Biological project risk assessment prediction method and system based on big data analysis

Abstract

The present invention discloses a biological project risk assessment prediction method and system based on big data analysis, which relates to the field of data processing technology. The method includes: performing experimental design disassembly and experimental behavior analysis on biological projects, obtaining multiple experimental process information, performing execution risk evaluation, and obtaining project execution risk coefficients; performing experimental process material risk analysis to obtain project material risk coefficients; reversely obtaining project cost risk coefficients based on multiple experimental process information, multiple biological material sets, and multiple sets of used chemicals; performing normalized calculations on multiple risk coefficients to obtain project comprehensive risk coefficients. The present invention solves the technical problem that the traditional biological project risk assessment in the prior art is based on a single dimension and has low comprehensiveness and accuracy of the assessment results, and achieves the technical effect of performing comprehensive risk assessment in multiple dimensions of biological projects based on big data analysis to improve the comprehensiveness and accuracy of project risk assessment.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a biological project risk assessment and prediction method and system based on big data analysis. Background Art

[0002] Biological project risk assessment is a complex technical task involving knowledge and technology in multiple fields. Biological projects usually include multiple links such as research and development, biological material management, experimental operation, safety management, etc., each of which has its own unique risk factors. With the rapid development of biotechnology, the complexity and risk of projects are also increasing. The risk assessment of biological projects needs to consider factors in multiple dimensions, which are interrelated and influence each other.

[0003] Traditional risk assessment of biological projects often focuses on a single dimension, such as the safety of the experimental process or the risk of material use, while ignoring other important factors. Therefore, the comprehensiveness and accuracy of risk assessment are low. Summary of the invention

[0004] The present application provides a biological project risk assessment prediction method and system based on big data analysis, which is used to solve the technical problems in the prior art that traditional biological project risk assessment is based on a single dimension and has low comprehensiveness and accuracy of assessment results.

[0005] The first aspect of the present application provides a biological project risk assessment and prediction method based on big data analysis, the method comprising: performing experimental design decomposition on the biological project to obtain multiple experimental projects; performing experimental behavior analysis on the multiple experimental projects to obtain multiple experimental process information, wherein each experimental process information includes experimental failure rate, experimental accident rate and equipment failure rate; pre-constructing an execution risk analysis model, and analyzing the multiple experimental process information through the execution risk analysis model to perform execution risk evaluation on the biological project to obtain a project execution risk coefficient; extracting and obtaining multiple biological material sets and multiple used chemical sets based on the multiple experimental projects; pre-constructing a material risk analysis model, and using the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and multiple used chemical sets to obtain a project material risk coefficient; reversely deducing the project cost risk coefficient based on the multiple experimental process information, multiple biological material sets and multiple used chemical sets; normalizing the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain a comprehensive project risk coefficient.

[0006] The second aspect of the present application provides a biological project risk assessment and prediction system based on big data analysis, the system comprising: an experimental design disassembly module, the experimental design disassembly module is used to perform experimental design disassembly on biological projects to obtain multiple experimental projects; an experimental behavior analysis module, the experimental behavior analysis module is used to perform experimental behavior analysis on the multiple experimental projects to obtain multiple experimental process information, wherein each experimental process information includes an experimental failure rate, an experimental accident rate and an equipment failure rate; an execution risk evaluation module, the execution risk evaluation module is used to pre-build an execution risk analysis model, and analyze the multiple experimental process information through the execution risk analysis model to perform execution risk evaluation on the biological project to obtain a project execution risk coefficient; a biological material information extraction module, the biological material information extraction module The material information extraction module is used to extract and obtain multiple biological material sets and multiple used chemical sets based on the multiple experimental projects; the material risk analysis module is used to pre-build a material risk analysis model, and use the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and multiple used chemical sets to obtain the project material risk coefficient; the project cost risk analysis module is used to reversely obtain the project cost risk coefficient based on the multiple experimental process information, multiple biological material sets and multiple used chemical sets; the project comprehensive risk calculation module is used to perform normalized calculation on the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain the project comprehensive risk coefficient.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The biological project risk assessment and prediction method and system based on big data analysis provided in the present application relate to the field of data processing technology. By performing experimental design disassembly and experimental behavior analysis on biological projects, multiple experimental process information is obtained, and execution risk evaluation, material risk analysis and project cost risk analysis are performed respectively to obtain the project execution risk coefficient, project material risk coefficient and project cost risk coefficient. Multiple risk coefficients are normalized and calculated to obtain the project comprehensive risk coefficient. This solves the technical problem that the traditional biological project risk assessment in the prior art is based on a single dimension and has low comprehensiveness and accuracy of the assessment results. This realizes the technical effect of conducting comprehensive risk assessment in multiple dimensions of biological projects based on big data analysis to improve the comprehensiveness and accuracy of project risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of a biological project risk assessment and prediction method based on big data analysis provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a process for calculating a project execution risk coefficient in a biological project risk assessment and prediction method based on big data analysis provided in an embodiment of the present application;

[0012] Figure 3 A schematic diagram of the structure of a biological project risk assessment and prediction system based on big data analysis provided in an embodiment of the present application.

