CN115935196A

CN115935196A - Matching degree calculation method, optimization method and device of process and production line

Info

Publication number: CN115935196A
Application number: CN202211448851.3A
Authority: CN
Inventors: 袁云浩; 唐思远; 王影; 沈克强; 陈智胜
Original assignee: Wuxi Yaoming Biotechnology Co ltd
Current assignee: Wuxi Yaoming Biotechnology Co ltd
Priority date: 2022-11-18
Filing date: 2022-11-18
Publication date: 2023-04-07

Abstract

The invention provides a method and a device for calculating the matching degree of a process and a production line and an optimization method and device. The calculation method comprises the following steps: obtaining key process parameters of each unit operation in M unit operations forming the process flow and probability distribution of each key process parameter, wherein M is an integer greater than 1; establishing a process flow model, wherein the process flow model comprises M unit operation models; inputting key process parameters of each unit operation and corresponding probability distribution into the process flow model based on Monte Carlo simulation to obtain output information, wherein the output information comprises output load capacity distribution, output volume distribution and output product retention time distribution; and calculating the total matching degree of the process and the production line according to the output information and the limiting parameters of the production line. The method for calculating the matching degree of the process and the production line fills the blank that the matching degree does not have quantitative indexes as references. The optimization method optimizes the mismatching degree on the basis of not changing the layout of the existing production line.

Description

Matching degree calculation method, optimization method and device of process and production line

Technical Field

The invention mainly relates to the field of manufacturing, in particular to a method, a device and a medium for calculating the matching degree of a process and a production line, an optimization method and an optimization device.

Background

In recent years, the cell culture titers of monoclonal antibodies (mAbs) have increased dramatically with improvements in cell lines, media components, and feeding strategies. Higher titer processes can present facility-adaptive challenges to traditional biopharmaceutical purification equipment whose capabilities are initially matched to lower titer processes. The bottleneck caused by the mismatch in equipment size, coupled with the process fluctuations after amplification, can result in the disposal of expensive products.

A current common strategy is to use production line adaptation software (e.g. superspro Designer) to calculate the match ratio of the process and the production line. Conventional adaptation software typically simulates the process flow using a single value (maximum or minimum), and may not be able to determine the correct match rate without taking into account the randomness of the process flow. Some worst-case combinations may result in production exceeding the capacity of the equipment, resulting in expensive products having to be discarded.

Disclosure of Invention

The invention aims to provide a method for calculating the matching degree of a process and a production line, an optimization method, a device and a computer readable medium, which can improve the matching accuracy.

In order to solve the technical problem, the invention provides a method for calculating the matching degree of a process and a production line, which comprises the following steps: obtaining a key process parameter of each unit operation in M unit operations forming the process flow and the probability distribution of each key process parameter, wherein M is an integer greater than 1; establishing a process flow model, wherein the process flow model comprises M unit operation models; inputting key process parameters of each unit operation and corresponding probability distribution into the process flow model based on Monte Carlo simulation to obtain output information, wherein the output information comprises output load capacity distribution, output volume distribution and output product retention time distribution; and calculating the total matching degree of the process and the production line according to the output information and the limiting parameters of the production line.

Optionally, key process parameters for each unit operation are obtained from expert knowledge or from a distribution fit of historical data.

Optionally, the step of obtaining the key process parameters of each unit operation according to the distribution fitting of the historical data comprises: judging whether the project to be produced has complete process flow information, if so, acquiring the key process parameters of each unit operation from the historical process flow information of the project, and if not, acquiring the key process parameters of each unit operation from the historical process flow information of the project similar to the project.

Optionally, the step of obtaining a probability distribution for each key process parameter comprises matching a most conforming distribution type among the plurality of mathematical distributions based on maximum likelihood estimation.

Optionally, the step of inputting the key process parameters of each unit operation and their corresponding probability distributions into the process flow model based on the monte carlo simulation comprises: step a: inputting the key process parameters of each unit operation and the probability distribution of the key process parameters into corresponding unit operation models based on Monte Carlo simulation, and outputting a group of output information by each unit operation model; step b: and repeating the step a for N times, wherein each unit operation model outputs N groups of output information, and the process flow model outputs N × M groups of output information, wherein N is a positive integer greater than 1.

Optionally, the step of calculating the total matching degree of the process and the production line according to the output information and the limiting parameters of the production line comprises: calculating the matching degree of the process flow and multiple dimensions of the production line according to the output information and the limiting parameters of the production line; and calculating the total matching degree of the process and the production line according to the matching degrees of the multiple dimensions.

Optionally, the matching degrees of the plurality of dimensions include a capacity matching degree, a product collection volume matching degree, and a product placement time matching degree.

Optionally, the total matching degree of the process and the production line is calculated by the following formula:

wherein P is the total matching degree, P _a For each dimension's degree of match, m is the total number of matching dimensions.

Optionally, the limiting parameters of the production lines include a maximum capacity, and the step of calculating the capacity matching degree according to the output information and the limiting parameters of the production lines includes: obtaining N predicted load capacities according to the output load capacity distribution of the N groups of output information of each unit operation; comparing the N predicted loads with the highest load, and counting the number of the N predicted loads which do not exceed the highest load; dividing the number which does not exceed the highest load by N to obtain the sub-load matching degree of each unit operation; and multiplying the sub-load capacity matching degrees of the M unit operations to obtain the load capacity matching degree.

Optionally, the limiting parameter of the production line includes a maximum collection volume, and the step of calculating the matching degree of the product collection volume according to the output information and the limiting parameter of the production line includes: obtaining N predicted product volumes according to the output volume distribution of the N groups of output information of each unit operation; comparing the N predicted product volumes with the maximum collection volume, and counting the number of products that do not exceed the maximum collection volume; dividing the number of the sub-product collection volumes which do not exceed the maximum collection volume by N to obtain the sub-product collection volume matching degree of each unit operation; and multiplying the matching degrees of the sub-product collection volumes of the M unit operations to obtain the matching degree of the product collection volume.

Optionally, the limiting parameter of the production line includes a maximum placing time, and the step of calculating the matching degree of the product placing time according to the output information and the limiting parameter of the production line includes: calculating N predicted product placement times based on the output product retention time distribution of the N sets of output information for each unit operation; comparing the N predicted product placement times with the maximum placement time, and counting the number of the products which do not exceed the maximum placement time; dividing the number which does not exceed the maximum placing time by N to obtain the matching degree of the placing time of the sub-products of each unit operation; and multiplying the matching degrees of the placing time of the sub-products of the M unit operations to obtain the matching degree of the placing time of the product.

Optionally, the predicted product placement time includes process elapsed time, resting time, and cycle time.

Optionally, the predicted product placement time is calculated using the following formula:

wherein, t _total Is the predicted product placement time, a is the step of beginning the product to enter the collection container, n is the step of ending the product placement,

it is the process of the jth step that consumes time, which is greater than or equal to>

Is the rest time of the jth step, is>

Is the round-robin time of the jth step, V ^j Is the predicted product volume, Q, of the jth step ^j Is the flow rate of the jth step,

is an estimation factor of the round-robin time of the jth step.

