A method and system for comprehensive evaluation of construction cost in whole process of existing building reconstruction

Abstract

The application discloses a kind of whole-process cost comprehensive evaluation method and system for existing building reconstruction, belong to building engineering cost management and intelligent evaluation technical field.The method includes: the four-stage evaluation index system of construction planning, design, construction and operation is built;Using differentiating expert weighted analytic hierarchy process, and the weight is given according to the professional title, working years and stage familiarity;Introducing fuzzy entropy quantization evaluation divergence, and the comprehensive weight is corrected with reliability weight;Introducing time attenuation factor for planning and design stage;Using ordered weighted average operator to calculate aggregation score;Based on historical data, set dynamic early warning threshold, identify weak links and generate improvement suggestions.The application also provides an online evaluation system for implementing the method, supports multi-user collaborative scoring, cloud deployment and cross-project knowledge accumulation.The application can scientifically quantify the whole-process cost management effect of existing building reconstruction, and provides a systematic evaluation tool for cost control of urban renewal projects.

Description

Technical Field

[0001] This invention belongs to the field of construction project cost management and intelligent evaluation technology. Specifically, it relates to a comprehensive cost evaluation method and system for the entire process of existing building renovation. It is applicable to the intelligent quantitative evaluation of cost control levels in urban renewal scenarios such as energy-saving renovation and building quality improvement. It can be deployed on the cloud or local server to achieve multi-user collaborative evaluation and knowledge accumulation. Background Technology

[0002] As my country's urbanization shifts from incremental expansion to stock renewal, the renovation of existing buildings has become an important aspect of urban renewal. Against this backdrop, the cost management of urban renewal projects exhibits high complexity, uncertainty, and diverse stakeholders, significantly increasing the difficulty of managing costs throughout the entire process.

[0003] Existing engineering cost evaluation methods have the following main shortcomings: their indicator systems are mostly geared towards new construction projects, failing to depict the unique cost risk characteristics of existing building renovations, and are often limited to cost control during the construction phase, lacking coverage of the entire process from planning to operation and maintenance; in determining weights, the traditional analytic hierarchy process (AHP) relies on a single expert group to make a global judgment on all stage indicators, ignoring the differences in professional understanding among different participating parties, resulting in insufficient effectiveness of weights; in fuzzy comprehensive evaluation, expert opinions are simply averaged or majority-based, failing to quantify the reliability and consistency of evaluation information, making them susceptible to interference from extreme opinions and affecting the objectivity of the results; the scores of each stage are weighted equally, failing to consider the decay effect of evaluation information over time, making it difficult to truly reflect the current management status of the project, and not conducive to the dynamic management of long-term projects; when aggregating secondary indicators, equal-weighted averaging is used, with insufficient attention paid to low-scoring indicators, making it difficult to effectively identify weak links in cost control. At the system level, existing evaluation systems have low integration and limited functionality. The modules are fragmented and lack cross-project knowledge accumulation and reuse capabilities. Furthermore, they do not support multi-user collaborative scoring and cloud deployment, making it difficult to adapt to the actual evaluation scenarios of large-scale renovation projects involving multiple participating units. This results in low evaluation efficiency and a lack of knowledge accumulation.

[0004] In summary, there is an urgent need to establish a fuzzy comprehensive evaluation method that can scientifically quantify management effectiveness and effectively identify weak links throughout the entire process of existing building renovation, as well as an integrated, intelligent online evaluation system that supports cross-project knowledge accumulation. Summary of the Invention

[0005] 1. This invention provides a comprehensive cost evaluation method for the entire process of renovating existing buildings, comprising the following steps:

[0006] Step S1: Construct an evaluation index system;

[0007] Step S11: Establish an evaluation index system consisting of four primary indicators (planning, design, construction, and operation and maintenance) and 15 corresponding secondary indicators;

[0008] The secondary indicators in the planning stage include: A1 The current status and safety performance assessment of the original building structure; A2 The sufficiency and depth of the renovation project planning scheme; A3 The level of detail in the on-site survey work; A4 The degree of identification and response to policies, regulations and norms; and A5 The clarity of project approval and funding arrangements.

[0009] The secondary indicators in the design phase include: B1 the economic rationality of the renovation design scheme and B2 the degree of design response to the structural constraints of the existing building.

[0010] The secondary indicators for the construction phase include: C1 Neighborhood relations management and disturbance control during construction; C2 Completeness of on-site visa and change management system during construction; C3 Existing building renovation project experience of construction companies; C4 Reasonableness of risk reserve funds for renovation projects; C5 Effectiveness of dynamic cost management mechanism for projects; C6 Completeness of original building structure protection and safe construction plan.

[0011] The secondary indicators in the operation and maintenance phase include: D1 the level of facility and equipment aging identification, maintenance and update management, and D2 the level of energy system operation efficiency and energy consumption management.

[0012] Step S2: Calculate the weights of the primary indicators;

[0013] After the evaluation index system is determined, the weights of the primary indicators are calculated. This step employs the differentiated expert weighted analytic hierarchy process to fully consider the differences in experts' understanding of the overall framework.