[0013] Explanation of the accompanying drawings: experimental design disassembly module 11, experimental behavior analysis module 12, execution risk assessment module 13, biomaterial information extraction module 14, material risk analysis module 15, project cost risk analysis module 16, project comprehensive risk calculation module 17. DETAILED DESCRIPTION

[0014] The present application provides a biological project risk assessment prediction method and system based on big data analysis, which is used to solve the technical problems in the prior art that traditional biological project risk assessment is based on a single dimension and has low comprehensiveness and accuracy of assessment results.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0017] Embodiment 1

[0018] like Figure 1 As shown, the present application provides a biological project risk assessment prediction method based on big data analysis, the method comprising:

[0019] 1. A biological project risk assessment and prediction method based on big data analysis, characterized in that the method comprises:

[0020] P10: Decompose the experimental design of biological projects and obtain multiple experimental projects.

[0021] It should be understood that the experimental design decomposition of biological projects is the basis of the entire risk assessment process. By decomposing biological projects into multiple specific experimental projects, the risk factors of each experimental project can be analyzed more finely to ensure the comprehensiveness and accuracy of risk assessment. Specifically, first decompose the overall goals and tasks of the biological project, determine the various subtasks or experiments included in the project, and convert complex project goals into multiple operational experimental projects. After the project is decomposed, identify the specific experimental tasks of each experimental project, that is, the specific operations and steps that need to be completed during the experiment, such as sample processing, reagent preparation, equipment operation, etc. By identifying the experimental tasks, the operating procedures and technical requirements of each experimental project can be clarified.

[0022] Furthermore, the experimental conditions of each experimental project are determined, including the experimental environment, experimental equipment, experimental materials, etc. The experimental environment refers to the various physical and chemical conditions that need to be met during the experiment, such as temperature, humidity, light, etc. And according to the experimental tasks and experimental conditions, the specific experimental plan of each experimental project is designed, including the operating steps and methods that need to be followed during the experiment, as well as the order of the experimental steps, the specific operating methods of each step, the data recording method, etc.

[0023] P20: Performing experimental behavior analysis on the multiple experimental items to obtain multiple experimental process information, wherein each experimental process information includes an experimental failure rate, an experimental accident rate, and an equipment failure rate.

[0024] Furthermore, step P20 of the embodiment of the present application further includes:

[0025] P21: Performing experimental design feature extraction on the multiple experimental projects to obtain multiple technical route information and multiple experimental process information; P22: Using the multiple technical route information and multiple experimental process information as constraints, performing network data call to obtain multiple historical project data sets; P23: Extracting the first experimental project sequentially from the multiple experimental projects, and extracting the first historical project data set from the multiple historical project data set mapping according to the first experimental project, wherein the first historical project data set includes multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical equipment failure rates of multiple approximate historical projects; P24: Configure multiple project weight coefficients according to multiple deadlines of the multiple approximate historical projects; P25: Using the multiple project weight coefficients to perform weighted calculation on the multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical equipment failure rates to obtain the first experimental process information, wherein the first experimental process information includes the experimental failure rate, the experimental accident rate and the equipment failure rate; P26: Similarly, performing experimental behavior data extraction on the multiple historical project data sets, and performing analysis and calculation based on the extraction results to obtain the multiple experimental process information.

[0026] Optionally, conduct a detailed experimental behavior analysis on the multiple experimental projects that have been disassembled, and deeply explore various potential risks in the experimental process. First, perform experimental design feature extraction on multiple experimental projects to obtain multiple technical route information and multiple experimental process information. Technical route information refers to the specific technical path adopted by the experimental project, such as experimental methods, operating steps, etc. Experimental process information refers to the various operation links and their sequence in the experimental process, usually including experimental preparation, operation process, data collection and result analysis, etc. Furthermore, using the multiple technical route information and multiple experimental process information as constraints, through the networked data call technology, retrieve multiple historical project data sets with high similarity to the current experimental project from the database.

[0027] Furthermore, for each experimental project (taking the first experimental project as an example), the experimental data of multiple similar historical projects closest to it are mapped and extracted from the corresponding historical project data set, including historical experimental failure rate, historical experimental accident rate and historical equipment failure rate. In order to more accurately reflect the reference value of different historical projects to the current experimental project, the corresponding project weight coefficients are configured according to the deadline (i.e., completion time) of these historical projects. The more recent the project, the more mature the technology, the higher the data credibility, and are usually given a higher weight. Furthermore, multiple project weight coefficients are used to perform weighted calculations on multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical equipment failure rates to obtain the experimental process information of the first experimental project, including experimental failure rate, experimental accident rate and equipment failure rate. Through weighted calculation, the risk level of the current experimental project can be more accurately reflected.

[0028] By analogy, this method is extended to all experimental projects, and similar experimental behavior data extraction and analysis calculations are performed on multiple historical project data sets, and finally detailed experimental process information of all experimental projects is obtained. Through this step-by-step extraction and analysis method, not only the accuracy and comprehensiveness of risk assessment are improved, but also strong data support is provided for subsequent risk management and control.

[0029] P30: Pre-construct an execution risk analysis model, and analyze the plurality of experimental process information through the execution risk analysis model to conduct an execution risk assessment on the biological project and obtain a project execution risk coefficient.

[0030] Further, such as Figure 2 As shown, step P30 of the embodiment of the present application also includes:

[0031] P31: Interactively obtain multiple sample experimental process information and multiple sample quantitative execution risks; P32: Use a linear regression model to construct the unoptimized execution risk analysis model; P33: According to the preset proportion parameters, divide the multiple sample experimental process information and multiple sample quantitative execution risk identifiers into a training set, a test set and a validation set; P34: Based on minimizing the error, use the training set to estimate the model parameters of the execution risk analysis model; P35: Based on the test set, evaluate the prediction accuracy of the execution risk analysis model; P36: Use the validation set to optimize the variables of the execution risk analysis model to complete the optimization of the risk analysis model; P37: Analyze the multiple experimental process information through the execution risk analysis model to evaluate the execution risk of the biological project and obtain the quantitative execution risks of multiple projects of the multiple experimental projects; P38: Calculate the mean of the multiple project quantitative execution risks as the project execution risk coefficient.

