Evaluation method and system of scientific and technological projects based on big data

Abstract

The present invention relates to the field of efficiency evaluation technology, specifically to a technology project evaluation method and system based on big data, comprising the following steps: based on an initialized technology project database, collecting historical execution data of existing projects, including execution results, quality levels and team feedback information, using MapReduce, establishing and sorting data indexes, and generating a project historical performance library. The present invention uses MapReduce technology to establish data indexes and classify and sort information, thereby achieving efficient data access and query response, identifying key performance indicators from a large amount of historical data through a random forest algorithm, constructing an accurate risk prediction model, improving the model's predictive ability and coverage of project risks, and using a support vector machine to adjust and optimize the performance and risk assessment of new projects, ensuring the pertinence and high adaptability of the assessment process, and making the assessment results more accurately reflect the actual situation of the project.

Description

Technical Field

[0001] The present invention relates to the technical field of efficiency evaluation, and in particular to a technology project evaluation method and system based on big data. Background Art

[0002] The field of efficiency evaluation technology involves a series of methods and tools used to quantify and analyze the efficiency of a project to determine the effectiveness and productivity of its resource utilization. This field is widely used in many industries such as industry, education, medical care, government and corporate management. The main purpose is to evaluate the economy of operations and the degree of optimization of processes through data analysis and efficiency indicators, so as to clarify the rationality of resource allocation, improve the scientific nature of decision-making and the efficiency of project execution.

[0003] Among them, the science and technology project evaluation method is a method specifically used to evaluate the efficiency and effectiveness of science and technology projects from concept to completion, aiming to determine whether the project can achieve its established goals and whether the use of resources during implementation is reasonable. The purpose of science and technology project evaluation includes helping project managers and decision makers to regularly check the progress, risks and potential value of projects, optimize resource allocation, improve project success rates, and support the decision-making process of future project planning and budget preparation.

[0004] In a big data environment, traditional methods have low efficiency in data indexing and querying, and cannot meet the needs of real-time data analysis. Traditional methods lack dynamic adaptability and cannot quickly adjust assessment models in response to various project changes, making risk predictions incomplete or out of date, affecting the scientific nature of decision-making and the accuracy of project management, leading to resource waste, project delays and even budget overruns in project management, which in turn affects the investment benefits and quality assurance of the entire project. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a technology project evaluation method and system based on big data.

[0006] In order to achieve the above purpose, the present invention adopts the following technical scheme: a scientific and technological project evaluation method based on big data, comprising the following steps:

[0007] S1: Based on the initialized science and technology project database, collect the historical execution data of existing projects, including execution results, quality levels and team feedback information, use MapReduce to establish and organize data indexes, and generate a project historical performance library;

[0008] S2: Based on the historical performance database of the project, the random forest algorithm is used to select key performance indicators in the data, build a risk prediction model, evaluate the relationship between historical performance and risk factors, and integrate multi-dimensional risk data to generate a risk and performance association model;

[0009] S3: Apply the risk and performance association model to the new project evaluation, use the support vector machine, combine the current project situation, perform performance and risk evaluation on the new project, analyze the potential risk points of the project, and draw a risk distribution map to obtain risk evaluation indicators;

[0010] S4: Based on the risk assessment indicators, conduct project decision analysis, analyze the potential risk control process in the project, implement project resource allocation adjustments, integrate resource and risk information, and obtain project adjustment proposals.

[0011] The present invention is improved in that the steps of acquiring the project historical performance library are specifically as follows:

[0012] S111: Extract historical execution data from the initialized science and technology project database, including project execution results, quality levels, and team feedback information, and classify the data by coding or marking;

[0013] S112: Apply MapReduce to organize the collected data, process and index each type of data through the Map function, and then aggregate the data through the Reduce function to form an intermediate data set in a unified format;

[0014] S113: Summarize the processed data set using the formula:

[0015]

[0016] Determine the contribution weight of the difference data category to the performance evaluation and build a project historical performance library, where x, y, and z represent the scores of execution results, quality level, and team feedback, respectively, and w x 、w y and w z is the corresponding weight coefficient, and t is the calculated performance indicator.

[0017] The present invention is improved in that the step of selecting the key performance indicators is specifically as follows:

[0018] S211: extracting performance data of historical projects from the project historical performance database, including project execution results, quality levels, and team feedback information;

[0019] S212: Perform statistical analysis on the extracted data to identify key performance data points that are correlated with project success or failure;

[0020] S213: Through the analysis results, select performance data points that have predictive power for project risks as key performance indicators, and identify delays and cost overruns as key factors;

[0021] S214: Integrate the selected performance data points using the formula:

[0022]

[0023] Calculate the key performance indicator set, where P represents the project delay index, which is an indicator to measure whether the project is completed on time; Q represents the cost overrun index, which measures the cost management efficiency of the project; R is the relationship between P and Q, which optimizes the influence of the two indicators in performance evaluation; S represents customer satisfaction, which is directly related to the customer's satisfaction with the project; T and U represent team satisfaction and supply chain efficiency, respectively, which balance the influence of the indicators.