Optionally, the calculation method further includes: and judging whether the total matching degree meets the requirement, if so, selecting the production line to carry out project production, and if not, switching other production lines to carry out matching degree calculation or optimizing the total matching degree of the production line.

In order to solve the technical problem, the invention provides a matching degree optimization method of a process and a production line, which comprises the following steps: calculating the matching degree of each dimension of each unit operation by the calculating method, and calculating the mismatching degree of each dimension according to the matching degree of each dimension; sorting all the mismatching degrees, and taking the unit operation corresponding to the highest mismatching degree as a first optimization unit operation; locating a process bottleneck parameter in the first optimization unit operation and a first unmatched parameter range of the process bottleneck parameter; and calculating a first optimized parameter range of the process bottleneck parameter according to the first unmatched parameter range and the allowable range of the process bottleneck parameter, and optimizing the operation of the first optimization unit according to the first optimized parameter range.

Optionally, the step of locating a process bottleneck parameter in the first optimization unit operation and a first unmatched parameter range of the process bottleneck parameter comprises: constructing an optimal binary decision tree and visualizing the optimal binary decision tree; and positioning the process bottleneck parameter and the first unmatched parameter range in the first optimization unit operation according to the selection operation of the optimal binary decision tree.

Optionally, the step of constructing an optimal binary decision tree includes: establishing a data set consisting of features and targets, wherein the features comprise key process parameters of each unit operation, and the target has a value of 0 or 1; constructing a binary decision tree based on the classification tree in the CART algorithm, and selecting the binary decision tree with the highest accuracy as the optimal binary decision tree.

Optionally, the value of the target is determined by: inputting the features into the process flow model based on a Monte Carlo simulation, and calculating the degree of mismatching of the first optimization unit operation; and judging whether the mismatching degree of the first optimization unit operation is larger than a first threshold value, if so, setting the value of the target to be 1, otherwise, setting the value of the target to be 0.

Optionally, the step of locating the process bottleneck parameter in the first optimization unit operation and the first unmatched parameter range of the process bottleneck parameter according to the optimal binary decision tree comprises: and positioning a branch from a node containing the most unmatched data at the bottom of the optimal binary decision tree to a root node, and taking all the optimal segmentation features and the optimal segmentation points on the branch as the process bottleneck parameter and the first unmatched parameter range respectively.

Optionally, the optimal segmentation feature and the optimal segmentation point are determined by using a kini index.

Optionally, before constructing the binary decision tree based on the classification tree in the CART algorithm, the feature is further filtered.

Optionally, the step of screening the features comprises process principle based screening, correlation based screening and manual special rule based screening.

Optionally, the screening based on relevance comprises calculating the relevance of each feature to the target, and removing features having a relevance below a second threshold.

Optionally, the step of optimizing the first optimization unit operation according to the first optimization parameter range comprises: optimizing the numerical value of the process bottleneck parameter according to the first optimization parameter range, inputting the optimized numerical value of the process bottleneck parameter into a unit operation model, and calculating the optimization mismatching degree of the first optimization unit operation; and judging whether the difference value of the highest mismatching degree and the optimized mismatching degree is larger than or equal to a third threshold, if so, finishing optimization by the first optimization unit operation.

Optionally, the optimization method further comprises: taking the unit operation corresponding to one of the other mismatching degrees as a second optimization unit operation; locating a second unmatched parameter range of the process bottleneck parameter and the process bottleneck parameter in the second optimization unit operation; calculating a second optimized parameter range according to the second unmatched parameter range and the allowable range of the process bottleneck parameter; and if the process bottleneck parameter in the first optimization unit operation is the same as the process bottleneck parameter in the second optimization unit operation, taking the intersection of the first optimization parameter range and the second optimization parameter range as a final optimization parameter range, and optimizing the first optimization unit operation and the second optimization unit operation according to the final optimization parameter range.

In order to solve the above technical problem, the present invention provides a device for calculating a matching degree between a process flow and a production line, comprising: a memory for storing instructions executable by the processor; a processor for executing the instructions to implement the computing method as described above.

In order to solve the above technical problems, the present invention provides a matching degree optimization device for a process flow and a production line, comprising: a memory for storing instructions executable by the processor; a processor for executing the instructions to implement the optimization method as described above.

To solve the above technical problem, the present invention provides a computer-readable medium storing computer program code, which when executed by a processor implements the calculation method and the optimization method as described above.

Compared with the prior art, the invention has the following advantages:

1. compared with the traditional mode of using the full upper limit and the full lower limit of all process parameters as the input of the same process flow model, the matching degree calculation method of the process and the production line uses the input of the Monte Carlo simulation process flow model, can better simulate the randomness of the process flow, can provide more abundant probability distribution for the output of the process flow model, and can control the possibility of extreme parameter combination through the execution times of Monte Carlo simulation.

2. The matching degree calculation method of the process and the production line is characterized in that the matching degree problem of the process and the production line is transformed into a specific multi-dimension mismatching problem, and the Monte Carlo simulation process flow model is input, so that the blank that the matching degree does not have a quantitative index as a reference is filled.

3. The method for optimizing the matching degree of the process and the production line can achieve the aim of improving the matching degree of the process and the production line by limiting key process parameters on the basis of not changing the layout of the existing production line.

4. The matching degree optimization method of the process and the production line can effectively solve the problem of unmatched production lines by positioning the process bottleneck parameters causing the unmatched production lines in a decision tree-based mode and adjusting or limiting the numerical values of the process bottleneck parameters.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the principle of the invention. In the drawings:

FIG. 1 is a schematic diagram of a hierarchy of processes according to an embodiment of the present invention;

FIG. 2 is a flow chart of a process to line match calculation method according to an embodiment of the present invention;

FIG. 3 is a system block diagram of a process flow model corresponding to the process of FIG. 1;

FIG. 4 is a schematic illustration of the input and output quantities of a process flow model according to one embodiment of the present invention;

FIG. 5 is a flow chart of a process to line match optimization method according to an embodiment of the present invention;

FIG. 6 is a diagram of an embodiment of a process and line matching optimization method of the present invention;

FIG. 7 is a diagrammatic illustration of a data set for constructing an optimal binary decision tree in accordance with an embodiment of the present invention;

FIG. 8 is a schematic diagram of an optimal binary decision tree according to an embodiment of the present invention;

FIG. 9 illustrates a process bottleneck parameter and the first mismatch parameter range in a first optimization unit operation according to one embodiment of the present invention;

FIG. 10 illustrates the process bottleneck parameter and the second mismatch parameter range in the second optimization unit operation in accordance with one embodiment of the present invention;

FIG. 11 is a schematic illustration of the final optimization parameter ranges of FIGS. 9 and 10;

FIG. 12 is a schematic diagram of a process line capability visualization according to an embodiment of the present invention;

fig. 13 is a system block diagram of a matching degree calculation device for a process and a production line according to an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only examples or embodiments of the application, from which the application can also be applied to other similar scenarios without inventive effort for a person skilled in the art. Unless otherwise apparent from the context, or otherwise indicated, like reference numbers in the figures refer to the same structure or operation.

As used in this application and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.

The relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise. Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description. Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate. In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

It should be noted that the terms "first", "second", and the like are used to define the components, and are only used for convenience of distinguishing the corresponding components, and the terms have no special meanings unless otherwise stated, so that the scope of the present application is not to be construed as being limited. Further, although the terms used in the present application are selected from publicly known and used terms, some of the terms mentioned in the specification of the present application may be selected by the applicant at his or her discretion, the detailed meanings of which are described in relevant parts of the description herein. Further, it is required that the present application is understood not only by the actual terms used but also by the meaning of each term lying within.

Flow charts are used herein to illustrate operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, various steps may be processed in reverse order or simultaneously. Meanwhile, other operations are added to or removed from these processes.

A project process often includes multiple unit operations, each including multiple process steps. Unit operations are a general term for a series of basic operations performed in the chemical and other process industries to comminute, convey, heat, cool, mix, and separate materials to produce desired physical changes. The various unit operations are based on different physical and chemical principles, and corresponding equipment is applied to achieve respective process purposes. For example, distillation can be used for the purpose of separating components in a liquid mixture or purifying a component according to the difference of volatilization capacities of the components in the liquid mixture. FIG. 1 is a schematic diagram of a hierarchy of processes according to an embodiment of the present application. As shown in fig. 1, process 1 comprises unit operation 11, unit operation 12, unit operation 13, unit operation 14, unit operation 15, and unit operation 16. Unit operation 13 includes process step 131, process step 132, process step 133, process step 134, and process step 135. Each unit operation may have one or more Critical process parameters (CCPs).

A manufacturing line is a combination of equipment used to perform a process to produce. The product is, for example, a chemical product, in particular a pharmaceutical. The production line has limiting parameters determined by its design characteristics, such as load capacity, product collection container volume, and the time limit for product stability (hold time limit). These limiting parameters do not necessarily satisfy the key process parameters of each unit operation in the process. Therefore, the matching degree of the process and the production line needs to be calculated.

Fig. 2 is a flowchart of a method for calculating a matching degree between a process and a production line according to an embodiment of the present invention. As shown in fig. 2, the method 200 for calculating the matching degree between a process and a production line includes the following steps:

step S21: obtaining key process parameters of each unit operation in M unit operations forming the process flow and probability distribution of each key process parameter, wherein M is an integer greater than 1;

step S22: establishing a process flow model, wherein the process flow model comprises M unit operation models;

step S23: inputting the key process parameters of each unit operation and the corresponding probability distribution into a process flow model based on Monte Carlo simulation to obtain output information, wherein the output information comprises output load capacity distribution, output volume distribution and output product retention time distribution;

step S24: and calculating the total matching degree of the process and the production line according to the output information and the limiting parameters of the production line.

The following describes steps S21 to S24 in detail.

In step S21, the key process parameters of each unit operation may be obtained according to expert knowledge, for example, a key process parameter template of each unit operation is preset according to expert knowledge and stored, and the key process parameters of each unit operation can be obtained by calling the template. For the project which lacks complete process information in the early stage, the method can calculate the production line with the highest matching rate by means of the currently known information so as to improve the success rate. Key process parameters for each unit operation may also be obtained by a distribution fit of historical data. Obtaining the probability distribution for each key process parameter may match the most conforming distribution type among the plurality of mathematical distributions based on maximum likelihood estimation. In the embodiment, the probability distribution of each key process parameter is obtained through distribution fitting, and compared with the method based on personal experience, the method can better support the calculation of the probability distribution in the subsequent process flow simulation process and better endow the simulation result with statistical significance.

In some embodiments, the step of obtaining key process parameters for each unit operation from a distribution fit of historical data comprises: and judging whether the project to be produced has complete process flow information, if so, fitting actual data distribution in the process flow information to obtain the key process parameters of each unit operation, and otherwise, fitting historical data distribution of similar projects to obtain the key process parameters of each unit operation.

In step S22, a process flow model composed of unit operation models is established based on the mass conservation law. Wherein the number of unit operation models corresponds to the number of unit operations. FIG. 3 is a system block diagram of a process flow model corresponding to the process of FIG. 1. As shown in FIG. 3, the process flow model 300 includes a unit operation model 31, a unit operation model 32, a unit operation model 33, a unit operation model 34, a unit operation model 35, and a unit operation model 36.

In step S23, the step of inputting the key process parameters and their corresponding probability distributions for each unit operation into the process flow model based on the monte carlo simulation comprises: and inputting the key process parameters of each unit operation and the probability distribution of the key process parameters into corresponding unit operation models based on Monte Carlo simulation, and outputting a group of output information by each unit operation model. And repeating the step N times, outputting N groups of output information by each unit operation model, and outputting N × M groups of output information by the process flow model, wherein N is a positive integer greater than 1. FIG. 4 is a schematic illustration of the input and output quantities of a process flow model according to one embodiment of the present invention. As shown in fig. 4, the input quantities to the process flow model 400 are the key process parameters and the probability distribution of the key process parameters for each unit operation. The probability distributions for the key process parameters include key process parameter 1 distribution 411, key process parameter 2 distribution 412, and key process parameter 3 distribution 413. The input quantities to the process flow model also include an input volume distribution 414 and an input mass distribution 415. The input volume distribution 414 and the input mass distribution 415 are the initial inputs or outputs of the last unit operation model. The process flow model 400 outputs a set of output information including, but not limited to, an output volume distribution 421, an output mass distribution 422, an output load distribution 423, and an output product retention time distribution 424.

The possibility of extreme parameter combinations can be controlled by controlling the number of Monte Carlo simulation executions N, which is closer to production practice. The output of a process flow model using the monte carlo method will have a corresponding numerical probability distribution. The numerical value probability distribution can be used for more accurately estimating the output in the process flow, including but not limited to upper and lower limits, expected values and the like, and provides a quantitative index based on probability calculation for the subsequent production line matching degree.

In step S24, the step of calculating the total matching degree of the process and the production line based on the output information and the limiting parameters of the production line is as follows. Firstly, the matching degree of the process flow and the production line in multiple dimensions is calculated according to the output information and the limiting parameters of the production line. Here, the degree of matching in a plurality of dimensions includes, but is not limited to, a degree of matching of a load amount, a degree of matching of a product collection volume, and a degree of matching of a product placement time. And then, calculating the total matching degree of the process and the production line according to the matching degrees of the multiple dimensions. Calculating the total matching degree of the process and the production line by the following formula:

wherein P is the total matching degree, P _a For each dimension's degree of match, m is the total number of matched dimensions.

The loading can be subdivided into volumetric and mass loadings, as is common in the calculation of unit operations associated with chromatography and filtration. When the product volume and mass input by the unit operation is out of the capacity range, product that exceeds the highest capacity is lost. In some embodiments, the limiting parameters of the manufacturing line include a maximum capacity, and the step of calculating the capacity match based on the output information and the limiting parameters of the manufacturing line includes:

(1) Obtaining N predicted load quantities according to the output load quantity distribution of the N groups of output information of each unit operation;

(2) Comparing the N predicted loading capacities with the highest loading capacity respectively, and counting the number exceeding the highest loading capacity;

(3) And dividing the number which does not exceed the highest load by N to obtain the sub-load matching degree of each unit operation. For example, it can be expressed by the following formula:

wherein, P _i Degree of sub-capacity match, N, for the ith unit operation _i Predicting the number of load exceeds the maximum load for the ith unit operation, N-N _i For the number of predicted loads in the ith unit operation that do not exceed the maximum load, N is the number of monte carlo simulations, which is equal to the number of output bursts.