[0032] It should be understood that a pre-constructed execution risk analysis model is used to analyze multiple experimental process information, evaluate the execution risk of biological projects, and obtain the project execution risk coefficient. Specifically, multiple sample experimental process information and multiple sample quantitative execution risks are first obtained in an interactive manner. Among them, the sample experimental process information includes data such as experimental failure rate, experimental accident rate, and equipment failure rate. The sample quantitative execution risk is the risk assessment result of historical projects, which serves as the label data of the model.

[0033] Furthermore, a linear regression model is used to construct an unoptimized execution risk analysis model. The linear regression model is a simple and effective statistical method that fits sample data by minimizing errors, thereby estimating the execution risk of the project. According to preset proportion parameters, multiple sample experimental process information and multiple sample quantitative execution risk identifiers are divided into training sets, test sets, and validation sets. The training set is used for model training, the test set is used for model performance evaluation, and the validation set is used for model optimization. Reasonable proportion division can ensure the generalization ability and prediction accuracy of the model. Based on the principle of minimizing errors, the training set is used to estimate the model parameters of the execution risk analysis model. Through iterative optimization, the parameter combination that minimizes the error is found, thereby constructing a preliminary risk analysis model.

[0034] Furthermore, the prediction accuracy of the execution risk analysis model is evaluated based on the test set. The performance of the model is evaluated by calculating the error between the predicted value and the actual value. If the error is large, the model needs to be further optimized. The validation set is used to optimize the variables of the execution risk analysis model. By adjusting the model variables and parameters, the error can be further reduced and the prediction accuracy of the model can be improved.

[0035] Furthermore, the experimental process information is input into the execution risk analysis model, and the multiple experimental process information is analyzed. By calculating the quantitative execution risk of each experimental project, the execution risk of the biological project is evaluated. Finally, the quantitative execution risk of multiple experimental projects is averaged, and the risk of each experimental project is comprehensively considered to provide an overall project risk evaluation index as the project execution risk coefficient. This coefficient comprehensively reflects the risk level that the entire biological project may encounter during the execution process, and provides an important reference for project decision-making.

[0036] P40: Extracting and obtaining a plurality of biological material sets and a plurality of used chemical sets based on the plurality of experimental projects.

[0037] Specifically, based on the specific needs of multiple experimental projects, multiple biological material sets and multiple sets of used chemicals are extracted to identify and classify all biological materials and chemicals required in the experimental projects for subsequent material risk analysis. Exemplarily, first collect detailed experimental plans and experimental step information for each experimental project, including data such as the types, quantities, specifications, etc. of all biological materials and chemicals required for the experiment. Then, according to the experimental plan, the required biological materials and chemicals are classified and sorted to obtain multiple biological material sets and multiple sets of used chemicals. The biological material set includes all biological samples used for the experiment, such as cell lines, tissue samples, DNA / RNA samples, etc. The chemical set includes all chemical reagents and consumables used in the experiment, such as solutions, buffers, enzymes, dyes, etc.

[0038] Furthermore, the collected multiple biological material sets and multiple used chemical sets are standardized to ensure data consistency. For example, different names of the same material or chemical are standardized to avoid data redundancy and confusion. The standardized biological material and chemical information is stored in a database. The database should contain detailed information on each material and chemical, including name, specification, supplier, storage conditions, safety information, etc. By building a database, efficient management and rapid retrieval of material and chemical information can be achieved.

[0039] P50: Pre-construct a material risk analysis model, and use the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and the multiple used chemical sets to obtain the project material risk coefficient.

[0040] Furthermore, step P50 of the embodiment of the present application also includes:

[0041] P51: Interactively obtain a quality weight engine; P52: Interactively obtain a biomaterial risk table and a chemical risk table; P53: Construct a biological risk analysis channel based on the quality weight engine and the biomaterial risk table; P54: Construct a chemical risk analysis channel based on the quality weight engine and the chemical risk table; P55: Preconfigure biological risk weight coefficients and chemical risk weight coefficients, and construct a risk comprehensive calculation channel based on the biological risk weight coefficients and the chemical risk weight coefficients; P56: Connect the biomaterial risk analysis channel and the chemical risk analysis channel in parallel, and connect the output ends of the biomaterial risk analysis channel and the chemical risk analysis channel to the input end of the risk comprehensive calculation channel to complete the construction of the material risk analysis model.

[0042] Optionally, a material risk analysis model is pre-built to perform material risk analysis on a set of biological materials and a set of chemicals during the experiment. Specifically, a quality weight engine is first obtained interactively. The quality weight engine is an important tool for evaluating the quality of materials and chemicals, and can weight and score materials and chemicals according to different quality indicators. Through this tool, reliable weight coefficients can be provided for subsequent risk analysis. Furthermore, a biomaterial risk table and a chemical risk table are obtained interactively. These risk tables contain risk data of various biological materials and chemicals, such as toxicity, flammability, reactivity, etc., which can provide basic data support for risk analysis of materials and chemicals.

