A tram operation risk assessment and optimization method

Abstract

The present application relates to a kind of tramcar operation risk assessment and optimization method, the present application is combined by risk matrix method and expert scoring method, to carry out the risk assessment of tramcar operation, to realize the comprehensive identification and evaluation of tramcar operation risk on the basis of incomplete risk event data, finally according to the evaluation result, tramcar operation is optimized, and the operation safety level is improved.Compared with prior art, the present application effectively avoids the problem that the risk cannot be comprehensively evaluated due to insufficient tramcar operation risk event data, or the problem that the subjectivity of risk assessment is too strong due to the dependence on expert evaluation, on the basis of realizing objective, comprehensive and comprehensive evaluation of tramcar operation risk, according to the risk assessment result, tramcar operation is proposed targeted operation risk improvement suggestion, so as to realize the improvement of tramcar operation safety level.

Description

Technical Field

[0001] This invention relates to the field of risk assessment, and in particular to a method for assessing and optimizing the operational risks of trams. Background Technology

[0002] As a supplement to the urban public transportation system, trams have developed rapidly in recent years. By the end of 2021, the total operating length of tram lines in my country reached 503.33 km. Due to the non-enclosed operating environment, facilities, and personnel of trams, safety incidents frequently occur. Therefore, the analysis and assessment of tram operation risks are receiving increasing attention, aiming to identify weak points and propose targeted improvement measures to enhance operational safety.

[0003] Currently, the methods for identifying and assessing operational risks of trams are usually based on and draw on relevant standards and methods in the fields of railways and urban rail transit. For example, fault tree analysis is used to identify hazards, risk matrix analysis is used to rate tram risk events, network hierarchy analysis and evidence theory are used to establish risk assessment models, and risk indicators are determined through expert interviews and questionnaires.

[0004] However, fault tree analysis and risk matrix analysis require solid data support, and simply using the analytic hierarchy process (AHP) and expert scoring methods is easily influenced by the subjectivity of the scorers. Furthermore, existing research is often limited to case studies of certain routes, resulting in an insufficiently comprehensive identification of risk sources. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method for assessing and optimizing the operational risks of trams. Considering the problem that it is usually difficult to obtain comprehensive data on dangerous events during tram operation, this invention makes full use of existing risk event data, combines the risk matrix with expert scoring, and comprehensively assesses the operational risks of trams based on fuzzy comprehensive evaluation method, identifies the weak links in tram operation, and proposes targeted improvement measures to enhance its operational safety.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A method for assessing and optimizing the operational risks of trams includes the following steps:

[0008] Identify all risk factors and their corresponding risk sources related to the safety of tram operation, and formulate tram operation risk assessment objectives;

[0009] Based on the objectives, risk factors and corresponding risk sources of the tram operation risk assessment, a framework for the tram operation risk assessment index system is constructed using the analytic hierarchy process.

[0010] Determine the weight coefficients of each level of indicators in the risk assessment indicator system to obtain the fuzzy comprehensive evaluation weight set corresponding to different levels of indicators;

[0011] Risk levels are assessed for each risk source in the indicator layer. For risk sources supported by risk event data, risk levels are classified based on the risk matrix method, and the risk sources are assessed using the risk matrix. For risk sources without risk event data, risk levels are classified based on expert scoring, and the risk sources are assessed using the fuzzy comprehensive evaluation matrix.

[0012] Based on the fuzzy comprehensive evaluation method, the fuzzy comprehensive evaluation set of indicators at different levels is calculated according to the fuzzy comprehensive evaluation weight set, risk matrix and fuzzy evaluation matrix.

[0013] Based on the aforementioned fuzzy comprehensive evaluation set, the risk sources and tram operation risks are assessed, and the assessment results are obtained.

[0014] Based on the assessment results, the operation of the tram will be optimized to improve operational safety.

[0015] Furthermore, the risk assessment indicator system framework includes an indicator layer, a criterion layer, and an objective layer;

[0016] The target layer includes the overall target of tram operation risk assessment;

[0017] The risk factors included in the criteria layer are called risk categories;

[0018] The risk factors included in the indicator layer are called risk sources.

[0019] Further, determining the weighting coefficients of each level of indicators in the risk assessment indicator system includes the following steps:

[0020] Construct the judgment matrix;

[0021] Calculate the weighting coefficients and perform a consistency check;

[0022] If the consistency check requirement is met, all steps are terminated; otherwise, the judgment matrix is ​​adjusted, the weight coefficients are recalculated, and a one-time check is performed.

[0023] Furthermore, the judgment matrix is ​​constructed based on expert scoring and the nine-scale method, resulting in a judgment matrix P. k :

[0024]

[0025] Where k = 0, 1, 2, ..., 6; a ij >0, a ijThis represents the ratio of the influence of the current layer's indicator on the corresponding indicator in the previous layer; a ii =1, a ij =1 / a ji Let i,j = 1,2,...,n, where n is the judgment matrix P. k The order of.