(4) And multiplying the sub-load matching degrees of the M unit operations to obtain the load matching degree. Can be expressed by the following formula:

wherein, P ₁ For degree of load matching, P _i The sub-capacity matching degree of the ith unit operation is, and M is the number of unit operations.

In some embodiments, the limiting parameters of the manufacturing line include a customized first load, the first load being less than the highest load. The first loading has a safety margin as compared to the highest loading. The step of calculating the load matching degree according to the output information and the limiting parameters of the production line comprises the following steps: obtaining N predicted load capacities according to N groups of output information of each unit operation; comparing the N predicted loads with the first load, and counting the number of the N predicted loads which do not exceed the first load; and dividing the number which does not exceed the first loading capacity by N to obtain the sub-loading capacity matching degree of each unit operation. And multiplying the sub-load matching degrees of the M unit operations to obtain the load matching degree.

Due to limitations in the design of the production line, there are limitations in the size and number of product collection containers within the space in which the unit operations are performed, which is referred to as product collection container volume. The maximum collection volume in the space in which it is located is the sum of the volumes of all placeable containers. When the product volume output by a unit operation exceeds the maximum collection volume supported by the space, the excess product volume is lost. In some embodiments, the limiting parameters of the production line include a maximum collection volume, and the step of calculating a product collection volume match based on the output information and the limiting parameters of the production line includes:

(1) Obtaining N predicted product volumes according to output volume distribution of N groups of output information of each unit operation;

(2) Comparing the N predicted product volumes with the maximum collection volume, and counting the number of the product volumes which do not exceed the maximum collection volume;

(3) And dividing the number which does not exceed the maximum collection volume by N to obtain the matching degree of the collection volume of the sub-products of each unit operation. For example, it can be expressed by the following formula:

wherein, P _j Collecting volume match, N, for sub-product of jth unit operation _j Predicting the number of product volumes exceeding the maximum collection volume, N-N, for the jth unit operation _j For the number of predicted product volumes in the jth unit operation that do not exceed the maximum collection volume, N is the number of Monte Carlo simulations, which is equal to the number of output message groups.

(4) And multiplying the matching degrees of the sub-product collection volumes of the M unit operations to obtain the matching degree of the product collection volume. Can be expressed by the following formula:

wherein, P ₂ For product collection volume matching, P _j The sub-product collection volume match for the jth unit operation, and M is the number of unit operations.

In some embodiments, the limiting parameters of the production line include a customized first collection volume, the first collection volume being less than the maximum collection volume. The first collection volume has a safety margin compared to the maximum collection volume. The step of calculating the product collection volume matching degree according to the output information and the limiting parameters of the production line comprises the following steps: obtaining N predicted product volumes from the N sets of output information for each unit operation; comparing the N predicted product volumes with the first collection volume, and counting the number of the predicted product volumes which do not exceed the first collection volume; and dividing the number of the sub-products which do not exceed the first collecting volume by N to obtain the matching degree of the collecting volume of the sub-products of each unit operation, and multiplying the matching degrees of the collecting volumes of the sub-products of the M unit operations to obtain the matching degree of the collecting volume of the product.

The time limit for product stability retention (hold time limit) is the maximum time of standing that a product will be tested for chemical stability during the process development phase and determine the stability. In a unit operation, the timing of the holding time is started from the first time of a process step in which a product enters the product collection container, and the timing of the holding time is ended when the product enters the next process step. One unit operation involves a plurality of process steps. The time consumed by each process step consists of a process-consumed time and a standing time portion. The total time of unit operation is not only the sum of the time consumed by the process steps, but also a part of the time used for the rotation among the process steps, and the part of the time can be obtained by carrying out statistics according to the historical data of unit operations of different types executed in different production lines and converting the statistics into an Estimation factor (Estimation factor). In some embodiments, the limiting parameters of the production line include a maximum placement time, and the step of calculating the product placement time match based on the output information and the limiting parameters of the production line includes:

(1) N predicted product placement times are calculated from the output product retention time distribution of the N sets of output information for each unit operation. The predicted product placement time includes process elapsed time, resting time, and cycle time. Wherein the process elapsed time can be obtained by dividing the predicted volume by the flow rate. The predicted product residence time is calculated using the following formula:

Is the resting time of the jth step,

is an estimation factor of the round-robin time of the jth step. Wherein the value of the evaluation factor is determined by the actual conditions of the production line together with the type of unit operation involved and is calculated on the basis of historical data. In the process stepSome of the process steps are pre-treatment, which is independent of the product, and the pre-treatment steps are assumed to be process steps 1 to 3. At process step 4, product is introduced into the collection container and product retention time is timed. Similarly, some of the process steps are post-treatments, which are product independent, and are assumed to be process step 21. The actual retention time of the product is then the time of process steps 4 to 20. .

(2) Comparing the N predicted product placement times with the maximum placement time, and counting the number of the products which do not exceed the maximum placement time;

(3) And dividing the number which does not exceed the maximum placing time by N to obtain the matching degree of the placing time of the sub-products of each unit operation. For example, it can be expressed by the following formula:

wherein, P _k Collecting volume match for k unit operation sub-product, N _k Predicting the number of product placement times exceeding the maximum placement time, N-N, for the kth unit operation _k For the number of predicted product placement times in the kth unit operation that do not exceed the maximum placement time, N is the number of monte carlo simulations, which is equal to the number of output message groups.

(4) And multiplying the matching degrees of the placing time of the sub-products of the M unit operations to obtain the matching degree of the placing time of the product. Can be expressed by the following formula:

wherein, P ₃ For the degree of adaptation of the product standing time, P _k The sub-product placement time matching degree for the kth unit operation, and M is the number of unit operations.

In some embodiments, the limiting parameters of the production line include a custom first placement time, the first placement time being less than the maximum placement time. The first placement time has a safety margin compared to the maximum placement time. The step of calculating the matching degree of the product placement time according to the output information and the limiting parameters of the production line comprises the following steps: calculating N predicted product placement times according to the N groups of output information of each unit operation; comparing the N predicted product placing times with the first placing time, and counting the number of the products which do not exceed the first placing time; dividing the number of the sub-products which do not exceed the first placing time by N to obtain the matching degree of the placing time of the sub-products of each unit operation; and multiplying the matching degrees of the placing time of the sub-products of the M unit operations to obtain the matching degree of the placing time of the product.

In some embodiments, the method for calculating the matching degree of the process and the production line further comprises the steps of: and judging whether the total matching degree meets the requirement, if so, selecting the production line to carry out project production, and if not, switching other production lines to carry out matching degree calculation or optimizing the total matching degree of the production line.