[0024] The present invention is improved in that the steps of obtaining the risk prediction model are specifically as follows:

[0025] S221: From the key performance indicators, select indicators associated with project risks, input the selected key indicators to build a model, and perform project risk prediction using the formula:

[0026] RI=a×KPI 2 +b×KPI+c

[0027] The risk index is calculated based on the square and linear combination of key performance indicators, where RI stands for risk index, KPI stands for key performance indicator, and a, b, and c are model adjustment parameters;

[0028] S222: Test the historical project data, analyze the correlation between the prediction results of the model and the project risk, and adjust the model parameters a, b and c according to the results of the risk correlation analysis to obtain the risk prediction model.

[0029] The present invention is improved in that the analysis steps of the potential risk points of the project are specifically as follows:

[0030] S311: Collect performance data on current projects, including time management, cost control, and quality indicators, and identify patterns and trends associated with project execution;

[0031] S312: Apply set operations to the collected data to identify key areas in the data that show potential problems, including delay trends, budget overruns, or quality decline trends, and mark them as potential risk points;

[0032] S313: Evaluate the identified potential risk points using the formula:

[0033]

[0034] Calculate the impact and probability of each risk point, and set the risk level based on the historical risk data and current project parameters. Among them, p1 represents the delay trend value, indicating the severity of project delay, p1 square term represents the accelerated growth of delay risk, p2 represents the budget overrun value, which is related to the efficiency of project cost control, p3 represents the quality decline value, p4 represents the project complexity, w1, w2, w3 and w4 are parameter weights, which are adjusted based on the impact analysis of historical data.

[0035] The present invention is improved in that the steps of obtaining the risk assessment index are specifically as follows:

[0036] S321: Summarize the potential risk points of each project, including the frequency of risk occurrence and the impact it has caused;

[0037] S322: Perform statistical analysis on each risk point and calculate the risk impact value r of each risk point based on the size and frequency of the risk. i , according to the historical data of the project and the risk management team, a criticality index i is assigned to each risk point i ;

[0038] S323: Combine the impact value and criticality index of each risk point and use the formula:

[0039]

[0040] Calculate the risk assessment index, where r i It represents the risk impact value of the i-th risk point, which is calculated by analyzing the risk point related data. i is the criticality index of the ith risk point, determined based on historical data, and n is the total number of identified risk points.

[0041] The present invention is improved in that the analysis steps of the risk control process are specifically as follows:

[0042] S411: Analyze various potential risks in the project using the risk assessment indicators, identify key risk points that affect the success of the project, and assign a severity score to each risk point k , assign expected efficiency e to each control process k ;

[0043] S412: Conduct cost-benefit analysis on the control process and calculate the cost-benefit coefficient c k , verify the financial feasibility of the control process, using the formula:

[0044]

[0045] Calculate the effect of the control process and adjust or select the optimal risk control process based on the results, where s krepresents the severity of the kth risk point, e k is the expected efficiency of the control process at the corresponding risk point, c k is the cost-effectiveness coefficient, and n is the total number of risk points identified.

[0046] The present invention is improved in that the steps of obtaining the project adjustment proposal are specifically as follows:

[0047] S421: Analyze the decision direction in the project using the risk control process, including the reallocation of costs, schedules and resources;

[0048] S422: Adjust project resource allocation based on decision analysis results, verify whether key risks are managed, and optimize resource utilization efficiency;

[0049] S423: Integrate the adjusted resource allocation and risk data, evaluate the impact of the adjustment on the project risk status and performability, and use the formula based on the integrated information:

[0050]

[0051] Get project adjustment proposals, where v j represents the impact value of the jth resource adjustment variable, d j is the dynamic adjustment factor of the jth variable, k j is the sensitivity coefficient of the impact factor, c j is a constant correction term used to balance the baseline impact or the preset adjustment deviation, and m is the total number of resource adjustment variables identified.