[0026] Furthermore, the calculation of weighting coefficients and consistency checks include:

[0027] Calculate the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer and perform a consistency test;

[0028] Calculate the weight coefficients of each indicator in the criterion layer to the overall target in the target layer and perform a consistency check;

[0029] Calculate the weight coefficients of each indicator in the indicator layer to the overall target in the target layer, and perform an overall consistency test;

[0030] The calculation of the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer and the consistency check include the following steps:

[0031] Construct the judgment matrix P k Given k = 1, 2, 3, ..., 6, calculate P. k The geometric mean vector M of each row vector in the vector. i Normalizing it yields the weight coefficient w of the corresponding indicator. i M i with w i The expressions are as follows:

[0032]

[0033]

[0034] Based on the weighting coefficient w i The fuzzy comprehensive evaluation weight set W of the indicator layer for each indicator in the criterion layer is obtained. k ;

[0035] The consistency index (CI) and consistency ratio (CR) of each indicator in the indicator layer with respect to the corresponding indicator in the criterion layer are calculated, and their expressions are as follows:

[0036]

[0037]

[0038] In the formula, λ max (P k ) is the judgment matrix P k The maximum eigenvalue;

[0039] RI is the average random consistency index of each indicator in the indicator layer with respect to the corresponding indicator in the criterion layer.

[0040] When CR < 0.1, the judgment matrix P is considered to be... k If the result is satisfactory and consistent, otherwise the judgment matrix is ​​adjusted and recalculated.

[0041] The calculation of the weight coefficients of each indicator in the criterion layer to the overall objective in the target layer and the consistency check include the following steps:

[0042] Construct a judgment matrix P0, calculate the geometric mean vector M0 of each row vector in P0, normalize it to obtain the weight coefficient w0 of the corresponding index, and obtain the fuzzy comprehensive evaluation weight set W of the criterion layer indexes on the overall target of the target layer. 0 ;

[0043] The consistency index (CI) of each indicator in the criterion layer with the overall objective in the target layer is calculated. 0 and consistency ratio CR 0 Their expressions are as follows:

[0044]

[0045]

[0046] In the formula, λ max (P0) is the largest eigenvalue of the judgment matrix P0;

[0047] RI 0 This is the average random consistency index of each indicator in the criterion layer with respect to the total objective in the target layer.

[0048] When CR 0 If the result is less than 0.1, the judgment matrix P0 is considered to have satisfactory consistency; otherwise, the judgment matrix is ​​adjusted and recalculated.

[0049] Calculate the weight coefficients of each indicator in the indicator layer to the overall goal in the target layer, i.e., the overall hierarchical ranking, and perform an overall consistency check, including the following steps:

[0050] The fuzzy comprehensive evaluation weight set w' of each indicator in the indicator layer relative to the total target in the target layer is recursively derived, i.e., the overall hierarchical ranking is:

[0051] w'=W k ×W 0

[0052] The overall ranking consistency ratio CR' of the indicator layer is calculated as follows:

[0053]

[0054] When CR' < 0.1, the overall hierarchical ranking is considered to have satisfactory consistency; otherwise, the judgment matrix needs to be adjusted and recalculated.

[0055] Furthermore, risk levels are classified for each risk source based on the risk matrix method, expert evaluation method, and fuzzy comprehensive evaluation method.

[0056] Furthermore, the risk level classification of each risk source includes the following steps:

[0057] For risk sources supported by risk event data, the risk level of the risk sources is classified based on the risk matrix method, and the risk sources are evaluated using the risk matrix.

[0058] For risk sources without supporting data on risk events, risk levels are classified based on expert scoring, and fuzzy comprehensive evaluation is used to assess the risk sources.

[0059] Furthermore, for the risk sources supported by risk event data, risk levels are classified and a fuzzy evaluation matrix is ​​constructed, including the following steps:

[0060] Obtain actual risk event data;

[0061] Based on the actual risk event data, calculate the average frequency of occurrence of the risk events. and the average number of casualties caused by risk events Their expressions are as follows:

[0062]

[0063]

[0064] Where N t,s Indicates risk source C s The frequency of dangerous events in year t; I t,s Indicates risk source C s The average number of casualties caused by dangerous events in year t;

[0065] Based on the frequency of risk events in rail transit, six risk frequency levels are set, from high to low: F6, F5, F4, F3, F2, and F1.

[0066] Based on the average frequency of occurrence of the aforementioned risk events Determine the risk frequency level of the risk source corresponding to the risk event;

[0067] Using casualties and economic losses as indicators, the severity of the consequences of risk events is classified into six levels, with the severity ranging from high to low as C6, C5, C4, C3, C2, and C1.

[0068] Based on the average number of casualties caused by the aforementioned risk events Determine the risk consequence level of the risk source corresponding to the risk event;

[0069] Based on the aforementioned risk frequency level and risk consequence level, a risk matrix is ​​constructed with the risk frequency level as the horizontal axis and the risk consequence level as the vertical axis. Each element of the risk matrix corresponds to the risk level of the risk source under a certain probability and consequence.

[0070] Furthermore, for risk sources without supporting data on risk events, risk levels are classified and a fuzzy evaluation matrix is ​​constructed, including the following steps:

[0071] Construct a fuzzy evaluation matrix composed of m single-index evaluation sets based on expert scoring:

[0072] R = (R1, R2, ..., R s ,...,R m ) T

[0073] The single-indicator evaluation set is the risk level classification result of risk sources without risk event data support, denoted as R. s =(r s1 ,r s2 ,r s3 ,r s4 ) T , s∈[1,m];

[0074] Where, r sx Indicates risk source C s For the evaluation v x The membership degree of x∈[1,4], and r sx ∈[0,1], Rating v x This provides the risk level that experts can choose.