[0014] Step S21: Calculate the expert weight w of the primary indicator. ij And integrate expert weight vector W 专家 ;

[0015] According to the expert scoring criteria, expert j's professional title level R... j Years of service Y j and the degree of familiarity with the problem in the i-th stage of the research F i,j Scoring is performed, a comprehensive score is calculated, and the score is normalized to obtain the expert weight w. ij The expert scoring criteria are shown in Table 1:

[0016] w ij =

[0017] In the formula: S j = R j + Y j + F i,j And satisfy .

[0018] Table 1 Expert Scoring Criteria

[0019] The expert weights are integrated to obtain the expert weight vector W for this stage. 专家 :

[0020] W 专家 =[w i1 ,w i2 ,…,w ij ] T

[0021] Step S211: Closed-loop feedback correction of expert weights;

[0022] This step is not performed during the initial evaluation. It is done after the evaluation of one or more actual projects is completed, based on the actual cost deviation rate δ at each stage of the project. actual The deviation rate δ between the predicted scores of each stage calculated in step S6 and the actual predicted scores is calculated. pred The differences between them are adjusted according to the following formula, which modifies the original weights of each expert involved in the project at that stage. :

[0023]

[0024] Where k is a preset coefficient, and it is recommended that the value range be 0.15~0.25.

[0025]

[0026] Where η is the preset learning rate and 0 < η ≤ 0.2; the corrected weights are normalized and stored in the expert weight database as the prior weights of the expert in the evaluation of subsequent projects of the same type.

[0027] Step S22: Establish the primary indicator judgment matrix A;

[0028] Based on the comparison scale table of judgment matrix elements, the importance of the four primary indicators of planning, design, construction, and operation and maintenance is compared pairwise to establish a primary indicator judgment matrix A. The comparison scale table of judgment matrix elements is shown in Table 2.

[0029] Table 2 Comparison Scale Table for Matrix Elements

[0030] Step S23: Calculate the weights of the primary indicator judgment matrix A using the geometric mean method. And integrate the first-level indicator weight matrix W 指标 :

[0031]

[0032]

[0033] In the formula: i is the i-th indicator, and j is the j-th expert.

[0034] Step S24: Consistency check:

[0035]

[0036] In the formula, CI is the consistency index; RI is the average random consistency index; λ max The largest eigenvalue of the judgment matrix is ​​CR. If CR < 0.1, then the judgment matrix passes the consistency test.

[0037] Step S25: Calculate the integrated weight vector W of the primary indicators. 一级 :

[0038] W 一级 =W 专家 W 指标 =[ W A W B W C W D ]

[0039] Step S3: Calculate the integrated weights of the secondary indicators for each stage;

[0040] For each secondary indicator under a primary indicator, the same differentiated weighted integration method as in step 2 is used to calculate the expert weights of the secondary indicators and construct a judgment matrix, thereby calculating the weights W of the secondary indicators. 二级 .

[0041] Step S4: Calculate the comprehensive weight based on the obtained primary and secondary indicator weights and classify them by importance;

[0042] Step S41: Calculate the overall weight W 综合 :

[0043] W 综合 =W 一级 W 二级

[0044] Step S42: Classify importance based on the classification boundaries determined by the numerical distribution of the comprehensive weight calculation results;

[0045] Based on preset thresholds, the combined weight values ​​of the 15 secondary indicators are divided into three levels: very important, important, and general.

[0046] Among them: a comprehensive weight value ≥ 0.1 is "very important"; 0.1 > comprehensive weight value ≥ 0.03 is "important"; and a comprehensive weight value < 0.03 is "average".

[0047] Step S5: Multi-level fuzzy comprehensive evaluation; introduce fuzzy entropy and reliability weights to quantify and correct the discrepancies in expert evaluation opinions, and construct a fuzzy evaluation matrix with integrated weights.

[0048] Step S51: Construct a five-level comment set and corresponding score ranges;

[0049] A five-level evaluation system was established: "Fully implemented, with significant control effect", "Good implementation, with obvious control effect", "Basically implemented, with average control effect", "Insufficient implementation, with weak control effect", and "Lack of implementation, with ineffective control effect".

[0050] The corresponding score intervals are [100, 90), [90, 80), [80, 60), [60, 40), and [40, 0); the midpoint of each score interval is taken as the score vector V= .

[0051] Step S52: Establish the fuzzy evaluation matrix :

[0052]

[0053] In the formula: is the original membership degree, and is the proportion of the number of people who gave the jth rating to the i-th factor out of the total number of participants.

[0054] Step S53: Calculate the fuzzy entropy E of each index. i :

[0055]

[0056] Step S54: Calculate the fusion weight W 融合 The weights were then normalized and fused in stages to obtain... :

[0057]

[0058] In the formula: ρ i For reliability weights and ρ i =1 E i .

[0059]

[0060] In the formula: is the weighting of the secondary indicators, and n is the number of secondary indicators in the period.

[0061] Step S6: Calculate the score for each level;

[0062] After constructing the fuzzy evaluation matrix, a time decay factor is introduced to correct the membership matrix Z for the planning and design stages, resulting in... Calculate the score S of the first-level indicator layer in sequence. 一级 Target layer score S T and the initial score S of the secondary indicator layer 二级 The aggregate score S is calculated using an ordered weighted average operator for the secondary indicator layer. OWA .

[0063] Step S61: Introduce the time decay factor to calculate the membership matrix of the first-level index layer. Compared with the first-level indicator score S 一级 :

[0064] Z=

[0065] In the formula: F is the original membership matrix of the secondary index.