[0052] A science and technology project evaluation system based on big data, the system comprising:

[0053] The project data index module extracts historical execution data from the science and technology project database, including execution results, quality levels, and team feedback information, creates data indexes through MapReduce, and organizes the data to generate a project historical performance database;

[0054] The performance indicator selection module uses the historical performance database of the project and the random forest algorithm to identify key performance indicators, build a risk prediction model, analyze the correlation between performance and risk factors, and establish a risk and performance correlation model;

[0055] The new project risk assessment module applies the risk and performance association model and uses a support vector machine to perform performance and risk assessment based on the current project situation, analyze potential risk points, and draw a risk distribution map to obtain risk assessment indicators;

[0056] The project decision analysis module analyzes the potential risks of the project based on the risk assessment indicators, evaluates and optimizes the risk control process, adjusts the risk control process, and generates a project risk control plan;

[0057] The resource allocation and integration module allocates and optimizes resources according to the project risk control plan, reallocates and integrates project resources, verifies whether the resource allocation matches the project requirements, and formulates resource adjustment proposals.

[0058] Compared with the prior art, the advantages and positive effects of the present invention are:

[0059] In the present invention, MapReduce technology is used to improve the establishment of data indexes and the classification and organization of information, thereby achieving efficient data access and query response. Key performance indicators are identified from a large amount of historical data through a random forest algorithm, and an accurate risk prediction model is constructed, which improves the model's predictive ability and coverage of project risks. Support vector machines are used to adjust and optimize the performance and risk assessment of new projects, ensuring the pertinence and high adaptability of the assessment process, so that the assessment results more accurately reflect the actual situation of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of a scientific and technological project evaluation method based on big data is proposed for the present invention;

[0061] Figure 2 A flowchart for obtaining a historical performance library of a project of the present invention;

[0062] Figure 3 A flowchart for selecting key performance indicators of the present invention;

[0063] Figure 4 A flow chart for obtaining the risk prediction model of the present invention;

[0064] Figure 5 It is a flow chart for analyzing potential risk points of the project of the present invention;

[0065] Figure 6 A flow chart for obtaining risk assessment indicators of the present invention;

[0066] Figure 7 The following is a flowchart of the risk control process of the present invention;

[0067] Figure 8 Adaptation of the proposed analytical flow chart for the present invention project. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0069] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0070] Example

[0071] See also Figure 1 The present invention provides a technical solution: a scientific and technological project evaluation method based on big data, comprising the following steps:

[0072] S1: Based on the initialized science and technology project database, collect the historical execution data of existing projects, including execution results, quality levels and team feedback information, use MapReduce to establish and organize data indexes, and generate a project historical performance library;

[0073] S2: Based on the project historical performance database, the random forest algorithm is used to select key performance indicators in the data, build a risk prediction model, evaluate the relationship between historical performance and risk factors, and integrate multi-dimensional risk data to generate a risk and performance association model;

[0074] S3: Apply the risk and performance association model to the evaluation of new projects. Use support vector machines to evaluate the performance and risks of new projects in combination with the current project situation, analyze the potential risk points of the project, and draw a risk distribution map to obtain risk assessment indicators.

[0075] S4: Conduct project decision analysis based on risk assessment indicators, analyze the potential risk control process in the project, implement project resource allocation adjustments, and then integrate resource and risk information to obtain project adjustment proposals.

[0076] The project historical performance database includes performance scores, team feedback indicators, and project completion time. The risk and performance association model includes performance factor weights, risk probability scores, and association strength. Risk assessment indicators include risk sensitive points, risk impact scope, and risk occurrence probability. Project adjustment proposals include adjusted budget plans, resource allocation optimization plans, and schedule adjustment results.

[0077] See also Figure 2 , the specific steps for obtaining the project historical performance library are:

[0078] S111: Extract historical execution data from the initialized science and technology project database, including project execution results, quality levels, and team feedback information, and classify the data by coding or marking;

[0079] S112: Apply MapReduce to organize the collected data, process and index each type of data through the Map function, and then aggregate the data through the Reduce function to form an intermediate data set in a unified format;

[0080] S113: Summarize the processed data set using the formula:

[0081]

[0082] Determine the contribution weight of the difference data category to the performance evaluation and build a project historical performance library, where x, y, and z represent the scores of execution results, quality level, and team feedback, respectively, and w x 、w y and w z is the corresponding weight coefficient, and t is the calculated performance indicator.

[0083] The completion status of each project is collected, such as whether the project is completed on time, whether the predetermined goals are achieved, etc. The collected data is quantified. For example, full marks (e.g. 10 points) are awarded for on-time completion, 2 points are deducted for delays of less than one week, and more points are deducted for delays of more than one week. The average score is calculated as x (execution result) through the execution data of all relevant projects.

[0084] Conduct a quality assessment on the final results of each project, including product quality inspection or service satisfaction survey, set quality assessment standards, such as a score from 1 to 10, and calculate the average quality score of all projects as y (quality level).