[0075] Furthermore, the step of obtaining fuzzy comprehensive evaluation sets of indicators at different levels based on the comprehensive fuzzy evaluation method, according to the fuzzy comprehensive evaluation weight set and the fuzzy evaluation matrix, includes the following steps:

[0076] The fuzzy evaluation matrix is ​​determined by the risk matrices of all risk sources in the indicator layer; among them, the evaluation matrix of risk sources with data support includes r. sx The fuzzy evaluation matrix for risk sources without data support is R, which takes the value of 0 or 1.

[0077] The fuzzy comprehensive evaluation weight sets of the criterion layer and the index layer are multiplied by the fuzzy evaluation matrices of all risk sources to obtain the fuzzy comprehensive evaluation sets of the two layers, respectively:

[0078] B = W × R

[0079] Based on the maximum membership degree method, the largest membership degree in B is selected as the evaluation result of the evaluation object.

[0080] Finally, the results of the comprehensive operational risk assessment of the tram system are analyzed, and optimization measures and suggestions are given from an overall perspective to optimize the operation of the tram system.

[0081] Compared with the prior art, the present invention has the following beneficial effects:

[0082] 1. This invention collects various risk factors affecting the safe operation of trams, making full use of actual data. For risk sources supported by risk event data, it assesses the risk sources based on the risk matrix method; for risk sources without risk event data support, it assesses them based on expert scoring and fuzzy comprehensive evaluation methods. This effectively avoids the problems of insufficient risk event data leading to an inability to comprehensively assess risks, or excessive subjectivity in risk assessment due to reliance solely on expert evaluation. Based on the objective, comprehensive, and integrated assessment of tram operation risks, it proposes targeted suggestions for improving tram operation risks based on the risk assessment results, thereby optimizing tram operation.

[0083] 2. This invention achieves a more comprehensive identification of risk sources by classifying each risk source in the indicator layer according to its risk level. Based on the classification of risk sources by risk level, targeted suggestions and improvement measures are proposed for risk sources with higher risks, which can better optimize the operation of trams and reduce the occurrence of risk events. Attached Figure Description

[0084] Figure 1 This invention relates to a method and process for assessing the operational risks of trams.

[0085] Figure 2 This invention relates to the classification of risk factors in tram operation;

[0086] Figure 3 This invention relates to a framework for a risk assessment index system for tram operation.

[0087] Figure 4 This invention relates to the application process of the analytic hierarchy process;

[0088] Figure 5 This invention relates to the risk matrix method evaluation process;

[0089] Figure 6 The results obtained in this embodiment of the invention are based on the expert evaluation method. Detailed Implementation

[0090] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0091] This invention constructs a risk assessment system based on a comprehensive identification of risk factors in tram operation. It makes full use of available hazardous event data, and combines expert scoring methods for any deficiencies with mature rail transit risk assessment theories to determine risk levels. This enables a comprehensive assessment of tram operation risks, thereby identifying weak links and proposing targeted improvement measures to enhance operational safety.

[0092] A method for assessing and optimizing tram operation risks, the implementation process of which is as follows: Figure 1 As shown, it includes the following steps:

[0093] Step 1: Identify and classify the risk factors in tram operation, and construct a framework for its operation risk assessment system based on the analytic hierarchy process (AHP).

[0094] Step 2: Determine the weight coefficients of each level of indicators in the tram operation risk assessment system, and obtain the fuzzy comprehensive evaluation weight set of each indicator in the indicator layer and criterion layer on the overall target of the target layer respectively;

[0095] Step 3: Based on whether there is supporting risk event data, assess the risk level of each risk source in the indicator layer using the risk matrix method and expert scoring method respectively;

[0096] Step 4: Use the fuzzy comprehensive evaluation method to calculate the fuzzy comprehensive evaluation set of indicators at different levels in the risk assessment system, so as to comprehensively assess the operational risks of the tram.

[0097] Step 5: Based on the assessment results of the tram operation risks, propose targeted suggestions and improvement measures for high-risk sources and the overall operation risks of the tram to reduce the occurrence of risk events.

[0098] 1. In step one, the risk factors affecting tram operation are identified and classified, such as... Figure 3 As shown, the specific process is as follows:

[0099] 1.1: Tram operation risk factors are divided into external factors and internal factors. External factors refer to risk factors not determined by the nature and status of the tram system itself, including risk factors caused by other traffic participants under semi-independent right-of-way and mixed right-of-way (hereinafter referred to as traffic environment factors) and natural environment factors. Internal factors refer to risk factors within the tram system, including equipment safety, facility safety, driver, and passenger personnel factors, etc.

[0100] 1.2: Among the external factors of the system, traffic environment factors can be further divided into three categories of risk factors based on other traffic participants: conflicts with motor vehicles, conflicts with non-motor vehicles, and conflicts with pedestrians. Natural environment factors include weather factors and geological factors. Among the internal factors of the system, equipment factors include vehicle system failures and roadside system failures. Facility factors include unreasonable design and damage to track and station facilities. Driver error and unsafe passenger behavior are also risk factors affecting tram operation.