[0066] The corrected membership matrix is ​​obtained by multiplying the membership matrices of the planning and design phases by the time decay factor. No time decay is applied during the construction and operation and maintenance phases, i.e. =Z.

[0067] =Z

[0068] in The time interval from the end of this phase to the evaluation point, in years; The preset attenuation coefficient has a value range of 0 < ≤0.1.

[0069] S 一级 =

[0070] Step S62: Calculate the membership degree Z of the target layer T. T With score S T :

[0071] Z T =W 一级 F T

[0072] In the formula: F T An integrated fuzzy evaluation matrix is ​​generated for the membership degree of each primary indicator.

[0073] S T =Z T

[0074] Step S63: Calculate the initial score S of the secondary indicator layer. 二级 ;

[0075] Step S631: Calculate the initial score S for each secondary indicator. 二级 And determine its evaluation level by combining the evaluation level score range table;

[0076] S 二级 =F V

[0077] Step S64: Introduce the ordered weighted average operator to calculate the aggregate score S of the secondary indicator layer. OWA ;

[0078] Step S641: Initial score S of secondary indicators 二级 Sort in descending order to get ;

[0079] Step S642: Calculate the position weight of each secondary indicator :

[0080] =

[0081] In the formula: j is the initial score in descending order, and n is the number of indicators.

[0082] Step S643: Calculate the combined weights of each secondary indicator. :

[0083]

[0084] In the formula: The weights are the weights corresponding to the comprehensive weights of the secondary indicators after sorting.

[0085] Step S644: Calculate the aggregate score S of the secondary indicators. OWA :

[0086] S OWA :

[0087] Step S7: Attribution Diagnosis and Improvement Suggestion Generation

[0088] Step S71: Attribution diagnosis;

[0089] Collect historical score data of similar existing building renovation projects within the past 5 years, calculate the mean and standard deviation of the score for each secondary indicator, and set dynamic early warning thresholds. Let k = 1.2. The initial score S for this project... 二级 Corresponding Compare.

[0090] Furthermore, the step S71, "collecting historical score data of existing building renovation projects of the same type within 5 years," specifically refers to the following: the historical score data comes from the evaluation result database of at least 5 existing building renovation projects of the same type that this method has previously applied to. The database stores the score values ​​of each secondary indicator of each project and its corresponding basic characteristics. If there are fewer than 5 historical data points, publicly released benchmark data for existing building renovation cost management evaluation in the industry will be used as a substitute.

[0091] Step S72: Generate improvement suggestions based on the pre-stored mapping rule base of "low score indicators - improvement measures - responsible entities".

[0092] The rule base is built based on post-evaluation reports and expert interviews of no fewer than 10 existing building renovation projects. It adopts a multi-dimensional matching strategy: when a low-scoring indicator exactly matches the indicator name in the rule base, the corresponding measures are directly output; when there is no exact match, fuzzy matching is performed according to the stage and weight level of the indicator, and the rule with the highest similarity is selected.

[0093] 2. This invention provides a full-process cost management and evaluation system for existing building renovation projects. The system includes:

[0094] Indicator system storage module: used to store the four-stage evaluation indicator system described in Example 1, including 4 primary indicators, 15 secondary indicators and corresponding comprehensive weight classification information;

[0095] Indicator scoring and data collection module: Receives the rating data of evaluators for each secondary indicator of the target project through the user interface, supports multiple users to score at the same time, and automatically performs data sorting and normalization.

[0096] Weight Calculation Module: Used to implement the analytic hierarchy process (AHP) and closed-loop feedback correction for differentiated expert weighted integration; this module includes: expert information management submodule, judgment matrix construction submodule, consistency verification submodule, weighted integration calculation submodule, and expert weight feedback update submodule;

[0097] Fuzzy comprehensive evaluation module: includes membership degree statistics submodule, fuzzy entropy calculation submodule, fusion weight calculation submodule, OWA synthesis submodule and time decay submodule;

[0098] Attribution diagnosis module: includes dynamic threshold calculation submodule, rule base storage submodule, and suggestion generation submodule;

[0099] This system can be deployed on a cloud server or a local server and accessed through a web browser or mobile app. It enables online evaluation and diagnosis of the cost management level of existing building renovation projects and supports cross-project expert weight learning and knowledge accumulation. Beneficial effects

[0100] Compared with the prior art, the present invention has the following beneficial effects:

[0101] A comprehensive cost evaluation index system covering the four stages of planning, design, construction, and operation and maintenance of existing building renovation has been established, enabling the evaluation results to fully reflect the cost control level of the renovation project throughout its entire life cycle.

[0102] The differentiated expert weighted analytic hierarchy process is adopted, which assigns weights to experts based on their professional titles, years of experience, and familiarity with the issues at specific stages. This makes the weights of indicators at each stage more aligned with the actual engineering project, and improves the scientific nature and relevance of the weight determination.

[0103] By introducing fuzzy entropy to quantify the degree of dispersion of membership, the weight contribution of indicators with low expert consensus is automatically reduced, thereby improving the robustness and objectivity of the evaluation results.

[0104] By introducing a time decay factor to correct the membership matrix of the planning and design phases, the comprehensive score is made to focus more on the recent performance of the construction and operation and maintenance phases, thereby reducing the unreasonable impact of early evaluation information on the current status assessment in long-cycle projects.