Compared with the traditional mode of using the full upper limit and the full lower limit of all process parameters as the input of the same process flow model, the method for simulating the process flow model by using the Monte Carlo can better simulate the randomness of the process, the output of the process flow model can provide probability distribution with richer information, and the possibility of extreme parameter combination can be controlled by the execution times of the Monte Carlo simulation. The method for calculating the matching degree of the process and the production line in the embodiment is used for imaging the problem of the matching degree of the process and the production line to a specific problem of mismatching of multiple dimensions, and fills a blank that the matching degree does not have a quantitative index as a reference by combining the input of a Monte Carlo simulation process flow model.

Fig. 5 is a flowchart of a matching degree optimization method of a process and a production line according to an embodiment of the present invention. As shown in fig. 5, the method 500 for optimizing the matching degree between the process and the production line includes the following steps:

step S51: the matching degree of each dimension of each unit operation is calculated by the process and production line matching degree calculation method, and the mismatching degree of each dimension is calculated according to the matching degree of each dimension.

Step S52: and sequencing all the mismatching degrees, and taking the unit operation corresponding to the highest mismatching degree as a first optimization unit operation.

Step S53: a first mismatch parameter range of the bottleneck parameter and the process bottleneck parameter in the first optimization unit operation are located.

Step S54: and calculating a first optimized parameter range of the process bottleneck parameter according to the first unmatched parameter range and the allowable range of the process bottleneck parameter, and optimizing the operation of the first optimization unit according to the first optimized parameter range.

FIG. 6 shows an embodiment of the method for optimizing the matching degree between the process and the production line according to the present invention. The following describes steps S51 to S54 in detail with reference to fig. 5 and 6.

In step S51, P is assumed _a For the matching degree of each dimension, calculating the mismatching degree of each dimension according to the matching degree of each dimension can be represented by a formula: 1-P _a And (4) obtaining. The calculated mismatch for each dimension of each unit operation is shown in fig. 6, and the process flow model 600 includes a unit operation model 61, a unit operation model 62, a unit operation model 63, a unit operation model 64, a unit operation model 65, and a unit operation model 66. Each unit operation model includes a product collection volume mismatch, a load mismatch, and a product placement time mismatch.

In step S52, all the mismatches are sorted from high to low, and the sorting result is: the product collection volume mismatch of the unit operation model 65 > the product collection volume mismatch of the unit operation model 66 > the product collection volume mismatch of the unit operation model 64. The mismatch for the other dimensions is 0. According to the sorting result, taking the unit operation 5 corresponding to the unit operation model 65 as a first optimization unit operation, and recording as S1; taking the unit operation 6 corresponding to the unit operation model 66 as a second optimization unit operation, which is marked as S2; the unit operation 4 corresponding to the unit operation model 64 is referred to as a third optimization unit operation as S3. First, the first optimization unit operation S1 is optimized. A common optimization is to change the production line layout and add collection containers. However, in actual production, it is difficult to change the layout of the production line due to space limitation or the limitation of the number of devices. The present embodiment describes optimizing the degree of mismatch without changing the existing production line layout. Specifically, by positioning the bottleneck parameters causing the production line mismatch in a decision tree-based manner, the problem of the production line mismatch can be effectively solved by adjusting or limiting the values of the bottleneck parameters.

In step S53, the step of locating the bottleneck parameter and the first unmatched parameter range of the bottleneck parameter in the first optimization unit operation is as follows. First, an optimal binary decision tree is constructed and visualized. Secondly, the process bottleneck parameter and the first unmatched parameter range in the first optimization unit operation are located according to the selection operation of the optimal binary decision tree.

In one example, the step of constructing an optimal binary decision tree includes:

a. a data set is created consisting of features including key process parameters for each unit operation and targets with values of 0 or 1. The value of the target is determined by: inputting the characteristics into a process flow model based on Monte Carlo simulation, and calculating the mismatching degree of the operation of the first optimization unit; and judging whether the mismatching degree of the first optimization unit operation is larger than a first threshold value, if so, setting the value of the target to be 1, otherwise, setting the value of the target to be 0. FIG. 7 is a diagrammatic illustration of a data set for constructing an optimal binary decision tree in accordance with an embodiment of the present invention. As shown in fig. 7, data set 700 includes feature X and object Y. Where the feature X is comprised of the key process parameters for each unit operation. The number of features is determined by the number of key process parameters corresponding to all unit operations involved in the process, and the number of rows of features X is equal to the number of monte carlo simulations. I.e. a in the feature X ₁₁ Is a key process parameter of unit operation 1, a _1n Is a key process parameter of unit operation n, a _mn Is the key process parameter for unit operation n at the mth Monte Carlo simulation. Each row of values in the object Y corresponds to a monte carlo random parameter combination of each row of the feature X, and the value of a specific matching problem (such as the first optimization unit S1) is assigned, the value of the matching value is 0, and the value of the mismatching value is 1.

In some embodiments, in order to achieve faster training and higher accuracy, and to exclude interference of irrelevant features with the binary decision tree, before constructing the binary decision tree based on the classification tree in the CART algorithm, the feature is further filtered. The step of screening features includes, but is not limited to, process-principle based screening, correlation-based screening, and manual special rule based screening. By process-principle-based screening is meant screening of the features based on mechanism and process principle, e.g. when the unit operation 5 is the first optimized unit operation S1, all key process parameters of the unit operation 6 may be removed from the features X. Given that the first optimization unit operation S1 to be optimized is related to the collection volume, key process parameters in all unit operations not participating in the collection volume calculation process, such as flow rates for calculating the settling time, etc., can be removed from the feature X. The screening based on the correlation includes calculating a correlation of each feature with the target, and removing features having a correlation below a second threshold. Specifically, the characteristic X and the target Y are put into a Logistic regression (Logistic regression) model, all the remaining key process parameters and targets are subjected to hypothesis test, and the p-value is calculated. Removing a critical process parameter with a p-value greater than a second threshold (typically 0.05) may remove the portion of the critical process parameter because it represents no statistical correlation between the critical process parameter and the target Y. The correlation is not limited to a linear correlation and is not subject to a normal distribution. The manual special rule-based screening may be to review the remaining key process parameters and manually remove the key process parameters that are difficult to control or cannot be modified after the process principle-based screening and the correlation-based screening are completed.

After feature screening is completed, a training set and a validation set are created from the data set. The training set is made, the balance condition in the training set is noticed, the accuracy of the decision tree is improved, and the bottleneck process parameters and the corresponding threshold values are positioned better. Specifically, in 100000 monte carlo simulations, the probability of mismatch of unit operation 5 is 3.5%, i.e., the target Y value of 3500 rows is 1. At this time, 3500 rows should be selected from the remaining 96500 rows of feature combinations with target Y value of 0 to form a 7000 rows training set, and the ratio of 0 to 1 is made to approach 50, instead of using the training set containing all 100000 rows, otherwise the binary decision tree can obtain 96.5% accuracy even if all feature combinations are predicted to be 0, which obviously cannot define the accurate bottleneck process parameters.

b. Constructing a binary decision tree based on a Classification tree in a CART (Classification and regression tree) algorithm, and selecting the binary decision tree with the highest accuracy as an optimal binary decision tree. The binary decision tree uses a Gini index (Gini index) to select and adopt the Gini index to determine the optimal segmentation characteristics and the optimal segmentation points, and K-fold cross validation (K fold cross-validation) is used for testing the robustness of the model. The method comprises the following specific steps:

i. segmenting the screened data set into k subsets, using k-1 subsets as a training set of the model, and using 1 subset as a verification set of the model;

recursively building a binary decision tree from the training set for each node starting from the root node by:

1) And (4) setting the data set entering the node as T, and calculating the Gini index of each feature to the data set T according to each feature X and each possible value X of each feature. Selecting the feature X with the smallest Gini index _opt And its value x _opt And as the optimal segmentation feature and the optimal segmentation point, generating two sub-nodes from the current node, segmenting the data set T into data sets T1 and T2 according to the condition that the numerical value of the optimal segmentation feature is greater than the optimal segmentation point or less than or equal to the optimal segmentation point, and distributing the data sets T into the two sub-nodes.