[0043] Furthermore, a biological risk analysis channel is constructed based on the quality weight engine and the biomaterial risk table. The biological risk analysis channel is an analysis path for assessing the risk of biomaterials. By combining the quality weight engine and the risk data of biomaterials, a comprehensive risk assessment is performed on each biomaterial. At the same time, a chemical risk analysis channel is constructed based on the quality weight engine and the chemical risk table. The chemical risk analysis channel is used to assess the risk of chemicals during the experiment and calculate the risk level of chemicals by analyzing the characteristics and risk data of chemicals.

[0044] Furthermore, according to the actual situation of the project and the needs of risk assessment, biological risk weight coefficients and chemical risk weight coefficients are configured, and a comprehensive risk calculation channel is constructed based on these weight coefficients. In this way, the biological risk and chemical risk are comprehensively evaluated to obtain the overall material risk coefficient of the project. The configuration of the weight coefficients can be flexibly adjusted according to the actual situation of the project and the needs of risk assessment.

[0045] Furthermore, the biomaterial risk analysis channel and the chemical risk analysis channel are connected in parallel, and the output ends of the biomaterial risk analysis channel and the chemical risk analysis channel are connected to the input end of the risk comprehensive calculation channel to complete the construction of the material risk analysis model. Finally, the material risk analysis model is used to perform experimental process material risk analysis on the multiple biomaterial sets and the multiple sets of used chemicals. For example, multiple biomaterial sets are used to traverse the risk table to determine the risk score, and the risk score is weighted in combination with the quality weight of the biomaterial. Finally, the overall risk coefficient of the biomaterial set is calculated, and then the risk coefficient of the overall risk comprehensive calculation channel and the set of used chemicals is further weighted and summed to finally obtain the project material risk coefficient.

[0046] P60: Obtain the project cost risk coefficient by reverse deduction based on the multiple experimental process information, multiple biological material sets and multiple used chemical sets.

[0047] Furthermore, step P60 of the embodiment of the present application also includes:

[0048] P61: interactively obtain a real-time material cost set, and calculate the conceptual material cost based on the real-time material cost set, multiple biological material sets and multiple used chemical sets; P62: interactively obtain a real-time equipment cost set, obtain multiple sets of experimental equipment lists from the multiple experimental projects, and calculate the conceptual equipment cost based on the real-time equipment cost set and the multiple sets of experimental equipment lists; P63: pre-build a safety stock analysis model, extract multiple experimental failure rates and multiple experimental accident rates from the multiple experimental process information, and analyze the multiple experimental failure rates and multiple experimental accident rates through the safety stock analysis model to obtain a safety stock index; P64 : Pre-construct a redundant equipment analysis model, extract multiple equipment failure rates from the multiple experimental process information, and analyze the multiple experimental accident rates and multiple equipment failure rates through the redundant equipment analysis model to obtain a redundant equipment index; P65: Multiply the conceptual material cost and the safety stock index to obtain the predicted material cost, and multiply the conceptual equipment cost and the redundant equipment index to obtain the predicted equipment cost; P66: Use the sum of the predicted material cost and the predicted equipment cost as the numerator, and the sum of the conceptual material cost and the conceptual equipment cost as the denominator to calculate and obtain the project cost risk coefficient.

[0049] It should be understood that the project cost risk coefficient is obtained by reverse analysis of the experimental process information, biological material set, used chemical set and real-time cost data. First, the real-time material cost set is obtained interactively, which contains the real-time price information of various biological materials and chemicals in the current market, and can generate a biological material purchase cost table and a chemical material purchase cost table. Subsequently, the conceptual material cost is calculated based on the real-time material cost set, multiple biological material sets and multiple used chemical sets. The conceptual material cost refers to the estimated total cost of materials and chemicals required in the experimental project without considering risk factors, such as when the equipment is not damaged and there are no defective products in the batch production process.

[0050] Furthermore, the real-time equipment cost set is obtained interactively, and multiple sets of experimental equipment lists are extracted from multiple experimental projects. Based on the real-time equipment cost set and multiple sets of experimental equipment lists, the conceptual equipment cost is calculated, that is, the estimated total cost of the equipment required in the experimental project without considering equipment failure and other risk factors.

[0051] Furthermore, in order to more comprehensively evaluate the cost risk, a safety stock analysis model and a redundant equipment analysis model are pre-built. The experimental failure rate and experimental accident rate are extracted from multiple experimental process information, and these data are analyzed by the safety stock analysis model to obtain the safety stock index. This index reflects the additional inventory level that needs to be maintained to deal with experimental failures and accidents, and has an important impact on material costs. The equipment failure rate is extracted from the experimental process information, and these data are analyzed by the redundant equipment analysis model to obtain the redundant equipment index. This index reflects the number of redundant equipment that needs to be equipped to reduce the impact of equipment failures on the experimental progress, which has a direct impact on equipment costs.

[0052] Furthermore, the conceptual material cost is multiplied by the safety stock index to obtain the predicted material cost; the conceptual equipment cost is multiplied by the redundant equipment index to obtain the predicted equipment cost. These two predicted values ​​respectively reflect the material cost and equipment cost that the project may face after considering the safety stock and redundant equipment. Finally, the sum of the predicted material cost and the predicted equipment cost is used as the numerator, and the sum of the conceptual material cost and the conceptual equipment cost is used as the denominator to calculate the project cost risk coefficient. This coefficient reflects the extent to which the actual cost of the project may deviate from the expected cost after considering various uncertainties. Through this coefficient, project managers can have a clearer understanding of the cost risk status of the project and formulate corresponding risk management strategies accordingly.