[0105] An ordered weighted average operator is used to aggregate the scores of secondary indicators, giving higher weights to low-scoring indicators, making the evaluation results more sensitive to weak links and providing a reliable basis for attribution diagnosis.

[0106] This integrated online evaluation system combines six core modules: indicator system storage, score collection, weight calculation, fuzzy comprehensive evaluation, attribution diagnosis, and visualization output. It supports multi-user collaborative scoring and cloud deployment, enables cross-project expert weight learning and historical data accumulation, and forms a complete closed loop from data collection to diagnostic recommendations. It provides a systematic, intelligent, and sustainable engineering evaluation tool for the cost management of existing building renovation projects. Attached Figure Description

[0107] Figure 1 This is an overall flowchart of the method of the present invention.

[0108] Figure 2 This is the flowchart for step S2.

[0109] Figure 3 This is the flowchart for step S5.

[0110] Figure 4 Here is the flowchart for step S6. (a) is the flowchart for step S6; (b) is the flowchart for step S64.

[0111] Figure 5 This is a block diagram of the system of the present invention. Detailed Implementation

[0112] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0113] Example 1:

[0114] This embodiment provides a comprehensive cost evaluation method for the entire process of existing building renovation, which can be applied to the cost management evaluation of commercial and residential building renovation projects.

[0115] The project is a frame-structured commercial and residential building, and the renovation goal is to transform the original old shopping mall and office space into high-end serviced apartments.

[0116] Step S1: Construct an evaluation index system;

[0117] Step S11: Establish an evaluation index system consisting of four primary indicators (planning, design, construction, and operation and maintenance) and their corresponding 15 secondary indicators, as shown in Table 3.

[0118] Table 3 Evaluation Index System

[0119] Step S2: Calculate the weights of the primary indicators;

[0120] Step S21: Calculate the expert weight vector W for the first-level indicators. 专家 ;

[0121] The evaluation is based on three dimensions: the expert's professional title, years of experience, and familiarity with the overall framework of whole-process cost management.

[0122] Taking Expert 1 as an example: Senior professional title: 6 points; over 10 years of professional experience: 6 points; very familiar with the overall framework of whole-process cost management: 6 points; Overall score S1 = 6 + 6 + 6 = 18. The expert weight vector W is obtained after normalizing the overall scores of the 13 experts. 专家 Among them, satisfying The comprehensive scores and normalized weights of the 13 experts in the evaluation of the primary indicators are shown in Table 4:

[0123] Table 4 Expert Composite Score and Normalized Weight

[0124] Step S22: Based on the comparison scale table of the judgment and comparison matrix elements, compare the importance of the four primary indicators—planning, design, construction, and operation and maintenance—pair by pair to establish judgment matrix A. For example, the judgment and comparison matrix constructed by expert 1... as follows:

[0125]

[0126] Step S23: Calculate and integrate the weight vector of the first-level indicators using the geometric mean method:

[0127]

[0128] Step S24: Consistency check:

[0129] [AW] 专家1 =[2.2413,0.5689,1.0288,0.2133] T

[0130]

[0131]

[0132] The average random consistency index (RI) values ​​are shown in Table 5.

[0133] Table 5 RI Value Table

[0134] Each expert constructed a judgment matrix, calculated the weight vector, and conducted a consistency test. The results of the first-level indicators are summarized in Table 6, and all of them passed the consistency test.

[0135] Table 6 Summary of Primary Indicators

[0136] Step S25: Transfer the expert weight vector W 专家 With the first-level indicator weight matrix W 指标 Multiplication is performed to obtain the integrated weights of the primary indicators. After normalization, the weights W of the primary indicators are calculated. 一级 As shown in Table 7:

[0137] Table 7 Weights of Primary Indicators

[0138] Step S3: Calculate the expert weights for the secondary indicators;

[0139] For the 15 secondary indicators, the comprehensive scoring dimensions of experts at each stage are the same as in step two. The difference is that the familiarity level refers to the familiarity with the problems at each specific stage, rather than the overall framework of the whole process.

[0140] (I) Calculation of weights for secondary indicators during the planning phase

[0141] The experts were comprehensively evaluated based on three dimensions: their professional title, years of experience, and familiarity with issues in the planning stage. The comprehensive scores of the 13 experts were then normalized to obtain a weight vector for the planning stage experts. As shown in Table 8:

[0142] Table 8 Expert Scores and Weights in the Planning Phase

[0143] Construct the judgment matrix for the secondary indicators and calculate the weight vector. Calculate the consistency index (CI) and consistency ratio (CR) for each expert. The calculated CR values ​​for the 13 experts range from 0.0324 to 0.0898, all less than 0.1, thus all passed the consistency test.

[0144] The weight matrix W is constructed by combining the weight vectors of the secondary indicators during the planning phase. 策划 Then, combined with the expert weight vector The integrated weights of each secondary indicator in the planning stage are obtained by performing multiplication operations, as shown in Table 9:

[0145] Table 9. Integrated Weights of Secondary Indicators in the Planning Phase

[0146] (II) Calculation of integrated weights for each secondary indicator during the design phase

[0147] The same method is used to calculate the expert weights during the design phase. After normalization, the expert weight vector for the design phase is obtained. Taking expert 5 as an example: Senior professional title: 6 points; over 10 years of experience: 6 points; very familiar with the design phase: 6 points; total score = 18; normalized weight is 0.1304. The second-order matrix RI is 0, so the expert's CR value is 0, satisfying consistency.