2) The operations of the previous steps are repeated for each child node and further segmentation is performed until the decision tree depth, typically at 3 levels, is controlled by the minimum number of samples required under the leaf node, the maximum number of leaf nodes, or the minimum reduction in purity required to perform the segmentation.

And iii, replacing the subset serving as the verification set with one subset in the training set, repeating ii, and outputting the average prediction accuracy after k times of execution.

And iv, adjusting the hyper-parameter, repeating the steps i, ii and iii, comparing the average prediction accuracy, and selecting the binary decision tree with the highest accuracy for visualization.

After visually presenting the binary decision tree to the user, the user can select an optimal binary decision tree on the binary decision tree. The method of this embodiment may locate the process bottleneck parameter and the first mismatch parameter range in the first optimization unit operation based on the selection operation on the optimal binary decision tree. Specifically, a branch from a node containing the most unmatched data at the bottom of the optimal binary decision tree to the root node is located, and all the optimal segmentation features and the optimal segmentation points on the branch are respectively used as a process bottleneck parameter and a first unmatched parameter range. FIG. 8 is a diagram of an optimal binary decision tree according to an embodiment of the present invention. As shown in fig. 8, if the branch from the node containing the most unmatched data in the bottom of the optimal binary decision tree 800 to the root node is a branch with a > 40 and B > 2.85, the optimal segmentation features a and B are both process bottleneck parameters, the first unmatched parameter range is marked as U1, and the value of U1 is a > 40 and B > 2.85.

Continuing back to fig. 1, in step S54, a first optimized parameter range of the bottleneck parameter is calculated according to the first unmatched parameter range and the allowable range of the bottleneck parameter. Specifically, the first optimized parameter range of the bottleneck parameter is a difference set between the allowable range of the bottleneck parameter and the first unmatched parameter range. FIG. 9 illustrates the process bottleneck parameter and first mismatch parameter ranges in a first optimization unit operation according to one embodiment of the present invention. As shown in fig. 9, the process bottleneck parameters of the first optimization unit operation S1 are a and B. The allowable range of the bottleneck parameter A is [0, 50], and the allowable range of the bottleneck parameter B is [2,5]. The value of the first mismatch parameter range U1 is { A > 40 and B > 2.85}, then the value of the first optimized parameter range G1 is {0 ≦ A ≦ 40 and 2 ≦ B ≦ 2.85}.

In some embodiments, the step of optimizing the first optimization unit operation according to the first optimization parameter range is as follows. Firstly, optimizing the numerical value of the process bottleneck parameter according to a first optimization parameter range, inputting the optimized numerical value of the process bottleneck parameter into a unit operation model, and calculating the optimization mismatching degree of the first optimization unit operation. Then, whether the difference between the highest mismatch and the optimized mismatch is greater than or equal to a third threshold is judged, and if so, the first optimization unit completes the optimization. In other words, it is determined whether there is a significant decrease between the optimized mismatch and the highest mismatch (not optimized), and if there is a significant decrease, the optimization is successful, otherwise, it is necessary to return to the feature screening step to re-screen the features. In practice, it is found that in most cases, the mismatch can be greatly reduced by limiting a plurality of bottleneck process parameters at the same time, optimizing the numerical value of the process bottleneck parameter according to the first optimized parameter range G1, and inputting the optimized numerical value of the process bottleneck parameter into the unit operation model, wherein the mismatch of the first optimized unit operation S1 is reduced from 3.5% to 0.07%.

In some embodiments, the method for optimizing the matching degree of the process and the production line further comprises the following steps. First, the unit operation corresponding to one of the other mismatch degrees is taken as the second optimized unit operation. Second, a second mismatch parameter range of the bottleneck parameter and the process bottleneck parameter in the second optimization unit operation is located. Then, a second optimized parameter range is calculated according to the second unmatched parameter range and the allowable range of the process bottleneck parameter. For example, as shown in FIG. 6, the product collection volume mismatch of the unit operation model 66 is the second highest mismatch. The unit operation 6 corresponding to the unit operation model 66 is regarded as a second optimization unit operation and is denoted as S2. And constructing an optimal binary decision tree for the second optimization unit operation S2, and positioning the process bottleneck parameters in the second optimization unit operation and a second mismatch parameter range of the process bottleneck parameters according to the optimal binary decision tree. FIG. 10 illustrates the process bottleneck parameter and second mismatch parameter ranges in a second optimization unit operation in accordance with one embodiment of the present invention. As shown in fig. 10, the process bottleneck parameters of the second optimization unit operation S2 are a, B and D. The allowable range of the bottleneck parameter A is [0, 50], the allowable range of the bottleneck parameter B is [2,5], and the allowable range of the bottleneck parameter D is [0,5]. The value of the second mismatch parameter range U2 is { A > 30 and B > 4.25 and D > 2.24}, then the value of the second optimized parameter range G2 is { 0. Ltoreq. A.ltoreq.30 and 2. Ltoreq. B.ltoreq.4.25 and 2. Ltoreq. D.ltoreq.2.24 }. And if the process bottleneck parameter in the first optimization unit operation is the same as the process bottleneck parameter in the second optimization unit operation, taking the intersection of the first optimization parameter range and the second optimization parameter range as a final optimization parameter range, and optimizing the first optimization unit operation and the second optimization unit operation according to the final optimization parameter range. Fig. 11 is a schematic diagram of the final optimization parameter ranges of fig. 9 and 10. As shown in fig. 11, the final optimized parameter range is denoted as Gfinal, and the value of Gfinal is the intersection of the first optimized parameter range G1 and the second optimized parameter range G2. The value of Gfinal is {0 ≦ A ≦ 30, 2 ≦ B ≦ 2.85, and 2 ≦ D ≦ 2.24}.

In some embodiments, when there is no intersection between the first optimization parameter range G1 and the second optimization parameter range G2, it indicates that the first optimization unit operation S1 and the second optimization unit operation S2 cannot be optimized simultaneously. The first optimization parameter range G1 may be used as the final optimization parameter range Gfinal to optimize the first optimization unit operation S1 preferentially, or the conflicting key process parameters may be deleted from the feature X, so that the first optimization unit operation S1 and the second optimization unit operation S2 are optimized simultaneously by other key process parameters.