[0085] Collect feedback from project team members on project processes, team collaboration, and project management, quantify the feedback through questionnaires, such as a score of 1 to 10, and average all collected feedback scores to obtain z (team feedback).

[0086] Assume that within a certain period of time, the following data is collected:

[0087] x (execution result): average score 8.0

[0088] y (quality level): average score 9.2

[0089] z (Team Feedback): Average score 7.5

[0090] Application formula:

[0091]

[0092] The calculation is as follows:

[0093]

[0094]

[0095]

[0096] t=8.5

[0097] The resulting comprehensive performance indicator t is 8.5, which comprehensively reflects the weighted average score of execution results, quality level and team feedback. In this way, the historical performance of each project can be quantified, and then the projects can be compared and analyzed.

[0098] See also Figure 3 , the steps for selecting key performance indicators are as follows:

[0099] S211: Extract various performance data of historical projects from the project historical performance database, including project execution results, quality levels, and team feedback information;

[0100] S212: Perform statistical analysis on the extracted data to identify key performance data points that are correlated with project success or failure;

[0101] S213: Through the analysis results, select performance data points that have predictive power for project risks as key performance indicators, and identify delays and cost overruns as key factors;

[0102] S214: Integrate the selected performance data points using the formula:

[0103]

[0104] Calculate the key performance indicator set, where P represents the project delay index, which is an indicator to measure whether the project is completed on time; Q represents the cost overrun index, which measures the cost management efficiency of the project; R is the relationship between P and Q, which optimizes the influence of the two indicators in performance evaluation; S represents customer satisfaction, which is directly related to the customer's satisfaction with the project; T and U represent team satisfaction and supply chain efficiency, respectively, which balance the influence of the indicators.

[0105] Evaluate a group of projects and record the number of days delayed for each project. For example, if the scheduled completion time of a project is 30 days and the actual completion time is 35 days, it is delayed by 5 days. The average number of days delayed for all projects is P.

[0106] Evaluate the budget and actual expenditure of the same set of projects. For example, if the project budget is 10,000 units and the actual expenditure is 12,000 units, the overspending is 2,000 units. Average the overspending percentages of all projects to get Q.

[0107] Define R as the project complexity score, which is based on the technical difficulty of the project, the resources required, and the size of the participating team, and rate the complexity of each project, for example from 1 to 10, and find the average of these scores;

[0108] Collect customer satisfaction scores on the project through questionnaire surveys, and average all received customer satisfaction S scores;

[0109] T is obtained through employee satisfaction survey, and U is obtained by evaluating the time efficiency and cost efficiency of the supply chain. Employee satisfaction T and supply chain efficiency U are scored separately and then the average is calculated.

[0110] Assume the following data:

[0111] P = 5 (average delay of 5 days)

[0112] Q = 20% (average cost overrun 20%)

[0113] R = 7 (average project complexity score 7)

[0114] S=8 (average customer satisfaction 8)

[0115] T=9 (average team satisfaction 9)

[0116] U=8 (average supply chain efficiency 8)

[0117] The calculation steps are:

[0118] Calculate the numerator: (P+2×Q)×R+3×S=(5+2×20)×7+3×8

[0119] Calculate the numerator: (5+40)×7+24=45×7+24=315+24=339

[0120] Calculate the denominator: T+U=9+8=17

[0121] Calculate KPI value:

[0122] KPI≈19.94 can be used to measure the key performance indicator of comprehensive project performance, which combines project delays, cost overruns, project complexity, customer and team satisfaction, and supply chain efficiency.

[0123] See also Figure 4 , the specific steps for obtaining the risk prediction model are:

[0124] S221: Select the indicators associated with project risk from the key performance indicators, input the selected key indicators to build a model, and perform project risk prediction using the formula:

[0125] RI=a×KPI 2 +b×KPI+c

[0126] The risk index is calculated based on the square and linear combination of key performance indicators, where RI stands for risk index, KPI stands for key performance indicator, and a, b, and c are model adjustment parameters;

[0127] S222: Test the historical project data, analyze the correlation between the prediction results of the model and the project risk, and adjust the model parameters a, b and c according to the results of the risk correlation analysis to obtain the risk prediction model.

[0128] Assume that there is a historical data set of projects, including the key performance indicators (KPIs) of each project. Through data analysis, it is found that the KPIs are highly correlated with the risk occurrence of the project.