[0148] The integrated weights of the secondary indicators in the design phase are: B1 = 0.1865, B2 = 0.8137.

[0149] (III) Calculation of integrated weights for each secondary indicator during the construction phase

[0150] The expert weights for the construction phase are calculated using the same method, and then normalized to obtain the expert weight vector for the construction phase. The CR values ​​of all expert judgment matrices ranged from 0.0379 to 0.0769, all less than 0.1, and all passed the consistency test. The stage weights of the secondary indicators in the construction phase are shown in Table 10.

[0151] Table 10 Integrated Weights of Secondary Indicators in the Construction Phase

[0152] (iv) Calculation of weights for each secondary indicator during the operation and maintenance phase

[0153] The same method is used to calculate the expert weights for the operation and maintenance phase, and after normalization, the expert weight vector for the operation and maintenance phase is obtained. Since the second-order matrix RI is 0, the expert CR values ​​are all 0, satisfying the consistency requirement.

[0154] The integrated weights of each secondary indicator during the operation and maintenance phase are: D1 = 0.6407, D2 = 0.3593.

[0155] Step S4: Calculate the overall weight and classify the importance;

[0156] Step S41: Multiply the stage weight of each secondary indicator by the weight of its corresponding primary indicator to obtain the overall weight of each secondary indicator throughout the entire process, such as... .

[0157] Step S42: Based on the overall weight value, the 15 secondary indicators are divided into three levels: an overall weight value ≥ 0.10 is "Very Important", an overall weight value between 0.03 and 0.10 is "Important", and an overall weight value < 0.03 is "General". The overall weight and importance classification results of the secondary indicators are shown in Table 11:

[0158] Table 11. Overall Weight and Importance Classification of Secondary Indicators Step S5: Multi-level fuzzy comprehensive evaluation;

[0159] Step S51: Construct a five-level comment set and corresponding score ranges;

[0160] A five-level evaluation system is constructed: "Sufficient implementation, significant control effect," "Good implementation, obvious control effect," "Basic implementation, average control effect," "Insufficient implementation, weak control effect," and "Missing implementation, ineffective control effect." The corresponding score ranges are [100, 90), [90, 80), [80, 60), [60, 40), and [40, 0). The midpoint of each score range is taken as the score vector V = .

[0161] Step S52: Establish the fuzzy evaluation matrix F;

[0162] Twenty experts were invited to evaluate the 15 secondary indicators based on the actual implementation of each stage of the project. The percentage of votes cast for each level of the secondary indicator was calculated to obtain the original membership degree. The results are shown in Table 12:

[0163] Table 12 Original Membership Degrees of Secondary Indicators

[0164] Step S53: Calculate the fuzzy entropy of each index;

[0165] Taking A1 as an example: 597.

[0166] Step S54: Calculate the fusion weight W 融合 And normalize and integrate weights in stages ;

[0167] Let the reliability weight ρ i =1−E i Then the reliability weight ρ of A1 i =1−0.9462 = 0.0403.

[0168] Use ρ i The combined weights are adjusted to obtain the fusion weight W. 融合 :

[0169]

[0170] The fusion weights are normalized in stages to obtain... The calculation results of the fuzzy entropy, reliability weight, fusion weight, and phased normalized fusion weight of each secondary indicator are shown in Table 13:

[0171] Table 13. Secondary Indicators: Fuzzy Entropy, Reliability Weight, Fusion Weight, and Normalized Fusion Weight

[0172] Step S6: Calculate the score for each level;

[0173] Step S61: Introduce a time decay factor to calculate the membership degree of the first-level index layer. Compared with the first-level indicator score S 一级 ;

[0174] The time interval between the end of the planning phase and the evaluation point The design phase will take 1.5 years. For one year, take =0.05.

[0175] A fuzzy evaluation is performed on primary indicator A, and its weight set is... and fuzzy evaluation matrix F A They are respectively:

[0176]

[0177] Calculate the membership degree of A :

[0178]

[0179]

[0180] Calculate A's score :

[0181] .

[0182] Similarly, calculate the membership vector and score for each stage: Based on the statistical results, perform fuzzy evaluation on the primary indicator B and calculate the membership degree of B:

[0183]

[0184]

[0185] Calculate B's score :

[0186] 20.

[0187] Based on the statistical results, a fuzzy evaluation is performed on the primary indicator C, and the membership degree of C is calculated:

[0188]

[0189] Calculate C's score :

[0190]

[0191] Based on the statistical results, a fuzzy evaluation is performed on the primary indicator D, and the membership degree of D is calculated:

[0192]

[0193] Calculate the score of D :

[0194] 50.

[0195] Step S62: Calculate the membership degree Z and score S of the target layer T. T ;

[0196] Its weight set and fuzzy evaluation matrix are W respectively 一级 and F T :

[0197]

[0198]

[0199] Calculate the membership degree of T :

[0200]

[0201] Calculate the score of T :

[0202] .

[0203] Step S63: Calculate the initial score S for each secondary indicator. 二级 ;

[0204] Step S631: Calculate the initial score S for each secondary indicator. 二级 The evaluation level is determined by combining the evaluation level score range table;

[0205] S 二级 =F

[0206] Such as S A1 =[0.15,0.20,0.30,0.25,0.10] [95,85,75,50,20] T =68.25. Similarly, calculate the initial scores for the remaining 14 secondary indicators.