[0053] Furthermore, step P63 of the embodiment of the present application also includes:

[0054] P63-1: Interactively obtain multiple historical inventory indexes of the multiple approximate historical projects in the first historical project data set; P63-2: Construct a standard inventory analysis branch based on a back propagation neural network; P63-3: Use the multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical inventory indices as training data to perform model training and parameter adjustment for the standard inventory analysis branch to obtain a first inventory analysis branch; P63-4: Similarly, construct multiple inventory analysis branches to obtain the multiple experimental projects; P63-5: Connect the multiple inventory analysis branches in parallel to obtain the safety inventory analysis model; P63-6: Map and synchronize the multiple experimental failure rates and the multiple experimental accident rates to the multiple inventory analysis branches in the safety inventory analysis model to obtain multiple experimental inventory indices; P63-7: After serializing the multiple experimental inventory indices, extract the maximum value as the safety inventory index.

[0055] Optionally, the specific process of constructing the safety inventory analysis model may be, first, interactively obtaining the historical inventory indexes of multiple approximate historical projects in the first historical project data set, that is, the safety and reliability indicators of the material inventory in the historical experimental project. Further, based on the back propagation neural network, a standard inventory analysis branch is constructed. The back propagation neural network is a commonly used deep learning algorithm that minimizes prediction errors by adjusting network weights and is suitable for complex nonlinear data analysis. In this step, BPNN is selected because it can handle complex nonlinear relationships and automatically adjust network parameters through the back propagation algorithm to minimize prediction errors.

[0056] Furthermore, multiple historical experimental failure rates, historical experimental accident rates, and historical inventory indexes are used as training data to train and adjust the model parameters of the standard inventory analysis branch. The historical data is input into the network, and the network weights and biases are adjusted through multiple iterations so that the network output (i.e., the predicted inventory index) is as close as possible to the actual inventory index, and the first inventory analysis branch suitable for the current experimental project is obtained. In the same way, multiple inventory analysis branches for multiple experimental projects are constructed. Each inventory analysis branch corresponds to a specific experimental project, and the accuracy and applicability of the analysis branch are ensured through independent training and parameter adjustment.

[0057] Furthermore, multiple inventory analysis branches are connected in parallel, that is, multiple independent analysis branches are combined together to form a safety inventory analysis model for comprehensive evaluation of inventory risks. Multiple experimental failure rates and multiple experimental accident rates are then mapped and synchronized to multiple inventory analysis branches in the safety inventory analysis model. Through this mapping, the data of the current experimental project is synchronously input into the model to obtain multiple experimental inventory indexes. Furthermore, after serializing multiple experimental inventory indexes, the maximum value is extracted as the safety inventory index to ensure that the entire project can still maintain a sufficient inventory level when facing the worst case.

[0058] P70: Perform normalized calculation on the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain the project comprehensive risk coefficient.

[0059] Furthermore, step P70 of the embodiment of the present application further includes:

[0060] P71: Pre-construct execution risk normalization coefficient, material risk normalization coefficient and cost risk normalization coefficient; P72: Use the execution risk normalization coefficient, material risk normalization coefficient and cost risk normalization coefficient to perform normalized calculation on the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain the project comprehensive risk coefficient; P73: Preset a project risk threshold. If the project comprehensive risk coefficient exceeds the project risk threshold, identify a risk label for the biological project.

[0061] In a possible embodiment of the present application, a comprehensive risk coefficient that fully reflects the risk status of the project is obtained by comprehensively considering the project execution risk, material risk and cost risk. Specifically, firstly, the execution risk normalization coefficient, material risk normalization coefficient and cost risk normalization coefficient are pre-constructed. The normalization coefficient is pre-set according to factors such as project characteristics, risk assessment objectives and historical experience, and is used to convert execution risk coefficients, material risk coefficients and cost risk coefficients of different dimensions and magnitudes to the same scale for comparison. The determination of the normalization coefficient usually involves expert consultation, data analysis and other methods to ensure its rationality and effectiveness.

[0062] Furthermore, the pre-built execution risk normalization coefficient, material risk normalization coefficient and cost risk normalization coefficient are used to normalize the corresponding project execution risk coefficient, project material risk coefficient and project cost risk coefficient, eliminating the dimensional differences and magnitude differences between different risk dimensions, so that each risk coefficient can be weighted summed or comprehensively evaluated on the same scale. The specific method of normalization calculation may include linear transformation, nonlinear transformation and other methods, and the specific choice depends on the needs of risk assessment and the characteristics of the data.

[0063] Furthermore, the normalized execution risk coefficient, material risk coefficient, and cost risk coefficient are weighted and summed or comprehensively processed to obtain the comprehensive risk coefficient of the project. This coefficient comprehensively reflects the risk status of the project in terms of execution, materials, and costs, and provides intuitive risk assessment results for project managers. At the same time, in order to better guide project decision-making and risk management, a project risk threshold can be preset, and the calculated comprehensive risk coefficient of the project can be compared with the preset risk threshold. If the comprehensive risk coefficient of the project exceeds the risk threshold, it indicates that the project has a high risk level and corresponding risk management measures need to be taken to reduce the risk. At this time, a risk label can be identified for the biological project to remind project managers and project team members to pay attention to risks and take corresponding countermeasures.