[0129] Collect key performance indicators (KPIs) of historical projects and corresponding project risk occurrences (e.g., whether the budget is exceeded by more than 10%, whether the delay exceeds the scheduled time by more than 10%, etc.);

[0130] Use the least squares method to fit the coefficients a, b, and c. It is necessary to divide the historical project data into a training set and a test set, and use the training set data to train the model to optimize the coefficient values.

[0131] Assume that through data analysis, we obtain a=0.05, b=0.3, c=2, and assume that the KPI of a specific project is 10. Calculate the risk index KPI of the project:

[0132] RI=0.05×10 2 +0.3×10+2

[0133] RI=0.05×100+3+2

[0134] RI=5+3+2=10

[0135] The RI value represents the risk assessment score of the project, and the risk level of the project is predicted through the set model.

[0136] See also Figure 5 , the analysis steps of the potential risk points of the project are as follows:

[0137] S311: Collect performance data on current projects, including time management, cost control, and quality indicators, and identify patterns and trends associated with project execution;

[0138] S312: Apply set operations to the collected data to identify key areas in the data that show potential problems, including delay trends, budget overruns, or quality decline trends, and mark them as potential risk points;

[0139] S313: Evaluate the identified potential risk points using the formula:

[0140]

[0141] Calculate the impact and probability of each risk point, and set the risk level based on the historical risk data and current project parameters. Among them, p1 represents the delay trend value, indicating the severity of project delay, p1 square term represents the accelerated growth of delay risk, p2 represents the budget overrun value, which is related to the efficiency of project cost control, p3 represents the quality decline value, p4 represents the project complexity, w1, w2, w3 and w4 are parameter weights, which are adjusted based on the impact analysis of historical data.

[0142] p1 (delay trend value) is obtained by calculating the average delay time of the project;

[0143] p2 (budget overrun) is calculated by calculating the percentage difference between the project's budget and actual spending;

[0144] p3 (quality reduction value) is calculated by the qualified rate of quality control points;

[0145] p4 (project complexity) is derived by evaluating factors such as the project's size, technical difficulty, and the number of teams involved;

[0146] w1, w2, w3 and w4 are determined through historical data analysis to ensure that the contribution of each risk parameter to the final risk score is appropriately reflected.

[0147] Assume the specific data is as follows:

[0148] p1=5 (average delay of 5 days)

[0149] p2 = 20% (average overspending 20%)

[0150] p3 = 10 (quality decreases by 10%)

[0151] p4=8 (project complexity score)

[0152] w1=0.5, w2=0.3, w3=0.1, w4=0.1

[0153] The calculation steps are:

[0154] Calculate the contribution of the deferred trend:

[0155] Calculate the contribution of budget overrun: p2 = w2 × p2 = 0.3 × 20 = 6

[0156] Calculate the contribution of quality degradation: p3 = w3 × e -p3 =0.1×e -10 ≈0.1×0.000045=0.0000045

[0157] Calculate the contribution of project complexity: p4 = w4 × log (p4) = 0.1 × log (8) ≈ 0.1 × 0.903 = 0.0903

[0158] Total risk score: RiskScore = 12.5 + 6 + 0.0000045 + 0.0903 ≈ 18.5903

[0159] Risk score ≈ 18.5903 represents the risk assessment score of the project, and a higher score indicates a greater risk.

[0160] See also Figure 6 , the specific steps for obtaining risk assessment indicators are:

[0161] S321: Summarize the potential risk points of each project, including the frequency of risk occurrence and the impact it has caused;

[0162] S322: Perform statistical analysis on each risk point and calculate the risk impact value r of each risk point based on the size and frequency of the risk. i , according to the historical data of the project and the risk management team, a criticality index i is assigned to each risk point i ;

[0163] S323: Combine the impact value and criticality index of each risk point and use the formula:

[0164]

[0165] Calculate the risk assessment index, where r i It represents the risk impact value of the i-th risk point, which is calculated by analyzing the risk point related data. i is the criticality index of the ith risk point, determined based on historical data, and n is the total number of identified risk points.

[0166] r iis the risk impact value of the ith risk point, which is obtained by quantitative analysis of the frequency and impact of risk occurrence. i is the importance index of the ith risk point, based on the weight distribution of historical data.

[0167] Suppose a project has three main risk points:

[0168] Delay risk: high frequency, medium impact;

[0169] Overspending risk: medium frequency, high impact;

[0170] Quality risk: low frequency, very high impact.