[0207] Step S64: Introduce the ordered weighted average operator to calculate the aggregate score S OWA ;

[0208] Step S641: Initial score S for each secondary indicator 二级 Sort in descending order;

[0209] The sorted order and their evaluation levels are shown in Table 14:

[0210] Table 14 Initial Scores of Secondary Indicators in Descending Order and Evaluation Levels

[0211] Step S642: Calculate position weights :

[0212] ,like =2(16-14) / 240=0.0167,

[0213] Step S643: Calculate the combined weights of each secondary indicator. :

[0214]

[0215] The calculation results of the position weights and combined weights of each secondary indicator are shown in Table 15:

[0216] Table 15. Position weights and combined weights of each secondary indicator

[0217] Step S644: Calculate the aggregate score S of the secondary indicators. OWA :

[0218]

[0219] The aforementioned 71.01 is the comprehensive score after attenuation at the primary indicator level, used to determine the overall cost management level of the project. The calculated 75.70 is the OWA aggregate score at the secondary indicator level, derived from the initial scores of 15 secondary indicators and weighted accordingly, used to identify high-performing indicator dimensions. These two scores occupy different levels within the evaluation system and serve different purposes: the former is used for level determination, while the latter is used for optimal ranking.

[0220] Step S7: Attribution Diagnosis and Improvement Suggestion Generation

[0221] Step S71: Attribution diagnosis;

[0222] Collect historical score data of similar existing building renovation projects within the past 5 years, calculate the mean and standard deviation of the score for each secondary indicator, and set dynamic early warning thresholds. Let k=1.2. The initial scores S for each indicator in this project are... 二级 Corresponding Compare.

[0223] Taking A1 as an example: the historical adjusted mean of this indicator is 75.6, and the adjusted standard deviation is 6.1. Therefore: = 75.6 − 1.2 × 6.1 = 68.28, the score for A1 in this project is 68.25, slightly lower than the warning threshold, triggering a weak link warning. Similarly, the dynamic warning thresholds for the remaining 14 indicators are calculated and compared one by one. Finally, the following 4 weak links are identified: A1, A3, C1, and C4.

[0224] Step S72: Based on the pre-stored "low score indicator - improvement measures - responsible entity" mapping rule base, generate improvement suggestions as shown in Table 16:

[0225] Table 16 Low-scoring indicators, improvement measures, and responsible parties

[0226] According to the closed-loop feedback correction mechanism in step S211, after the comprehensive evaluation of this project is completed, the difference between the actual cost deviation rate of this project and the predicted score conversion deviation rate in step S6 can be used to correct the expert weights participating in the evaluation, so as to provide more accurate prior information on expert weights for the evaluation of similar projects in the future.

[0227] (1) The actual cost deviation rate was calculated based on the project settlement data. The total budgeted cost was RMB 14,348,122.58, the total settled cost was RMB 15,229,255.70, the overspending amount was RMB 881,133.12, and the overall overspending rate was:

[0228]

[0229] Since the budget and settlement for each stage of this project are not separated, the overall overspending rate is used here as an approximation of the actual cost deviation rate for each stage.

[0230] (2) Predicted score conversion deviation rate

[0231] To convert the score into a cost deviation rate, a conversion factor of k = 0.2 is set. The conversion formula is as follows:

[0232]

[0233] For example, in the planning stage: δ pred =(100−71.19) / 100×0.2 0.0576 = 5.76%

[0234] (3) Example of the revised expert weights

[0235] )

[0236] Taking expert 1 in the planning phase as an example, its original weight was 0.1192. With a learning rate η = 0.05 (satisfying 0 < η ≤ 0.2), the corrected base weight is:

[0237]

[0238] After correcting the weights of all experts in the planning phase in the same way, they are then normalized to obtain the updated expert weight vector for the planning phase. Similarly, correction factors for the design, construction, and operation and maintenance phases can be calculated separately, and the corresponding expert weights can be updated accordingly. The corrected weights will be stored in the system for use in subsequent project weight calculations, enabling dynamic learning and continuous optimization of expert weights.

[0239] Example 2:

[0240] This embodiment provides a full-process cost management and evaluation system for existing building renovation projects, used to automate the operation and result output of the method described in Embodiment 1 above. The system includes an indicator system storage module, an indicator scoring acquisition module, a weight calculation module, a fuzzy comprehensive evaluation module, an attribution diagnosis module, and an output module. These modules work together to form a complete online evaluation and diagnosis platform.

[0241] Indicator system storage module: Used to store the four-stage evaluation indicator system described in Example 1, including 4 primary indicators, 15 secondary indicators and corresponding comprehensive weight classification information, and supports dynamic updates and version management of the indicator system.

[0242] Indicator scoring and data collection module: Receives evaluation data of each secondary indicator of the target project from evaluators through the user interface, supports multiple users to score online at the same time, automatically performs data processing, deduplication and normalization, and generates the original membership matrix of each indicator.