[0101] 1.3: "Tram operation risk" is used as the overall assessment objective and is classified as the primary risk category. Traffic environment, natural environment, equipment, facilities, drivers, and passengers are classified as secondary risk categories. Motor vehicle conflicts, non-motor vehicle conflicts, pedestrian conflicts, weather, geology, vehicle system malfunctions, roadside system malfunctions, unreasonable track and station design and damage, driver misconduct, and unsafe passenger behavior are classified as tertiary risk categories.

[0102] 1.4: Based on the characteristics of different operating lines and existing operating experience, all risk sources of the tram operation system are included in the above three-level risk classification.

[0103] Step one establishes a framework for a tram operation risk assessment system based on the analytic hierarchy process, such as... Figure 2 As shown, the specific process is as follows:

[0104] 1.5: Determine the indicator levels of the evaluation system based on the hierarchical structure of the Analytic Hierarchy Process (AHP), and divide the evaluation system into the target layer A, the criterion layer B, and the indicator layer C.

[0105] 1.6: The evaluation system with three levels of indicators obtained in 1.4 corresponds one-to-one with the three tiers in 1.5. That is, "tram operation risk" is the total target of target layer A; the six risk categories of traffic environment B1, natural environment B2, equipment B3, facilities B4, driver B5, and passenger B6 are the indicators of criterion layer B; the three-level classification indicators obtained by combining experience and specific circumstances are all risk sources of tram operation, and are the indicators Cs in indicator layer C, s∈[1,22];

[0106] The risk sources under the traffic environment B1 include motorcycle risk C1, other public transportation risk C2, car risk C3, train risk C4, non-motorized vehicle risk C5, and pedestrian risk C6; the risk sources under the natural environment B2 include severe weather risk C7 and earthquake risk C8; the risk sources under equipment B3 include turnout system risk C9, signaling system risk C10, power supply system risk C11, traction equipment risk C12, power equipment risk C13, and mechanical equipment risk C14; the risk sources under facilities B4 include track risk C15 and station risk C16; the risk sources under drivers B5 include signal violation risk C17 and train control mismanagement risk C18; and the risk sources under passengers B6 include risk of falling inside the vehicle C19, risk of falling off the train C20, risk of being trapped inside the vehicle C21, and risk of being stuck in the door C22.

[0107] 2. In step two, determine the weight coefficients of each level of indicators in the tram operation risk assessment system, such as... Figure 4 As shown, it includes the following steps:

[0108] (1) Establish a hierarchical structure;

[0109] (2) Construct the judgment matrix;

[0110] (3) Calculate the weighting coefficients;

[0111] (4) Consistency check;

[0112] (5) If the consistency test requirement is met, then all steps are terminated; otherwise, the judgment matrix is ​​adjusted and the process returns to step (3).

[0113] The specific process is as follows:

[0114] 2.1: The importance between any two indicators within the indicator layer is evaluated using expert scoring and the nine-scale method shown in Table 1, constructing seven judgment matrices P. k Let P be an integer (k = 0, 1, 2, ..., 6). k It is A-[B1-B6], B1-[Ca-Cb], B2-[Cc-Cd], B3-[Ce-Cf], B4-[Cg-Ch], B5-[Cl-Co], B6-[Cp-Cq];

[0115] Table 1 Nine-Scale Method

[0116]

[0117] The expression for the judgment matrix is:

[0118]

[0119] Where k = 0, 1, 2, ..., 6; a ij >0, a ij Indicates the index x of this layer i With indicator x j The ratio of the impact on the corresponding indicator at the previous level; a ii =1, a ij =1 / a ji , i,j=1,2,...,n.

[0120] 2.2: Calculate the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer, and the weight coefficients of the six indicators in the criterion layer to the total target in the target layer, and perform consistency checks.

[0121] The calculation of the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer and the consistency test specifically include the following steps:

[0122] 2.2.1: Calculate the six judgment matrices P from 2.1 k k = 1, 2, ..., 6, i.e., B1-[Ca-Cb], B2-[Cc-Cd], B3-[Ce-Cf], B4-[Cg-Ch], B5-[Cl-Co], B6-[Cp-Cq], and the geometric mean vector M of the row vectors of each judgment matrix. i Normalization yields the weight coefficient w of the corresponding indicator. i As shown in equation (1), the weight set W of the indicator layer on each indicator of the criterion layer is obtained. k ;

[0123]

[0124] 2.2.2: Calculate the consistency index CI and consistency ratio CR of the indicator layer to the criterion layer, as shown in equation (2);

[0125]

[0126] In the formula, λ max (P k ) is the judgment matrix P k The largest eigenvalue, n is the judgment matrix P k The order of;

[0127] The average random consistency index RI can be obtained from Table 2.

[0128] Table 2 Average Random Consistency Index

[0129]

[0130] Calculate the consistency ratio CR as shown in equation (3);

[0131]

[0132] When CR < 0.1, the judgment matrix is ​​considered to have satisfactory consistency; otherwise, the judgment matrix needs to be adjusted and recalculated.