[0171] Assume that the risk impact value and importance index are as follows:

[0172] r1=5(delay risk impact value)

[0173] r2=7(overspending risk impact value)

[0174] r3=9(quality risk impact value)

[0175] i1=0.6 (Delay risk importance index)

[0176] i2=0.8 (Overspending risk importance index)

[0177] i3=1.0 (Quality Risk Importance Index)

[0178] Calculate the contribution of each risk point:

[0179] Delay risk contribution = r1 × i1 = 5 × 0.6 = 3

[0180] Overspending risk contribution = r2×i2=7×0.8=5.6

[0181] Quality risk contribution = r3 × i3 = 9 × 1.0 = 9

[0182] Comprehensive calculation of risk assessment indicators:

[0183] RiskIndex=3+5.6+9=17.6

[0184] The calculated comprehensive risk assessment indicator RiskIndex is 17.6, which is calculated by quantifying the impact of each risk point and multiplying its importance. It can provide a risk metric for the project team to help prioritize and manage key risk points in the project.

[0185] See also Figure 7 , the analysis steps of the risk control process are as follows:

[0186] S411: Use risk assessment indicators to analyze the various potential risks in the project, identify the key risk points that affect the success of the project, and assign a severity score to each risk point k , assign expected efficiency e to each control process k ;

[0187] S412: Conduct cost-benefit analysis on the control process and calculate the cost-benefit coefficient c k , verify the financial feasibility of the control process, using the formula:

[0188]

[0189] Calculate the effect of the control process and adjust or select the optimal risk control process based on the results, where s k represents the severity of the kth risk point, e k is the expected efficiency of the control process at the corresponding risk point, c k is the cost-effectiveness coefficient, and n is the total number of risk points identified.

[0190] s k is the severity of the kth risk point, which is usually obtained through quantitative analysis in the risk assessment process, such as the possible loss measurement;

[0191] e k is the expected efficiency of the control process for the kth risk point, estimated based on historical data of similar situations and expert evaluation;

[0192] c k is the cost-effectiveness coefficient, which calculates the economic benefit of risk reduction per unit of input and is obtained through financial analysis;

[0193] n is the total number of risk points identified in the project.

[0194] Assume the following parameters:

[0195] Risk point 1:

[0196] s1 = 8 (e.g., 8 represents the loss due to delay), e1 = 0.9 (expected efficiency of the control process), c1 = 0.5 (cost-effectiveness coefficient);

[0197] Contribution of risk point 1 = 8 × 0.9 × 0.5 = 3.6.

[0198] Risk point 2:

[0199] s2 = 5 (e.g., 5 represents the loss due to overspending), e2 = 0.7, c2 = 0.3;

[0200] Contribution of risk point 2 = 5 × 0.7 × 0.3 = 1.05.

[0201] Risk point 3:

[0202] s3=6 (for example, 6 represents the loss due to quality problems), e3=0.8, c3=0.4;

[0203] Contribution of risk point 3 = 6 × 0.8 × 0.4 = 1.92.

[0204] Calculate the control process effectiveness score:

[0205] Control process effectiveness score = 3.6 + 1.05 + 1.92 = 6.57

[0206] The calculated control process effectiveness score of 6.57 represents the overall control effect on project risks. The score provides quantitative indicators of project risk management through risk point severity, expected efficiency of the control process, and cost-effectiveness coefficient assessment.

[0207] See also Figure 8 , the specific steps for obtaining the project adjustment proposal are:

[0208] S421: Use the risk control process to analyze decision directions in the project, including cost, schedule, and resource reallocation;

[0209] S422: Adjust project resource allocation based on decision analysis results, verify whether key risks are managed, and optimize resource utilization efficiency;

[0210] S423: Integrate the adjusted resource allocation and risk data, evaluate the impact of the adjustment on the project risk status and performability, and use the formula based on the integrated information:

[0211]

[0212] Get project adjustment proposals, where v j represents the impact value of the jth resource adjustment variable, d j is the dynamic adjustment factor of the jth variable, k j is the sensitivity coefficient of the impact factor, c j is a constant correction term used to balance the baseline impact or the preset adjustment deviation, and m is the total number of resource adjustment variables identified.

[0213] v j The impact value of the jth resource adjustment variable, including the quantitative value of capital flow, human resources or material allocation, is evaluated based on the planned and actual use of each resource;

[0214] d j It is a dynamic adjustment factor that reflects the impact of changes in the project phase or external environment on resource demand, and is derived through analysis of historical data and current market conditions;

[0215] k j is the sensitivity coefficient, which indicates the weight of the impact of resource adjustment on project success and is set based on previous experience and risk assessment of the project;

[0216] c j is a constant correction term used to balance the baseline effect or preset adjustment deviation, which is set based on the specification standard;

[0217] m is the total number of resource adjustment variables identified, determined based on the complexity and diversity of the project.

[0218] Assume that a project has three key resource adjustment variables: cash flow, human resources, and material allocation.