The matching degree optimization method of the process and the production line of the embodiment optimizes the mismatching degree by limiting key process parameters on the basis of not changing the layout of the existing production line, and achieves the purpose of improving the matching degree of the process and the production line. The technology bottleneck parameters causing the unmatched production lines are positioned in a decision tree-based mode, and the unmatched production line problem can be effectively solved by adjusting or limiting the values of the technology bottleneck parameters.

In some embodiments, whether the project using the platform process can be adapted to the target production line can be quickly determined according to the mismatch parameter range of the key process bottleneck parameter and the process bottleneck parameter of the process on the production line, and the key process parameter selected by the experiment of the project developed by using the platform process can be limited and guided under the condition of determining the target production line so as to ensure that the capacity of the production line can be completely matched. FIG. 12 is a schematic diagram illustrating the visualization of the production line capability according to one embodiment of the present invention. As shown in fig. 12, due to different production line designs, the ranges of mismatch parameters of the key bottleneck parameters and the bottleneck parameters obtained by the same process on different production lines may be different. The process 1 has a mismatch parameter range 121 of the key bottleneck parameter and the bottleneck parameter on the production line P1, and the process 1 has a mismatch parameter range 122 of the key bottleneck parameter and the bottleneck parameter on the production line P2. When the project of the process 1 is planned to be used, the production line P1 and the production line P2 can be quickly matched only by comparing the bottleneck process parameters, and the production line with the bottleneck process parameter value range meeting the project requirement is selected for production. The method and the device can analyze the key process parameters of each fixed process for each existing production line, and the analysis result can be like production line capability, so that the aim of assisting in quickly matching the process or guiding process development is fulfilled.

Fig. 13 is a system block diagram of a process-to-production-line matching degree calculation apparatus 1300 (hereinafter, referred to as the calculation apparatus 1300) according to an embodiment of the invention. Referring to fig. 13, the computing device 1300 may include an internal communication bus 1301, a processor 1302, a Read Only Memory (ROM) 1303, a Random Access Memory (RAM) 1304, and a communication port 1305. When implemented on a personal computer, the computing device 1300 may also include a hard disk 1306. The internal communication bus 1301 may enable data communication among the components of the computing device 1300. The processor 1302 may make the determination and issue the prompt. In some embodiments, processor 1302 may be comprised of one or more processors. The communication port 1305 may enable data communication between the computing device 1300 and the outside. In some embodiments, the computing device 1300 may send and receive information and data from a network through the communication port 1305. The computing device 1300 may also include various forms of program storage units and data storage units, such as a hard disk 1306, read Only Memory (ROM) 1303 and Random Access Memory (RAM) 1304, capable of storing various data files used in computer processing and/or communications, and possibly program instructions executed by the processor 1302. The processor executes these instructions to implement the main parts of the method. The results processed by the processor are communicated to the user device through the communication port and displayed on the user interface. The above-described operation method may be implemented as a computer program, stored in the hard disk 1306, and loaded into the processor 1302 to be executed, so as to implement the matching degree calculation method and optimization method of the process and the production line of the present application.

The invention also comprises a computer readable medium having stored thereon computer program code which, when executed by a processor, implements the foregoing process to line match calculation and optimization methods.

The process matching degree calculation method and optimization method for a production line may be embodied as a computer program, and may be stored in a computer-readable storage medium as an article of manufacture. For example, computer-readable storage media can include but are not limited to magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact Disk (CD), digital Versatile Disk (DVD)), smart cards, and flash memory devices (e.g., electrically Erasable Programmable Read Only Memory (EPROM), card, stick, key drive). In addition, various storage media described herein can represent one or more devices and/or other machine-readable media for storing information. The term "machine-readable medium" can include, without being limited to, wireless channels and various other media (and/or storage media) capable of storing, containing, and/or carrying code and/or instructions and/or data.

The invention also provides a matching degree optimizing device of the process and the production line, and the structure of the matching degree optimizing device of the process and the production line can refer to the matching degree calculating device 1300 of the process and the production line, which is not described herein again.

Aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. The processor may be one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital signal processing devices (DAPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or a combination thereof. Furthermore, aspects of the present application may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media. For example, computer-readable media can include, but are not limited to, magnetic storage devices (e.g., hard disk, floppy disk, magnetic tape \8230;), optical disks (e.g., compact disk CD, digital versatile disk DVD \8230;), smart cards, and flash memory devices (e.g., card, stick, key drive \8230;).

The computer-readable medium may comprise a propagated data signal with the computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, and the like, or any suitable combination. The computer readable medium can be any computer readable medium that can communicate, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable medium may be propagated over any suitable medium, including radio, electrical cable, fiber optic cable, radio frequency signals, or the like, or any combination of the preceding.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single disclosed embodiment.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing disclosure is by way of example only, and is not intended to limit the present application. Various modifications, improvements and adaptations to the present application may occur to those skilled in the art, though not expressly described herein. Such alterations, modifications, and improvements are intended to be suggested herein and are intended to be within the spirit and scope of the exemplary embodiments of this application.

Also, this application uses specific language to describe embodiments of the application. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the present application is included in at least one embodiment of the present application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the present application may be combined as appropriate.

Although the present application has been described with reference to the present specific embodiments, it will be recognized by those skilled in the art that the foregoing embodiments are merely illustrative of the present application and that various changes and substitutions of equivalents may be made without departing from the spirit of the application, and therefore, it is intended that all changes and modifications to the above-described embodiments that come within the spirit of the application fall within the scope of the claims of the application.

Claims

1. A matching degree calculation method of a process and a production line is characterized by comprising the following steps:

obtaining a key process parameter of each unit operation in M unit operations forming the process flow and the probability distribution of each key process parameter, wherein M is an integer greater than 1;

establishing a process flow model, wherein the process flow model comprises M unit operation models;

inputting key process parameters of each unit operation and corresponding probability distribution into the process flow model based on Monte Carlo simulation to obtain output information, wherein the output information comprises output load capacity distribution, output volume distribution and output product retention time distribution;

and calculating the total matching degree of the process and the production line according to the output information and the limiting parameters of the production line.

2. The method of claim 1, wherein the key process parameters for each unit operation are obtained from expert knowledge or a distribution fit of historical data.

3. The method of claim 2, wherein the step of obtaining key process parameters for each unit operation from a distribution fit of historical data comprises:

judging whether the project to be produced has complete process flow information, if so, acquiring the key process parameters of each unit operation from the historical process flow information of the project, and if not, acquiring the key process parameters of each unit operation from the historical process flow information of the project similar to the project.

4. The method of claim 1, wherein the step of obtaining a probability distribution for each key process parameter comprises matching a most conforming distribution type among the plurality of mathematical distributions based on maximum likelihood estimation.

5. The method of claim 1, wherein inputting the key process parameters and their corresponding probability distributions for each unit operation into the process flow model based on a monte carlo simulation comprises:

step a: inputting the key process parameters of each unit operation and the probability distribution of the key process parameters into corresponding unit operation models based on Monte Carlo simulation, and outputting a group of output information by each unit operation model;

step b: and (c) repeatedly executing the step a for N times, outputting N groups of output information by each unit operation model, and outputting N × M groups of output information by the process flow model, wherein N is a positive integer greater than 1.