[0133] The calculation of the weight coefficients of the six indicators at the criterion layer to the overall objective at the target layer and the consistency test include the following steps:

[0134] 2.2.3: Calculate the weighting coefficients of the six indicators at the criterion layer to the overall objective at the target layer:

[0135] Calculate the geometric mean vector M0 of the row vectors of the judgment matrix P0 in 2.1, i.e., A-[B1-B6], and normalize it to obtain the weights w0 of the corresponding indicators, as shown in Equation (1). This yields the weight vector W of the six indicators of the criterion layer to the total target of the target layer. 0 ;

[0136] 2.2.4: Calculate the consistency index (CI) of the six indicators at the criterion level with the overall objective at the target level. 0 and consistency ratio CR 0 The specific process is the same as in 2.2.2;

[0137] 2.3: Calculate the weight coefficients of each indicator in the indicator layer to the overall goal in the target layer, i.e., the overall hierarchical ranking, and perform an overall consistency check, specifically including the following steps:

[0138] 2.3.1: Recursively derive the fuzzy comprehensive evaluation weight set w' of each indicator in the indicator layer relative to the total target of the target layer, i.e., the overall hierarchical ranking, as shown in equation (4);

[0139] w'=W k ×W 0 (4)

[0140] 2.3.2: Calculate the overall ranking consistency ratio CR' of the indicator layer, as shown in equation (5);

[0141]

[0142] In the formula, CI 0 The vector is composed of the consistency indices corresponding to the six indicators in the criteria layer. Each component can be calculated using equation (2); RI 0 The vector is composed of the average random consistency index corresponding to the 6 indicators in the criterion layer, which can be obtained from Table 2;

[0143] When CR' < 0.1, the overall hierarchical ranking is considered to have satisfactory consistency; otherwise, the judgment matrix needs to be adjusted and recalculated.

[0144] 3. In step three, the risk matrix method is used to classify the risk levels of risk sources supported by hazardous event data in the indicator layer, such as... Figure 5 As shown, it includes the following steps:

[0145] (1) Obtain risk event data;

[0146] (2) Obtain the frequency of risk events and the number of casualties caused by risk events respectively;

[0147] (3) Based on the frequency of occurrence of risk events, classify the risk frequency level of the corresponding risk source; based on the number of casualties caused by risk events, classify the risk consequence level of the corresponding risk source.

[0148] (4) Construct a risk matrix;

[0149] (5) The risk levels are determined.

[0150] The specific process is as follows:

[0151] 3.1: Determine the risk frequency level of the risk source. The specific process is as follows:

[0152] 3.1.1: Calculate the frequency of hazardous events based on actual data. As in equation (6);

[0153]

[0154] In the formula, Risk source C at the indicator layer s The corresponding average frequency of occurrence of hazardous events, N t,s Indicates risk source C s The frequency of dangerous events in year t;

[0155] 3.1.2: The frequency level of hazardous events is set to 6 levels based on the frequency of occurrence of hazardous events, as shown in Table 3;

[0156] Table 3 Risk Frequency Level Classification

[0157]

[0158] 3.1.3: Obtain the risk frequency level of the risk source corresponding to a certain hazardous event based on Table 3;

[0159] 3.2: Determine the risk consequence level of the risk source. The specific process is as follows:

[0160] 3.2.1: Calculate the average number of serious injuries or deaths caused by hazardous events based on actual data. As in equation (7);

[0161]

[0162] In the formula, Risk source C at the indicator layer s The average number of serious injuries or deaths per dangerous event, I t,s Indicates risk source C s The number of serious injuries or deaths caused by dangerous events in year t;

[0163] 3.2.2: Using casualties and economic losses as indicators, the severity of the consequences is divided into 6 levels, as shown in Table 4:

[0164] Table 4 Classification of Risk Consequence Levels

[0165]

[0166] 3.2.3: Obtain the risk consequence level of the risk source corresponding to a certain hazardous event based on Table 4;

[0167] 3.3: Construct a risk matrix with frequency level and consequence level as the horizontal and vertical axes. Each element of the matrix corresponds to the risk assessment value under a certain frequency and consequence, and is divided into four levels: A. Unacceptable, B+ Undesirable, B. Tolerable, and C. Acceptable, as shown in Table 5.

[0168] Table 5 Risk Matrix

[0169]

[0170]

[0171] The definitions of the four risk levels, A, B+, B, and C, are shown in Table 6.

[0172] Table 6 Risk Level Definitions

[0173]

[0174] 3.4: Determine the frequency and severity of risk events based on actual hazardous event data, classify the risk levels of the corresponding risk sources in the indicator layer, and obtain the risk levels of the data-supported risk sources under the six indicators in the criterion layer, k∈{1,2,...,6};

[0175] In step three, the risk level of risk sources without supporting data on dangerous events in the indicator layer is classified using an expert scoring method. The specific process is as follows:

[0176] 3.5: Constructing a fuzzy evaluation matrix composed of m single-index evaluation sets based on expert scoring method

[0177] R = (R1, R2, ..., R s ,...,R m ) T

[0178] Among them, the single-index evaluation set R s =(r s1 ,r s2 ,r s3 ,r s4 ) T This refers to the risk level classification results for risk sources that lack data support; r sx ∈[0,1], indicating that the index C s For the evaluation v x The degree of membership, and Rating v x These are evaluation options that experts can choose, consistent with the risk level definitions in Table 5.

[0179] 4. In step four, the fuzzy comprehensive evaluation method is used to calculate the fuzzy comprehensive evaluation sets at different levels in the evaluation system. The specific process is as follows:

[0180] 4.1: The risk assessment results of the six indicators at the criteria level are calculated. The specific process is as follows:

[0181] 4.1.1: Determine the weight set W Z , which is the weight set W of the indicator layer for the six indicators of the criterion layer in step 2.2.1. k ;

[0182] 4.1.2: Determine the fuzzy evaluation matrix R Z It consists of the risk matrix of all risk sources in the indicator layer. Among them, the risk matrix of risk sources with data support includes r... sx The fuzzy evaluation matrix for risk sources with values ​​of 0 or 1 and no data support is the fuzzy evaluation matrix R in section 3.5.