[0064] In summary, the embodiments of the present application have at least the following technical effects:

[0065] This application obtains multiple experimental process information by performing experimental design disassembly and experimental behavior analysis on biological projects, and performs execution risk evaluation, material risk analysis and project cost risk analysis respectively to obtain the project execution risk coefficient, project material risk coefficient and project cost risk coefficient. Multiple risk coefficients are normalized and calculated to obtain the comprehensive project risk coefficient.

[0066] The technical effect of conducting comprehensive risk assessment in multiple dimensions of biological projects based on big data analysis has been achieved, thereby improving the comprehensiveness and accuracy of project risk assessment.

[0067] Embodiment 2

[0068] Based on the same inventive concept as the biological project risk assessment prediction method based on big data analysis in the aforementioned embodiment, Figure 3 As shown, the present application provides a biological project risk assessment and prediction system based on big data analysis, and the system and method embodiments in the present application embodiments are based on the same inventive concept. Among them, the system includes:

[0069] The experimental design disassembly module 11 is used to disassemble the experimental design of the biological project to obtain multiple experimental projects.

[0070] The experimental behavior analysis module 12 is used to perform experimental behavior analysis on the multiple experimental items to obtain multiple experimental process information, wherein each experimental process information includes an experimental failure rate, an experimental accident rate and an equipment failure rate.

[0071] The execution risk assessment module 13 is used to pre-build an execution risk analysis model and analyze the multiple experimental process information through the execution risk analysis model to perform execution risk assessment on the biological project and obtain a project execution risk coefficient.

[0072] The biomaterial information extraction module 14 is used to extract and obtain multiple biomaterial sets and multiple used chemical sets based on the multiple experimental items.

[0073] The material risk analysis module 15 is used to pre-build a material risk analysis model, and use the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and the multiple used chemical sets to obtain the project material risk coefficient.

[0074] The project cost risk analysis module 16 is used to reversely obtain the project cost risk coefficient according to the multiple experimental process information, multiple biological material sets and multiple used chemical sets.

[0075] The project comprehensive risk calculation module 17 is used to perform normalized calculation on the project execution risk coefficient, the project material risk coefficient and the project cost risk coefficient to obtain the project comprehensive risk coefficient.

[0076] Furthermore, the experimental behavior analysis module 12 is also used to perform the following steps:

[0077] Perform experimental design feature extraction on the multiple experimental projects to obtain multiple technical route information and multiple experimental process information; use the multiple technical route information and multiple experimental process information as constraints to perform network data call to obtain multiple historical project data sets; extract the first experimental project sequentially from the multiple experimental projects, and extract the first historical project data set from the multiple historical project data set mapping according to the first experimental project, wherein the first historical project data set includes multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical equipment failure rates of multiple approximate historical projects; configure multiple project weight coefficients according to multiple deadlines of the multiple approximate historical projects; use the multiple project weight coefficients to perform weighted calculation on the multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical equipment failure rates to obtain first experimental process information, wherein the first experimental process information includes experimental failure rate, experimental accident rate and equipment failure rate; and so on, perform experimental behavior data extraction on the multiple historical project data sets, and perform analysis and calculation based on the extraction results to obtain the multiple experimental process information.

[0078] Furthermore, the execution risk assessment module 13 is further configured to perform the following steps:

[0079] Interactively obtain multiple sample experimental process information and multiple sample quantitative execution risks; use a linear regression model to construct the unoptimized execution risk analysis model; according to preset proportion parameters, divide the multiple sample experimental process information and multiple sample quantitative execution risk identifiers into a training set, a test set and a validation set; based on minimizing the error, use the training set to estimate the model parameters of the execution risk analysis model; based on the test set, evaluate the prediction accuracy of the execution risk analysis model; use the validation set to optimize the variables of the execution risk analysis model to complete the optimization of the risk analysis model; analyze the multiple experimental process information through the execution risk analysis model to evaluate the execution risk of the biological project and obtain multiple project quantitative execution risks of the multiple experimental projects; calculate the mean of the multiple project quantitative execution risks as the project execution risk coefficient.

[0080] Furthermore, the material risk analysis module 15 is also used to perform the following steps:

[0081] Interactively obtain a quality weight engine; interactively obtain a biomaterial risk table and a chemical risk table; construct a biological risk analysis channel based on the quality weight engine and the biomaterial risk table; construct a chemical risk analysis channel based on the quality weight engine and the chemical risk table; preconfigure a biological risk weight coefficient and a chemical risk weight coefficient, and construct a risk comprehensive calculation channel based on the biological risk weight coefficient and the chemical risk weight coefficient; connect the biomaterial risk analysis channel and the chemical risk analysis channel in parallel, and connect the output ends of the biomaterial risk analysis channel and the chemical risk analysis channel to the input end of the risk comprehensive calculation channel to complete the construction of the material risk analysis model.

[0082] Furthermore, the project cost risk analysis module 16 is also used to perform the following steps:

[0083] Interactively obtain a real-time material cost set, and calculate the conceptual material cost based on the real-time material cost set, multiple biological material sets and multiple used chemical sets; interactively obtain a real-time equipment cost set, obtain multiple groups of experimental equipment lists from the multiple experimental projects, and then calculate the conceptual equipment cost based on the real-time equipment cost set and the multiple groups of experimental equipment lists; pre-construct a safety stock analysis model, extract multiple experimental failure rates and multiple experimental accident rates from the multiple experimental process information, and analyze the multiple experimental failure rates and multiple experimental accident rates through the safety stock analysis model to obtain a safety stock index; pre-construct a redundant equipment analysis model, extract multiple equipment failure rates from the multiple experimental process information, and analyze the multiple experimental accident rates and multiple equipment failure rates through the redundant equipment analysis model to obtain a redundant equipment index; multiply the conceptual material cost and the safety stock index to obtain a predicted material cost, and multiply the conceptual equipment cost and the redundant equipment index to obtain a predicted equipment cost; use the sum of the predicted material cost and the predicted equipment cost as the numerator, and the sum of the conceptual material cost and the conceptual equipment cost as the denominator to calculate the project cost risk coefficient.

