] Rework costs: During the R&D or early mass production phases, insufficient process expertise and skill levels lead to low yields, further increasing rework volume. Along with rework, corresponding material, labor, and overhead costs also increase; the calculation method for material, labor, and overhead costs is the same as described above. Let the average yield of a certain seamless tube model in the initial mass production phase be... If t tons need to be produced, and one rework is required (first feeding t1 tons, second feeding t2 tons), then Let c be the cost of material handling and waste loss calculated using the above formula. Then, the cost of one rework is... ;

[0099] From this point on, we can accurately track the time from raw material loading to employee reporting work on the equipment, calculate the working hours, and calculate the cost of electricity and gas consumed by the equipment within that time frame. By dividing this cost by the reported weight, we can obtain the real-time cost per ton, providing a highly reliable basis for the sales department to quote prices.

[0100] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A seamless real-time cost accounting method based on a production management system, characterized in that, Applications include seamless tube production scenarios, including: Obtain the raw material specifications and input time of the target production order, establish a raw material purchase price matrix with time period as the horizontal axis and supplier and specification as the vertical axis, match the corresponding purchase unit price in the raw material purchase price matrix according to the input time, and obtain the raw material cost based on the purchase unit price and the raw material input weight of the target production order. The operator collects the start and end times of the target production order, calculates the difference between the two to obtain the working hours, obtains the preset piece-rate price corresponding to the process, and outputs the labor cost. Obtain the total energy consumption reading of the workshop within a preset statistical period, and simultaneously obtain the running time of each production equipment connected to the meter that generates the total energy consumption reading, to obtain the energy cost; Obtain the yield rate of each process node through which the target production order passes. Based on the input weight of the raw materials and the yield rate of each process node, calculate the accumulated loss weight according to the process flow sequence, obtain the current waste recycling unit price, and obtain the waste recycling value. The unit real-time production cost of seamless tubes is obtained based on the raw material cost, labor cost, energy cost, waste recycling value, and the reported weight of the target production order.

2. The seamless pipe real-time cost accounting method based on a production management system according to claim 1, characterized in that, Specific methods for establishing a raw material procurement price matrix include: Obtain purchase order data from the ERP system, which includes purchase time, supplier information, specifications and purchase price; The procurement time is processed using the sliding window method, which divides the continuous time axis into several discrete time window segments. The purchase prices within each time window are filled into a matrix to form the raw material purchase price matrix, so that the corresponding purchase unit price can be indexed by the time window in which the input time falls.

3. The seamless pipe real-time cost accounting method based on a production management system according to claim 1, characterized in that, It also includes a step for forecasting raw material costs for specific suppliers: When the target production order specifies a particular supplier, and that supplier has not yet generated a current purchase order, obtain multiple historical quotation data of that particular supplier within a historical time period; Establish a multinomial regression model of price quotes with time and volatility factors; The estimated price for a future preset time period is calculated using the polynomial regression model, and this estimated price is used as the purchase unit price.

4. The seamless pipe real-time cost accounting method based on a production management system according to claim 3, characterized in that, The specific formula for calculating labor costs is as follows: Labor cost = (completion time - start time) × time quota of the machine for this process; The start and completion times are obtained in real time from the work reports on the production site by the production management system.

5. The seamless pipe real-time cost accounting method based on a production management system according to claim 1, characterized in that, The specific steps for determining the average operating power of each production device include: Construct a linear equation with multiple variables: ; in, This represents the total energy consumption reading per unit time. Let be the running time of the nth device per unit time. Let n be the average operating power of the nth device; Collect multiple sets of total energy consumption readings per unit time and the corresponding operating time of each device, and construct a system of linear equations; The linear equations are solved in groups to obtain multiple solutions for the average operating power of each device. The average operating power is obtained by averaging the solutions.

6. The seamless pipe real-time cost accounting method based on a production management system according to claim 5, characterized in that, The specific formula for calculating the accumulated weight loss is as follows: ； in, Input weight for raw materials, to This refers to the yield rate of each process node arranged in the process sequence.

7. The seamless pipe real-time cost accounting method based on a production management system according to claim 1, characterized in that, It also includes the step of predicting the scrap rate based on a hybrid model: Obtain characteristic parameters that affect the yield, wherein the characteristic parameters include at least: The quality factors of the raw material supplier, the pressure parameters of the piercing machine, the temperature parameters of the piercing machine, and the heat treatment time parameters; Establish a functional relationship between the waste rate and the characteristic parameters; A prediction model is established using a hybrid algorithm of decision tree and multinomial regression. Based on the current production process parameters, the future scrap rate is predicted, and the loss weight is calculated using the predicted scrap rate.

8. The seamless pipe real-time cost accounting method based on a production management system according to claim 1 or 7, characterized in that, It also includes the rework cost accounting steps: When the production management system detects rework, it records the initial material input quantity. and additional material input for rework ; Obtain the average yield of this type of seamless tube. ; Based on the aforementioned unit real-time production cost The formula for calculating the increased cost of a single rework is as follows: 。

Images