[0183] 4.1.3: Based on weight set W Z and fuzzy evaluation matrix R Z The fuzzy comprehensive evaluation set B of the six indicators of the criterion layer is calculated according to equation (8). Z This yields the risk assessment results for each indicator in the criteria layer.

[0184] B Z =W Z ×R Z (8)

[0185] 4.2: The risk assessment results for the overall target at the target level are calculated, and the specific process is as follows:

[0186] 4.2.1: Determine the weight set W M , is the weight set w' of the indicator layer on the total target of the target layer in step 2.2.1;

[0187] 4.2.2: Determine the fuzzy evaluation matrix RM The fuzzy evaluation matrix R in step 4.1.2 Z Consistent;

[0188] 4.2.3: Calculate the fuzzy comprehensive evaluation set B of the total objective at the target layer. M The risk assessment results of tram operation risk are obtained, as shown in equation (9);

[0189] B M =W M ×R M (9)

[0190] Further risk assessments are conducted for different risk sources, and the specific process is as follows:

[0191] 4.3: Selecting B using the maximum membership method Z The highest membership degree is used as the evaluation result of each risk source in the criterion layer, as shown in equation (10);

[0192] V = {v} x |v x →max(b x )} (10)

[0193] 4.4: Regarding B Z The risk source with the highest membership degree is rated A or B+. The causes of dangerous incidents in tram operation and corresponding solutions are analyzed.

[0194] Step four involves a comprehensive assessment of the tram's operational risks. The specific process is as follows:

[0195] 4.5: Selecting B using the maximum membership method M The highest degree of membership is used as the evaluation result of the comprehensive risk of tram operation, as shown in equation (10);

[0196] 4.6: Analyze the results of the comprehensive operational risk assessment of tram systems and provide measures and suggestions from an overall perspective.

[0197] The specific implementation includes the following steps:

[0198] I. Determining the risk level of each indicator based on the assessment system framework

[0199] This embodiment statistically analyzes operational accident information for Changchun Tram Routes 54 and 55 from 2015 to 2019, obtaining actual data on accidents caused by various risk indicators under traffic environment B1, natural environment B2, and passenger risk B6. The data includes the annual number of accidents for each indicator under B1, B2, and B6, as well as the corresponding downtime or number of casualties for each accident. Based on the obtained actual risk event data, a risk matrix method is used to assess the risk level of each risk source indicator.

[0200] The risk level assessment of the above-mentioned risk indicators supported by actual data, using the risk matrix method, is shown in Table 7.

[0201] Table 7 Risk Level Assessment Based on Risk Matrix Method

[0202]

[0203]

[0204] For risk indicators that lack actual data support, namely the risk level assessment of each indicator under equipment B3, facility B4, and driver B5, the expert evaluation method is used to assess the risk level of each risk source indicator.

[0205] Evaluation results obtained based on expert evaluation method, such as Figure 6 As shown in the diagram, the size of the color block for each risk indicator represents the degree of membership of that indicator to the corresponding risk level. For example, the degree of membership of a turnout fault to level A is 0.1, to level B+ is 0.5, and to level B is 0.4. According to the maximum membership method, the risk level corresponding to this risk indicator can be determined to be level B+.

[0206] II. Risk Assessment of Criteria-Level and Target-Level Indicators

[0207] (1) For criterion layer B, based on the above formulas (1)-(8), the wind level of the indicators under criterion layers B1, B2, and B6, i.e., the fuzzy evaluation matrix R, is given in Table 2. The risk level of the indicators under criterion layers B3, B4, and B5, i.e., the fuzzy evaluation matrix R, is calculated according to... Figure 4 The process is given by the expert evaluation method, and the fuzzy comprehensive evaluation set of the criterion layer B is calculated as shown in Table 8.

[0208] Table 8 Risk Assessment at the Criterion Level

[0209]

[0210] The results show that the risk levels of traffic environment B1, natural environment B2, and equipment B3 in the criterion layer are relatively high, classified as B+, indicating undesirable risks; the rest are classified as B, indicating acceptable risks. In the indicator layer, the risk levels of non-motorized vehicles / pedestrians (C1), cars (C2) encroaching on tracks, collisions with non-motorized vehicles / pedestrians (C3), other trams (C4), buses / freight vehicles (C5), cars (C6), severe weather (C7), equipment switches (C9), and signal system malfunctions (C10) are all relatively high.

[0211] (2) Risk assessment of target-level indicators

[0212] For target layer A, the overall hierarchy and risk level matrix are obtained based on the above formulas (1)-(9);

[0213] The membership degree of the target layer A tram operation risk index was calculated as follows:

[0214] B = [0.1159, 0.4068, 0.4404, 0.0359], where the membership degree of level B is the largest at 0.4404. According to the maximum membership degree method, the overall risk level of tram operation in Changchun City can be considered to be level B, indicating that the current operational risk of trams is generally acceptable after adopting sufficient control and management measures.