[0243] The weight calculation module is used to implement the analytic hierarchy process (AHP) and closed-loop feedback correction for differentiated expert weighted integration. This module further includes the following sub-modules: an expert information management sub-module, used to store information such as experts' professional titles, years of experience, and familiarity levels at each stage; a judgment matrix construction sub-module, guiding users to perform pairwise comparisons using a nine-level scale; a consistency check sub-module, automatically calculating the CI and CR values ​​of the judgment matrix and issuing correction prompts for matrices with CR ≥ 0.1; a weighted integration calculation sub-module, calculating the expert weight vector based on the expert comprehensive score and multiplying it with the indicator weight matrix to obtain the integrated weight; and an expert weight feedback update sub-module, updating the weight according to the difference between the actual project cost deviation rate and the predicted score, using a correction formula. The expert weights are dynamically updated to achieve continuous optimization.

[0244] Fuzzy Comprehensive Evaluation Module: Used for implementing multi-level fuzzy comprehensive evaluation and score calculation. This module includes the following sub-modules: Membership Degree Statistics Sub-module, which automatically converts the collected scoring data into the membership degree of each indicator to the five-level comment set; Fuzzy Entropy Calculation Sub-module, which calculates the membership degree of each indicator to the five-level comment set; and Fuzzy Entropy Calculation Sub-module, which calculates the membership degree of each indicator to the five-level comment set. Calculate the fuzzy entropy and reliability weight ρ of each indicator. i =1 E i The fusion weight calculation submodule multiplies the comprehensive weight and the reliability weight to obtain the fusion weight. The process is then normalized in stages; the time decay submodule applies a decay factor exp(−γ) to the membership matrix of the planning and design stages. Δt); OWA synthesis submodule, using ordered weighted average operator. = The aggregated scores of the secondary indicators are calculated to make the evaluation results more sensitive to low-scoring indicators.

[0245] Attribution Diagnosis Module: Used to identify weaknesses and generate improvement suggestions. This module includes the following sub-modules: Dynamic Threshold Calculation Sub-module, which calculates the mean and standard deviation of scores for each indicator based on historical databases and sets dynamic warning thresholds. σ i Let k=1.2. The system compares the scores of each indicator in this project with the threshold one by one, automatically marking weak indicators that are below the threshold; the rule base storage submodule stores the mapping relationship between "low-scoring indicators - improvement measures - responsible entities" and the cost-benefit data of each measure; the suggestion generation and sorting submodule automatically matches the corresponding improvement measures based on the triggered weak indicators and generates an attribution diagnosis report.

[0246] This system can be deployed on a cloud server or a local server and accessed through a web browser or mobile app. It enables online evaluation and diagnosis of the cost management level of existing building renovation projects and supports cross-project expert weight learning and knowledge accumulation.

[0247] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A comprehensive cost evaluation method for the entire process of renovating existing buildings, characterized in that, Includes the following steps: Step S1: Construct an evaluation index system; An evaluation index system was established, consisting of four primary indicators (planning, design, construction, and operation and maintenance) and 15 corresponding secondary indicators. Step S2: Calculate the weights of the primary indicators using the differentiated expert weighted analytic hierarchy process, including: Step S21: A comprehensive score is calculated based on the expert's professional title, years of experience, and familiarity with the issues at the corresponding stage, and then normalized to obtain the expert weight w. ij Integrated expert weight vector W 专家 ; Step S22: Establish the primary indicator judgment matrix A; Step S23: Calculate the weights of the primary indicator judgment matrix A using the geometric mean method, and integrate the primary indicator weight matrix W. 指标 ; Step S24: Consistency check; Step S25: Place W 专家 With W 指标 Multiplying them together yields the integrated weight vector W of the primary indicators. 一级 ; Step S3: Calculate the weights of the secondary indicators for each stage; For each secondary indicator under a primary indicator, the same differentiated weighted integration method as in step 2 is used to calculate the expert weights of the secondary indicators and construct a judgment matrix, thereby calculating the weights W of the secondary indicators. 二级 . Step S4: Place W 一级 With W 二级 Multiplying them together yields the combined weight W for each secondary indicator. 综合 And importance is classified according to the comprehensive weight value; Step S5: Perform multi-level fuzzy comprehensive evaluation, including: Step S51: Construct a five-level comment set and corresponding score intervals, and take the midpoint of each score interval as the score vector V; Step S52: Establish the fuzzy evaluation matrix F, using the original membership degrees This represents the proportion of people who gave the jth rating for the i-th factor out of the total number of participants. Step S53: Calculate the fuzzy entropy of each index. ; Step S54: Let the reliability weight ρ i =1 E i Calculate the fusion weights And obtained by phased normalization . Step S6: Calculate the score for each level, including: Step S61: Introduce a time decay factor to the planning and design phases. The corrected membership matrix is ​​obtained. Calculate the score S of the first-level indicator layer. 一级 = No time decay is applied during the construction and operation and maintenance phases. The time interval from the end of this phase to the evaluation point, in years; The preset attenuation coefficient has a value range of 0 < ≤0.。; Step S61: Calculate the target layer score S T =W 一级 F T ; Step S61: Calculate the initial score S of the secondary indicator layer. 二级 =F V; Step S62: Calculate the aggregate score S of the secondary indicator layer using the ordered weighted average operator. OWA : ,in The scores are the initial scores of the secondary indicators sorted in descending order. For combined weights; Step S7: Attribution diagnosis and improvement suggestion generation, including: Step S71: Collect historical score data of similar existing building renovation projects, calculate the mean and standard of each secondary indicator, and set dynamic early warning thresholds. ,in Using preset coefficients, the initial scores of each secondary indicator in this project are compared with the corresponding thresholds to identify weak links that are below the thresholds; Step S72: Generate improvement suggestions based on the pre-stored "low score indicator - improvement measures - responsible entity" mapping rule base.