[0084] Furthermore, the project cost risk analysis module 16 is also used to perform the following steps:

[0085] Interactively obtain multiple historical inventory indexes of the multiple approximate historical projects in the first historical project data set; construct a standard inventory analysis branch based on a back propagation neural network; use the multiple historical experimental failure rates, multiple historical experimental accident rates and multiple historical inventory indexes as training data to perform model training and parameter adjustment of the standard inventory analysis branch to obtain a first inventory analysis branch; and so on, construct multiple inventory analysis branches to obtain the multiple experimental projects; connect the multiple inventory analysis branches in parallel to obtain the safety inventory analysis model; map and synchronize the multiple experimental failure rates and the multiple experimental accident rates to the multiple inventory analysis branches in the safety inventory analysis model to obtain multiple experimental inventory indexes; after serializing the multiple experimental inventory indexes, extract the maximum value as the safety inventory index.

[0086] Furthermore, the project comprehensive risk calculation module 17 is also used to perform the following steps:

[0087] Pre-construct an execution risk normalization coefficient, a material risk normalization coefficient, and a cost risk normalization coefficient; and use the execution risk normalization coefficient, the material risk normalization coefficient, and the cost risk normalization coefficient to perform normalized calculations on the project execution risk coefficient, the project material risk coefficient, and the project cost risk coefficient to obtain the project comprehensive risk coefficient; preset a project risk threshold, and if the project comprehensive risk coefficient exceeds the project risk threshold, identify a risk label for the biological project.

[0088] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0090] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A biological project risk assessment and prediction method based on big data analysis, characterized in that: The method comprises: Conduct experimental design and disassemble biological projects to obtain multiple experimental projects; Performing experimental behavior analysis on the multiple experimental items to obtain multiple experimental process information, wherein each experimental process information includes an experimental failure rate, an experimental accident rate, and an equipment failure rate; Pre-constructing an execution risk analysis model, and analyzing the plurality of experimental process information through the execution risk analysis model to perform an execution risk assessment on the biological project and obtain a project execution risk coefficient; Extracting and obtaining a plurality of biological material sets and a plurality of used chemical sets based on the plurality of experimental projects; Pre-constructing a material risk analysis model, and using the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and the multiple used chemical sets to obtain a project material risk coefficient; Reversely obtain the project cost risk coefficient based on the multiple experimental process information, multiple biological material sets and multiple used chemical sets; Normalizing and calculating the project execution risk coefficient, the project material risk coefficient and the project cost risk coefficient to obtain a comprehensive project risk coefficient; performing experimental behavior analysis on the multiple experimental projects to obtain multiple experimental process information, the method further includes: Extracting experimental design features from the multiple experimental projects to obtain multiple technical route information and multiple experimental process information; Taking the multiple technical route information and multiple experimental process information as constraints, network data call is performed to obtain multiple historical project data sets; Extracting a first experimental project from the plurality of experimental projects in sequence, and extracting a first historical project data set from the plurality of historical project data sets according to the first experimental project, wherein the first historical project data set includes a plurality of historical experimental failure rates, a plurality of historical experimental accident rates, and a plurality of historical equipment failure rates of a plurality of approximate historical projects; configuring a plurality of project weight coefficients according to a plurality of deadlines of the plurality of approximate historical projects; Using the multiple project weight coefficients to perform weighted calculation on the multiple historical experiment failure rates, the multiple historical experiment accident rates, and the multiple historical equipment failure rates to obtain first experiment process information, wherein the first experiment process information includes the experiment failure rate, the experiment accident rate, and the equipment failure rate; By analogy, the experimental behavior data of the multiple historical project data sets are extracted, and analysis and calculation are performed based on the extraction results to obtain the multiple experimental process information; an execution risk analysis model is pre-constructed, and the multiple experimental process information is analyzed by the execution risk analysis model to perform execution risk evaluation on the biological project to obtain a project execution risk coefficient. The method also includes: Interactively obtain multiple sample experimental process information and multiple sample quantitative execution risks; Using a linear regression model to construct the unoptimized execution risk analysis model; According to a preset ratio parameter, the multiple sample experimental process information and the multiple sample quantitative execution risk identifiers are divided into a training set, a test set and a validation set; Based on minimizing the error, using the training set to estimate the model parameters of the execution risk analysis model; Performing a prediction accuracy evaluation of the execution risk analysis model based on the test set; Using the validation set to perform variable optimization on the execution risk analysis model, thereby completing the optimization of the risk analysis model; Analyzing the plurality of experimental process information by using the execution risk analysis model to perform execution risk assessment on the biological project and obtain quantitative execution risks of the plurality of experimental projects; An average of the quantified execution risks of the multiple projects is calculated as the project execution risk coefficient.