6. The method of claim 5, wherein the step of calculating an overall process-to-line match based on the output information and the line constraint parameters comprises:

calculating the matching degree of the process flow and multiple dimensions of the production line according to the output information and the limiting parameters of the production line;

and calculating the total matching degree of the process and the production line according to the matching degrees of the plurality of dimensions.

7. The method of claim 6, wherein the degree of matching in the plurality of dimensions comprises a capacity degree of matching, a product collection volume degree of matching, and a product placement time degree of matching.

8. The method of claim 6, wherein the total match of the process to the production line is calculated by the formula:

9. The method of claim 7, wherein the constraint parameters of the production lines include a maximum load, and the step of calculating the load match based on the output information and the constraint parameters of the production lines includes:

obtaining N predicted load quantities according to the output load quantity distribution of the N groups of output information of each unit operation;

comparing the N predicted loads with the highest load, and counting the number of the N predicted loads which do not exceed the highest load;

dividing the number which does not exceed the highest load by N to obtain the sub-load matching degree of each unit operation;

and multiplying the sub-load capacity matching degrees of the M unit operations to obtain the load capacity matching degree.

10. The method of claim 7, wherein the production line limiting parameters include a maximum collection volume, and wherein calculating the product collection volume match based on the output information and the production line limiting parameters includes:

obtaining N predicted product volumes according to the output volume distribution of the N groups of output information of each unit operation;

comparing the N predicted product volumes with the maximum collection volume, and counting the number of products that do not exceed the maximum collection volume;

dividing the number of the sub-product collection volumes which do not exceed the maximum collection volume by N to obtain the sub-product collection volume matching degree of each unit operation;

and multiplying the matching degrees of the sub-product collection volumes of the M unit operations to obtain the matching degree of the product collection volume.

11. The method of claim 7, wherein the production line limiting parameters include a maximum placement time, and wherein calculating the product placement time match based on the output information and production line limiting parameters comprises:

calculating N predicted product placement times based on the output product retention time distribution of the N sets of output information for each unit operation;

comparing the N predicted product placement times with the maximum placement time, and counting the number of the products which do not exceed the maximum placement time;

dividing the number of the sub-products which do not exceed the maximum placing time by N to obtain the matching degree of the placing time of the sub-products of each unit operation;

and multiplying the matching degrees of the placing time of the sub-products of the M unit operations to obtain the matching degree of the placing time of the product.

12. The method of claim 11, wherein the predicted product placement time comprises process elapsed time, resting time, and cycle time.

13. The method of claim 12, wherein the predicted product placement time is calculated using the formula:

Is the rest time of the jth step,

is the round-robin time of the jth step, V ^j Is the predicted product volume, Q, of the jth step ^j Is the flow rate of the jth step, f _e ^j Is an estimation factor of the round-robin time of the jth step.

14. The method of claim 1, further comprising:

and judging whether the total matching degree meets the requirement, if so, selecting the production line to carry out project production, and if not, switching other production lines to carry out matching degree calculation or optimizing the total matching degree of the production line.

15. A matching degree optimization method of a process and a production line is characterized by comprising the following steps:

calculating a degree of matching for each dimension of each unit operation by a method according to any one of claims 6 to 13, calculating a degree of mismatching for each dimension based on the degree of matching for each dimension;

sorting all the mismatching degrees, and taking the unit operation corresponding to the highest mismatching degree as a first optimization unit operation;

locating a process bottleneck parameter in the first optimization unit operation and a first unmatched parameter range of the process bottleneck parameter;

and calculating a first optimized parameter range of the process bottleneck parameter according to the first unmatched parameter range and the allowable range of the process bottleneck parameter, and optimizing the operation of the first optimizing unit according to the first optimized parameter range.

16. The method of claim 15, wherein the step of locating a process bottleneck parameter in the first optimization unit operation and a first unmatched parameter range of the process bottleneck parameter comprises:

constructing an optimal binary decision tree and visualizing the optimal binary decision tree;

and positioning the process bottleneck parameter and the first unmatched parameter range in the first optimization unit operation according to the selection operation of the optimal binary decision tree.

17. The method of claim 16, wherein the step of constructing an optimal binary decision tree comprises:

establishing a data set consisting of features and targets, the features including key process parameters for each unit operation, the targets having values of 0 or 1;

constructing a binary decision tree based on the classification tree in the CART algorithm, and selecting the binary decision tree with the highest accuracy as the optimal binary decision tree.

18. The method of claim 17, wherein the value of the target is determined by:

inputting the features into the process flow model based on Monte Carlo simulation, and calculating the degree of mismatch of the first optimization unit operation;

and judging whether the mismatching degree of the first optimization unit operation is larger than a first threshold, if so, setting the value of the target to be 1, otherwise, setting the value of the target to be 0.

19. The method of claim 16, wherein the step of locating a process bottleneck parameter in the first optimized unit operation and a first unmatched parameter range of the process bottleneck parameter according to the optimal binary decision tree comprises: and positioning a branch from a node containing the most unmatched data at the bottom of the optimal binary decision tree to a root node, and taking all the optimal segmentation features and the optimal segmentation points on the branch as the process bottleneck parameter and the first unmatched parameter range respectively.

20. The method of claim 19, wherein the optimal segmentation feature and the optimal segmentation point are determined using a kini index.

21. The method of claim 17, further comprising filtering the features prior to constructing a binary decision tree based on the classification tree in the CART algorithm.

22. The method of claim 21, wherein the step of screening the features comprises process-principle based screening, correlation-based screening, and manual special-rule based screening.

23. The method of claim 22, wherein the relevance-based screening comprises computing a relevance of each feature to the target, and removing features having a relevance below a second threshold.

24. The method of claim 15, wherein optimizing the first optimization unit operation based on the first optimization parameter range comprises:

optimizing the numerical value of the process bottleneck parameter according to the first optimization parameter range, inputting the optimized numerical value of the process bottleneck parameter into a unit operation model, and calculating the optimization mismatching degree of the first optimization unit operation;

and judging whether the difference value of the highest mismatching degree and the optimized mismatching degree is larger than or equal to a third threshold, if so, finishing optimization by the first optimization unit operation.

25. The method of claim 15, further comprising:

taking the unit operation corresponding to one of the other mismatching degrees as a second optimization unit operation;

locating a second unmatched parameter range of the process bottleneck parameter and the process bottleneck parameter in the second optimization unit operation;

calculating a second optimized parameter range according to the second unmatched parameter range and the allowable range of the process bottleneck parameter;

and if the process bottleneck parameter in the first optimization unit operation is the same as the process bottleneck parameter in the second optimization unit operation, taking the intersection of the first optimization parameter range and the second optimization parameter range as a final optimization parameter range, and optimizing the first optimization unit operation and the second optimization unit operation according to the final optimization parameter range.

26. A matching degree calculation device for a process and a production line comprises:

a memory for storing instructions executable by the processor;

a processor for executing the instructions to implement the method of any one of claims 1-14.

27. A matching degree optimizing device for a process and a production line comprises:

a memory for storing instructions executable by the processor;

a processor for executing the instructions to implement the method of any of claims 15-25.

28. A computer-readable medium having stored thereon computer program code which, when executed by a processor, implements the method of any of claims 1-25.