[0219] Assume the following parameters:

[0220] v1=100, v2=50, v3=30 (the impact value of funds, manpower, and materials)

[0221] d1=1.1, d2=1.0, d3=0.9 (dynamic adjustment factor)

[0222] k1=0.5, k2=0.3, k3=0.2 (sensitivity coefficient)

[0223] c1=5, c2=3, c3=2 (constant correction term)

[0224] Calculate the contribution of each resource variable:

[0225] Cash flow contribution = (100 × 1.1 × 0.5 + 5) = 55 + 5 = 60

[0226] Human resource contribution = (50 × 1.0 × 0.3 + 3) = 15 + 3 = 18

[0227] Contribution of material allocation = (30 × 0.9 × 0.2 + 2) = 5.4 + 2 = 7.4

[0228] Comprehensive calculation of adjusted scores:

[0229] Adjustment Score=60+18+7.4=85.4

[0230] The calculated Adjustment Score = 84 reflects the combined impact of the current project status and resource adjustment variables, helping the project management team evaluate the effectiveness of different resource adjustments.

[0231] A science and technology project evaluation system based on big data, the system includes:

[0232] The project data index module extracts historical execution data from the science and technology project database, including execution results, quality levels, and team feedback information, creates data indexes through MapReduce, and organizes the data to generate a project historical performance database;

[0233] The performance indicator selection module uses the project historical performance database and the random forest algorithm to identify key performance indicators, build a risk prediction model, analyze the correlation between performance and risk factors, and establish a risk and performance correlation model;

[0234] The new project risk assessment module applies the risk and performance association model and uses support vector machines to conduct performance and risk assessment based on the current project situation, analyze potential risk points, and draw risk distribution maps to obtain risk assessment indicators;

[0235] The project decision analysis module analyzes potential project risks based on risk assessment indicators, evaluates and optimizes the risk control process, adjusts the risk control process, and generates a project risk control plan;

[0236] The resource allocation and integration module allocates and optimizes resources according to the project risk control plan, reallocates and integrates project resources, verifies whether the resource allocation matches project requirements, and formulates resource adjustment proposals.

[0237] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A scientific and technological project evaluation method based on big data, characterized in that: The following steps are involved: Based on the initialized science and technology project database, the historical execution data of existing projects, including execution results, quality levels and team feedback information, are collected. Using MapReduce, data indexes are established and sorted to generate a project historical performance database. Based on the historical performance database of the project, the random forest algorithm is used to select key performance indicators in the data, build a risk prediction model, evaluate the relationship between historical performance and risk factors, and integrate multi-dimensional risk data to generate a risk and performance association model; Apply the risk and performance association model to the new project evaluation, use the support vector machine, combine the current project situation, conduct performance and risk evaluation on the new project, analyze the potential risk points of the project, draw a risk distribution map, and obtain the risk evaluation index; Based on the risk assessment indicators, conduct project decision analysis, analyze the potential risk control process in the project, implement project resource allocation adjustments, integrate resource and risk information, and obtain project adjustment proposals; The steps for selecting the key performance indicators are as follows: Extracting performance data of historical projects from the project historical performance database, including project execution results, quality levels, and team feedback information; Conduct statistical analysis on the extracted data to identify key performance data points that correlate with project success or failure; By analyzing the results, we select performance data points that have predictive power for project risks as key performance indicators, and identify delays and cost overruns as key factors; To integrate the selected performance data points, use the formula: Calculate the key performance indicator set, where P represents the project delay index, which measures whether the project is completed on time; Q represents the cost overrun index, which measures the cost management efficiency of the project; R is the relationship between P and Q, which optimizes the influence of the two indicators in performance evaluation; S represents customer satisfaction, which is directly related to the customer's satisfaction with the project; T and U represent team satisfaction and supply chain efficiency, respectively, which balance the influence of indicators; The steps for obtaining the risk prediction model are specifically as follows: From the key performance indicators, select indicators associated with project risk, input the selected key indicators to build a model, and perform project risk prediction using the formula: RI=a×KPI 2 +b×KPI+c The risk index is calculated based on the square and linear combination of key performance indicators, where RI stands for risk index, KPI stands for key performance indicator, and a, b, and c are model adjustment parameters; Test historical project data, analyze the correlation between the model's prediction results and project risks, and adjust the model parameters a, b and c according to the results of the risk correlation analysis to obtain a risk prediction model; The steps for obtaining the risk assessment indicators are specifically as follows: Summarize the potential risk points of each project, including the frequency of risk occurrence and the impact it has caused; Perform statistical analysis on each risk point and calculate the risk impact value r of each risk point based on the size and frequency of the risk i , according to the historical data of the project and the risk management team, a criticality index i is assigned to each risk point i ; Combine the impact value and criticality index of each risk point and use the formula: Calculate the risk assessment index, where r i It represents the risk impact value of the i-th risk point, which is calculated by analyzing the risk point related data. i is the criticality index of the ith risk point, determined based on historical data, and n is the total number of identified risk points.