2. The biological project risk assessment and prediction method based on big data analysis according to claim 1, characterized in that: Pre-building a material risk analysis model, the method further comprising: Interactively obtain quality weight engine; Interactively obtain the biological material risk table and chemical risk table; Constructing a biological risk analysis channel based on the quality weight engine and the biological material risk table; Constructing a chemical risk analysis channel based on the quality weight engine and the chemical risk table; Preconfigure biological risk weight coefficients and chemical risk weight coefficients, and construct a comprehensive risk calculation channel based on the biological risk weight coefficients and chemical risk weight coefficients; The biomaterial risk analysis channel and the chemical risk analysis channel are connected in parallel, and the output ends of the biomaterial risk analysis channel and the chemical risk analysis channel are connected to the input end of the risk comprehensive calculation channel to complete the construction of the material risk analysis model.

3. The biological project risk assessment and prediction method based on big data analysis according to claim 1, characterized in that: Obtaining a project cost risk coefficient by reverse deduction based on the plurality of experimental process information, the plurality of biological material sets and the plurality of used chemical sets, the method further comprising: Interactively obtain a real-time material cost set, and calculate and obtain a conceptual material cost based on the real-time material cost set, a plurality of biological material sets, and a plurality of used chemical sets; Interactively obtain a real-time equipment cost set, obtain multiple groups of experimental equipment lists from the multiple experimental projects, and then calculate and obtain a conceptual equipment cost based on the real-time equipment cost set and the multiple groups of experimental equipment lists; Pre-constructing a safety stock analysis model, extracting multiple experimental failure rates and multiple experimental accident rates from the multiple experimental process information, and analyzing the multiple experimental failure rates and multiple experimental accident rates through the safety stock analysis model to obtain a safety stock index; Pre-constructing a redundant equipment analysis model, extracting multiple equipment failure rates from the multiple experimental process information, and analyzing the multiple experimental accident rates and multiple equipment failure rates through the redundant equipment analysis model to obtain a redundant equipment index; The conceptual material cost and the safety stock index are multiplied to obtain the predicted material cost, and the conceptual equipment cost and the redundant equipment index are multiplied to obtain the predicted equipment cost; The project cost risk coefficient is calculated by taking the sum of the predicted material cost and the predicted equipment cost as the numerator and the sum of the conceptual material cost and the conceptual equipment cost as the denominator.

4. The biological project risk assessment and prediction method based on big data analysis as claimed in claim 3, characterized in that: Pre-constructing a safety stock analysis model, extracting multiple experimental failure rates and multiple experimental accident rates from the multiple experimental process information, and analyzing the multiple experimental failure rates and multiple experimental accident rates through the safety stock analysis model to obtain a safety stock index, the method further comprising: interactively obtaining a plurality of historical inventory indexes of the plurality of similar historical items in the first historical item data set; Build a standard stock analysis branch based on back-propagation neural network; Using the multiple historical experiment failure rates, multiple historical experiment accident rates, and multiple historical inventory indexes as training data to perform model training and parameter adjustment for the standard inventory analysis branch, to obtain a first inventory analysis branch; By analogy, multiple inventory analysis branches are constructed to obtain the multiple experimental projects; Connecting the multiple inventory analysis branches in parallel to obtain the safety inventory analysis model; Synchronize the mapping of the multiple experimental failure rates and the multiple experimental accident rates to the multiple inventory analysis branches in the safety inventory analysis model to obtain multiple experimental inventory indexes; After serializing the multiple experimental inventory indexes, a maximum value is extracted as the safety inventory index.

5. The biological project risk assessment and prediction method based on big data analysis according to claim 1, characterized in that: Normalizing and calculating the project execution risk coefficient, the project material risk coefficient and the project cost risk coefficient to obtain a comprehensive project risk coefficient, the method further comprising: Pre-construct execution risk normalization factor, material risk normalization factor and cost risk normalization factor; And using the execution risk normalization coefficient, material risk normalization coefficient and cost risk normalization coefficient to perform normalized calculation on the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain the project comprehensive risk coefficient; A project risk threshold is preset, and if the comprehensive risk factor of the project exceeds the project risk threshold, a risk label is assigned to the biological project.

6. A biological project risk assessment and prediction system based on big data analysis, characterized in that: The system is used to execute the method according to any one of claims 1 to 5, and the system comprises: An experimental design disassembly module, wherein the experimental design disassembly module is used to disassemble the experimental design of a biological project to obtain multiple experimental projects; An experimental behavior analysis module, wherein the experimental behavior analysis module is used to perform experimental behavior analysis on the multiple experimental items to obtain multiple experimental process information, wherein each experimental process information includes an experimental failure rate, an experimental accident rate, and an equipment failure rate; An execution risk assessment module, wherein the execution risk assessment module is used to pre-build an execution risk analysis model and analyze the plurality of experimental process information through the execution risk analysis model to perform an execution risk assessment on the biological project and obtain a project execution risk coefficient; A biomaterial information extraction module, the biomaterial information extraction module is used to extract and obtain multiple biomaterial sets and multiple used chemical sets based on the multiple experimental items; A material risk analysis module, wherein the material risk analysis module is used to pre-build a material risk analysis model, and use the material risk analysis model to perform experimental process material risk analysis on the multiple biological material sets and the multiple used chemical sets to obtain a project material risk coefficient; A project cost risk analysis module, wherein the project cost risk analysis module is used to reversely obtain a project cost risk coefficient based on the plurality of experimental process information, the plurality of biological material sets and the plurality of used chemical sets; The project comprehensive risk calculation module is used to perform normalized calculation on the project execution risk coefficient, project material risk coefficient and project cost risk coefficient to obtain the project comprehensive risk coefficient.

Images