[0215] III. Suggestions for optimizing tram operation

[0216] (1) Considering that both tram lines in Changchun are low-capacity rail transit lines without independent right-of-way or block system, in order to avoid potential hazards caused by the high risk indicators under traffic environment B1, it is recommended to adopt mixed right-of-way and set up guardrails and dedicated passenger platforms, increase warning signs in accident-prone areas, and remind passengers more often when getting on and off the tram; carry out publicity and education for the general public, reminding them to be more vigilant when crossing tram operating sections and not to compete with trams; for the phenomenon of private cars occupying the road, relevant departments can use electronic police, tram on-board monitoring and other methods to take pictures and fine violating vehicles; strengthen the training of staff, reasonably schedule and operate trains, and avoid train collisions.

[0217] (2) Northeast cities may experience severe weather such as rainstorms in summer and heavy snow in winter. Therefore, in order to cope with the impact of severe weather under the natural environment B2, the drainage and anti-skid performance of the tracks and roads should be improved. It is also necessary to predict extreme weather in a timely manner and prepare snow removal measures in advance to ensure the normal operation of trams.

[0218] (3) For the tram system related equipment B3, strengthen the maintenance and health management of key signal and power supply equipment in daily operation, and restrict the height of large freight cars passing through the tram operating section; for key equipment such as tram on-board and trackside, adopt operation and maintenance strategies such as condition maintenance, fault prediction and health management technology to avoid insufficient or excessive maintenance as much as possible and actively reduce the possibility of equipment failure.

[0219] This embodiment identifies the main risks in tram operation in northern Chinese cities and establishes a framework for a tram operation risk assessment index system. Given the limitations of actual event data, a risk matrix method combined with expert scoring is used to assess the risk level at the index level, and a fuzzy comprehensive evaluation method is employed to evaluate operational risks. Based on operational event data of Changchun trams from 2015 to 2019, the assessment results show that the overall operational risk of Changchun trams is generally controllable. However, the risk levels of traffic environment, natural environment, and equipment at the criterion level are relatively high, belonging to the undesirable risk level. Based on the assessment results, optimization suggestions are proposed: for systems located in northern cities with similar tram operation scenarios to Changchun, necessary risk prevention physical measures should be implemented during construction, such as installing guardrails and warning signs. Regulations for the travel behavior of other traffic participants should be refined, passengers should be reminded more frequently when boarding and alighting, and penalties should be imposed for obstructing the tracks when necessary. The severe cold weather in northern China also has a significant impact on tram operation. Timely monitoring and handling of snow accumulation and freezing / thawing issues on the tracks should be strengthened, and road surface drainage facilities in tram operating sections should be improved to reduce the impact of summer water accumulation on passenger travel.

[0220] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for assessing and optimizing the operational risks of trams, characterized in that, Includes the following steps: Identify risk factors related to the safety of tram operations; Based on the aforementioned tram operation risk factors, a framework for a tram operation risk assessment index system is constructed using the analytic hierarchy process (AHP). Determine the weight coefficients of each level of indicators in the risk assessment indicator system to obtain the fuzzy comprehensive evaluation weight set corresponding to different levels of indicators; Risk levels are assessed for each risk source in the indicator layer. For risk sources supported by risk event data, risk levels are classified based on the risk matrix method, and the risk sources are assessed using the risk matrix. For risk sources without risk event data, risk levels are classified based on expert scoring, and the risk sources are assessed using the fuzzy comprehensive evaluation matrix. Based on the fuzzy comprehensive evaluation method, the fuzzy comprehensive evaluation set of indicators at different levels is calculated according to the fuzzy comprehensive evaluation weight set, risk matrix and fuzzy evaluation matrix. Based on the aforementioned fuzzy comprehensive evaluation set, the risk sources and tram operation risks are assessed, and the assessment results are obtained. Based on the evaluation results, the tram operation was optimized.

2. The method for assessing and optimizing tram operation risks according to claim 1, characterized in that, The risk assessment indicator system framework includes an indicator layer, a criterion layer, and an objective layer; The target layer includes the overall target of tram operation risk assessment; The risk factors included in the criteria layer are called risk categories; The risk factors included in the indicator layer are called risk sources.

3. The method for assessing and optimizing tram operation risks according to claim 2, characterized in that, Determining the weighting coefficients of each level of indicators in the risk assessment indicator system includes the following steps: Construct the judgment matrix; Calculate the weighting coefficients and perform a consistency check; If the consistency check requirement is met, all steps are terminated; otherwise, the judgment matrix is ​​adjusted, the weight coefficients are recalculated, and a one-time check is performed.

4. The method for assessing and optimizing tram operation risks according to claim 3, characterized in that, The judgment matrix is ​​constructed based on expert scoring and the nine-scale method. : in, Determine if all elements in the matrix are greater than 0. a 11 This indicates the ratio of the impact of the first indicator on itself. a 1n This represents the ratio of the influence of the first indicator on the nth indicator. a n1 This represents the ratio of the influence of the nth indicator to the first indicator. a nn This represents the ratio of the influence of the nth-level indicator on itself. , To determine the matrix The order of.

5. The method for assessing and optimizing tram operation risks according to claim 4, characterized in that, The calculation of weighting coefficients and consistency checks include: Calculate the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer and perform a consistency test; Calculate the weight coefficients of each indicator in the criterion layer to the overall target in the target layer and perform a consistency check; Calculate the weight coefficients of each indicator in the indicator layer to the overall target in the target layer, and perform an overall consistency test.