2. The scientific and technological project evaluation method based on big data according to claim 1 is characterized in that: The specific steps for obtaining the project historical performance library are as follows: Extract historical execution data from the initialized science and technology project database, including project execution results, quality levels, and team feedback information, and classify the data through coding or labeling; Use MapReduce to organize the collected data, process and index each type of data through the Map function, and then aggregate the data through the Reduce function to form an intermediate data set in a unified format; Summarize the processed data set using the formula: Determine the contribution weight of the difference data category to the performance evaluation and build a project historical performance library, where x, y, and z represent the scores of execution results, quality level, and team feedback, respectively, and w x 、w y and w z is the corresponding weight coefficient, and t is the calculated performance indicator.

3. The scientific and technological project evaluation method based on big data according to claim 1 is characterized in that: The specific steps for analyzing the potential risk points of the project are as follows: Collect performance data on current projects, including time management, cost control, and quality metrics, and identify patterns and trends associated with project execution; Apply set operations to the collected data to identify key areas in the data that show potential problems, including delay trends, budget overruns, or quality decline trends, and mark them as potential risk points; Evaluate the identified potential risk points using the formula: Calculate the impact and probability of each risk point, and set the risk level based on the historical risk data and current project parameters. Among them, p1 represents the delay trend value, indicating the severity of project delay, p1 square term represents the accelerated growth of delay risk, p2 represents the budget overrun value, which is related to the efficiency of project cost control, p3 represents the quality decline value, p4 represents the project complexity, w1, w2, w3 and w4 are parameter weights, which are adjusted based on the impact analysis of historical data.

4. The scientific and technological project evaluation method based on big data according to claim 1 is characterized in that: The analysis steps of the risk control process are specifically as follows: Use the risk assessment indicators to analyze the various potential risks in the project, identify the key risk points that affect the success of the project, and assign a severity score to each risk point k , assign expected efficiency e to each control process k ; Conduct a cost-benefit analysis on the control process and calculate the cost-benefit coefficient c k , verify the financial feasibility of the control process, using the formula: Calculate the effect of the control process and adjust or select the optimal risk control process based on the results, where s k represents the severity of the kth risk point, e k is the expected efficiency of the control process at the corresponding risk point, c k is the cost-effectiveness coefficient, and n is the total number of risk points identified.

5. The scientific and technological project evaluation method based on big data according to claim 1 is characterized in that: The specific steps for obtaining the project adjustment proposal are as follows: Utilize the risk control process to analyze decision making in the project, including reallocation of costs, schedules, and resources; Adjust project resource allocation based on decision analysis results, verify whether key risks are managed, and optimize resource utilization efficiency; Integrate the adjusted resource allocation and risk data, evaluate the impact of the adjustment on the project risk profile and performability, and based on the integrated information, use the formula: Get project adjustment proposals, where v j represents the impact value of the j-th resource adjustment variable, d j is the dynamic adjustment factor of the jth variable, k j is the sensitivity coefficient of the impact factor, c j is a constant correction term used to balance the baseline impact or the preset adjustment deviation, and m is the total number of resource adjustment variables identified.

6. The science and technology project evaluation system based on big data is characterized by: According to any one of claims 1 to 5, the method for evaluating scientific and technological projects based on big data is implemented, and the system comprises: The project data index module extracts historical execution data from the science and technology project database, including execution results, quality levels, and team feedback information, creates data indexes through MapReduce, and organizes the data to generate a project historical performance database; The performance indicator selection module uses the historical performance database of the project and the random forest algorithm to identify key performance indicators, build a risk prediction model, analyze the correlation between performance and risk factors, and establish a risk and performance correlation model; The new project risk assessment module applies the risk and performance association model and uses a support vector machine to perform performance and risk assessment based on the current project situation, analyze potential risk points, and draw a risk distribution map to obtain risk assessment indicators; The project decision analysis module analyzes the potential risks of the project based on the risk assessment indicators, evaluates and optimizes the risk control process, adjusts the risk control process, and generates a project risk control plan; The resource allocation and integration module allocates and optimizes resources according to the project risk control plan, reallocates and integrates project resources, verifies whether the resource allocation matches the project requirements, and formulates resource adjustment proposals.

Images