2. The method according to claim 1, characterized in that, Step S1 establishes an evaluation index system consisting of four primary indicators (planning, design, construction, and operation and maintenance) and 15 corresponding secondary indicators. The secondary indicators for the planning stage include: A1 Assessment of the current state and safety performance of the original building structure; A2 Sufficiency and depth of the renovation project planning scheme; A3 Detail of the on-site survey work; A4 Identification and responsiveness to policies, regulations, and standards; and A5 Clarity of project approval and funding arrangements. The secondary indicators for the design stage include: B1 Economic rationality of the renovation design scheme; and B2 Design responsiveness to the constraints of the existing building structure. The secondary indicators for the construction stage include: C1 Neighborhood relations management and disturbance control during construction; C2 Completeness of on-site visa and change management systems during construction; C3 Experience of the construction company in existing building renovation projects; C4 Reasonableness of risk reserve funds for the renovation project; C5 Effectiveness of the dynamic cost management mechanism for the project; and C6 Completeness of the original building structure protection and safe construction scheme. The secondary indicators for the operation and maintenance stage include: D1 Aging identification and maintenance / update management level of facilities and equipment; and D2 Energy system operating efficiency and energy consumption management level.

3. The method according to claim 1, characterized in that, The consistency check step in step S2 is as follows: CI is the consistency index; RI is the average random consistency index; λ max The largest eigenvalue of the judgment matrix is ​​CR. If CR < 0.1, then the judgment matrix passes the consistency test.

4. The method according to claim 1, characterized in that, Step S2 also includes closed-loop feedback correction S211 for expert weights: This step is not performed during the initial evaluation, but after the evaluation of one or more actual projects is completed, based on the actual cost deviation rate δ of each stage of the project. actual The deviation rate δ between the predicted scores of each stage calculated in step S6 and the actual predicted scores is calculated. pred The differences between them should be corrected by adjusting the original weights of each expert involved in the project at that stage using the following formula: η is the preset learning rate and 0 < η ≤ 0.2; the corrected weights are normalized and stored in the expert weight database for subsequent project evaluation.

5. The method according to claim 1, characterized in that, The importance classification in step S4 is as follows: a comprehensive weight value ≥ 0.1 is "very important"; 0.1 > comprehensive weight value ≥ 0.03 is "important"; and comprehensive weight value < 0.03 is "average".

6. The method according to claim 1, characterized in that, In step S5, the five-level evaluation set and score ranges are constructed as follows: "Sufficient implementation, significant control effect", "Good implementation, obvious control effect", "Basic implementation, average control effect", "Insufficient implementation, weak control effect", and "Missing implementation, ineffective control effect". The corresponding score ranges are [100, 90), [90, 80), [80, 60), [60, 40), and [40, 0). The midpoint of each score range is taken as the score vector V = .

7. The method according to claim 1, characterized in that, The position weights of the ordered weighted average operator described in step S64 are... = Calculate, where j is the initial score in descending order, and n is the number of indicators; the combined weights are calculated according to... calculate.

8. The method according to claim 1, characterized in that, In step S7, the historical data for setting the dynamic threshold based on historical data comes from the evaluation result database of at least 5 existing building renovation projects of the same type that this method has previously applied to. If there are fewer than 5 data, the benchmark data for cost management evaluation of existing building renovation projects published in the industry will be used instead.

9. The method according to claim 1, characterized in that, In step S7, improvement suggestions are generated based on a pre-stored mapping rule base. This rule base is constructed based on post-evaluation reports and expert interviews from no fewer than 10 existing building renovation projects, and adopts a multi-dimensional matching strategy: when a low-scoring indicator precisely matches the indicator name in the rule base, the corresponding measures are directly output. When no exact match is found, fuzzy matching is performed based on the stage and weight level of the indicator, and the rule with the highest similarity is selected.

10. A comprehensive cost evaluation system for the entire process of renovating existing buildings. The system includes: Indicator system storage module: used to store the four-stage evaluation indicator system described in Example 1, including 4 primary indicators, 15 secondary indicators and corresponding comprehensive weight classification information; Indicator scoring and data collection module: Receives the rating data of evaluators for each secondary indicator of the target project through the user interface, supports multiple users to score at the same time, and automatically performs data sorting and normalization. Weight Calculation Module: Used to implement the analytic hierarchy process (AHP) and closed-loop feedback correction for differentiated expert weighted integration; this module includes: expert information management submodule, judgment matrix construction submodule, consistency verification submodule, weighted integration calculation submodule, and expert weight feedback update submodule; Fuzzy comprehensive evaluation module: includes membership degree statistics submodule, fuzzy entropy calculation submodule, fusion weight calculation submodule, OWA synthesis submodule and time decay submodule; Attribution diagnosis module: includes dynamic threshold calculation submodule, rule base storage submodule, and suggestion generation submodule.

11. This system can be deployed on a cloud server or a local server and accessed through a web browser or mobile terminal APP to realize online evaluation and diagnosis of the cost management level of existing building renovation projects, and support cross-project expert weight learning and knowledge accumulation.

Images