6. The method for assessing and optimizing tram operation risks according to claim 5, characterized in that, The process of calculating the weight coefficients of each indicator in the indicator layer to the corresponding indicator in the criterion layer and performing a consistency check includes the following steps: Construct the judgment matrix , ,calculate Geometric mean vector of each row vector in the vector Normalizing them yields the weight coefficients of the corresponding indicators. , and The expressions are as follows: in: , Indicates the first i The first indicator for the first j The ratio of the influence of each indicator ; Based on the weighting coefficients The fuzzy comprehensive evaluation weight set of each indicator in the criterion layer is obtained from the indicator layer. ; The consistency index of each indicator in the calculation indicator layer with the corresponding indicator in the criterion layer. and consistency ratio Their expressions are as follows: In the formula, To determine the matrix The maximum eigenvalue; This is the average random consistency index of each indicator in the indicator layer with respect to the corresponding indicators in the criterion layer. when When, the judgment matrix is ​​considered to be... If the result is satisfactory and consistent, otherwise the judgment matrix is ​​adjusted and recalculated. The calculation of the weight coefficients of each indicator in the criterion layer to the overall target in the target layer and the consistency verification include the following steps: Construct the judgment matrix ,calculate The geometric mean vector of each row vector in the vector. Normalization yields the weight coefficients of the corresponding indicators. The fuzzy comprehensive evaluation weight set of the criteria layer indicators on the overall objective of the target layer is obtained. ; The consistency index of each indicator in the calculation criteria layer with the overall goal of the target layer. and consistency ratio Their expressions are as follows: In the formula, To determine the matrix The maximum eigenvalue; This is the average random consistency index of each indicator in the criterion layer with respect to the total objective in the target layer. when When, the judgment matrix is ​​considered to be... If the result is satisfactory and consistent, then the judgment matrix is ​​adjusted and recalculated.

7. The method for assessing and optimizing tram operation risks according to claim 6, characterized in that, The calculation of the weight coefficients of each indicator in the indicator layer to the overall target of the target layer, i.e., the overall hierarchical ranking, and the overall consistency check, includes the following steps: The fuzzy comprehensive evaluation weight set of each indicator in the indicator layer relative to the total target in the target layer is recursively derived and calculated. The overall hierarchical ranking is as follows: Calculate the overall ranking consistency ratio of the indicator layer. for: when If the overall hierarchical ranking is satisfactory, then the overall ranking is considered to have satisfactory consistency; otherwise, the judgment matrix needs to be adjusted and recalculated.

8. The method for assessing and optimizing tram operation risks according to claim 7, characterized in that, For the risk sources supported by risk event data, risk levels are classified and a fuzzy evaluation matrix is ​​constructed, including the following steps: Obtain actual risk event data; Based on the actual risk event data, calculate the average frequency of occurrence of the risk events. and the average number of casualties caused by risk events Their expressions are as follows: in Indicates risk source exist The frequency of dangerous events caused by annual occurrence; Indicates risk source exist The average number of casualties caused by dangerous events in a given year; Based on the frequency of risk events in rail transit, six risk frequency levels are set, from high to low: F6, F5, F4, F3, F2, and F1. Based on the average frequency of occurrence of the aforementioned risk events Determine the risk frequency level of the risk source corresponding to the risk event; Using casualties and economic losses as indicators, the severity of the consequences of risk events is classified into six levels, with the severity ranging from high to low as C6, C5, C4, C3, C2, and C1. Based on the average number of casualties caused by the aforementioned risk events Determine the risk consequence level of the risk source corresponding to the risk event; Based on the aforementioned risk frequency level and risk consequence level, a risk matrix is ​​constructed with the risk frequency level as the horizontal axis and the risk consequence level as the vertical axis. Each element of the risk matrix corresponds to the risk level of the risk source under a certain probability and consequence.

9. The method for assessing and optimizing tram operation risks according to claim 8, characterized in that, For risk sources without supporting data on risk events, risk levels are classified and a fuzzy evaluation matrix is ​​constructed, including the following steps: Based on expert scoring method A fuzzy evaluation matrix composed of individual indicator evaluation sets: The single-indicator evaluation set is the risk level classification result of risk sources without risk event data support, denoted as... , ; in, Indicates risk source For evaluation membership degree ,and , ;evaluate This provides a risk level that experts can choose.

10. A method for assessing and optimizing tram operation risks according to claim 9, characterized in that, The method based on comprehensive fuzzy evaluation, which obtains fuzzy comprehensive evaluation sets of indicators at different levels according to the fuzzy comprehensive evaluation weight set and fuzzy evaluation matrix, includes the following steps: The fuzzy evaluation matrix is ​​determined by comprising the risk matrices of all risk sources in the indicator layer; among them, the evaluation matrices of risk sources with data support are... The fuzzy evaluation matrix for risk sources with values ​​of 0 or 1 and no data support is: ; The fuzzy comprehensive evaluation weight sets of the criterion layer and the index layer are multiplied by the fuzzy evaluation matrices of all risk sources to obtain the fuzzy comprehensive evaluation sets of the two layers, respectively: Based on the maximum membership degree method, the largest membership degree in B is selected as the evaluation result of the evaluation object.

Images