Cross-domain maintenance decision-making method for coordinated optimization of maintenance service outlets, activities, and resources

Abstract

The present invention relates to the technical field of avionics equipment, and disclosed is a cross-domain maintenance decision-making method for coordinated optimization of maintenance service outlets, activities, and resources. The method comprises the steps of: on the basis of a fault phenomenon, acquiring a set of a variety of maintenance activities; determining whether resources required for a faulty product are missing at a current service outlet, and if the resources are missing, separately computing a resource scheduling time and cost required for each maintenance activity for which resources are missing; computing a fault repair time, repair costs, and repair probability of a certain maintenance activity; and determining a goal of a maintenance decision, and outputting an optimal maintenance scheme. The present invention can assist technicians in making maintenance decisions during maintenance of electronic products, reduce the waste of time and resources caused by the inconsistencies and uncertainties of maintenance activities, guide maintenance activities of remote operation and maintenance, reduce maintenance time, and reduce operation and maintenance costs.

Description

Cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization TECHNICAL FIELD

[0001] The present application relates to the technical field of avionics, more specifically, to a cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization. BACKGROUND

[0002] In modern application scenarios with varying forms, the reliability and operation and maintenance service of equipment are the key to victory. When the key equipment fails or has problems in the field, emergency repair and maintenance work is usually needed to reduce downtime or loss. The traditional maintenance process first reports the fault by the field staff, then the maintenance personnel of the design factory go to the scene, judge the possible fault type of the equipment according to the actual situation of the equipment on site, and carry out maintenance activities. The maintenance decision is usually based on the subjective judgment and experience of the on-site maintenance personnel, which may cause inconsistency and uncertainty of the decision, thereby causing waste of time and resources, including waiting for maintenance personnel to arrive and unnecessary or missing spare parts, instruments and tool preparation, causing unnecessary downtime extension, and seriously affecting necessary work and service. However, due to the problems of different environments and highly integrated functions of aviation equipment, the following difficulties exist in the formulation of maintenance decision:

[0003] (1) High complexity of products and faults, including many product varieties, various fault types and random fault occurrence;

[0004] (2) Uncertainty of resource distribution at the time of failure;

[0005] (3) Differences in user requirements, such as time, cost, etc. SUMMARY

[0006] The purpose of the present application is to overcome the shortcomings of the prior art and provide a cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization, which can guide remote operation and maintenance activities, reduce maintenance time and reduce operation and maintenance cost.

[0007] The purpose of the present application is achieved by the following scheme:

[0008] A cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization, comprising the following steps:

[0009] According to the fault phenomenon, a set of various maintenance activities is obtained;

[0010] Determine whether the required resources of the fault product are missing at the current network point, if there is a resource missing condition, calculate the resource scheduling time and cost required for each maintenance activity lacking resources;

[0011] calculating the fault repair time, repair cost and repair probability of a certain maintenance activity;

[0012] determining the target of the maintenance decision and outputting the optimal maintenance scheme.

[0013] Further, the set of multiple maintenance activities includes being obtained from an operation and maintenance platform.

[0014] Further, the resource scheduling time and cost required by each resource-lacking maintenance activity are calculated respectively, specifically including: using a mixed integer programming model to calculate the resource scheduling time and cost required by each resource-lacking maintenance activity respectively.

[0015] Further, the target of the maintenance decision is determined and the optimal maintenance scheme is outputted, specifically including: comprehensively considering the cost and time factors, determining the target of the maintenance decision, setting the objective function, using an optimization algorithm to traverse all possible maintenance activity sequences, and outputting the optimal maintenance scheme.

[0016] Further, the set of multiple maintenance activities is obtained according to the fault phenomenon, specifically including the sub-steps: according to the fault phenomenon, using retrieval or knowledge reasoning to obtain the fault reason and its bound troubleshooting activities and the required spare parts or instrument tooling.

[0017] Further, the required resources of the fault product at the current network site are determined, specifically including the sub-steps: if there is no resource shortage, the fault repair time, repair cost and repair probability of a certain maintenance activity are calculated.

[0018] Further, the fault repair time, repair cost and repair probability of a certain maintenance activity are calculated, specifically including the sub-steps:

[0019] Taking the maintenance decision model or the resource demand type and quantity selected by the user as input, based on the resource distribution and use state in the support network, according to the selected resource scheduling strategy, a resource scheduling scheme is provided, specifically including:

[0020] First, a mathematical model of the problem is established, and a resource scheduling objective function is constructed:

[0021] Where w1 and w2 are the weight proportions of cost and time, respectively. If w1=1 and w2=0, the resource scheduling cost is minimized. If w1=0 and w2=1, the scheduling time is minimized. min represents the minimum value function, i represents the network site, k represents the network site, t represents the transportation mode, I represents the number of network sites, and z ikt represents whether to call resources from i network site to k network site by transportation mode t; cost tikCijk (t) represents the transportation cost from i to k using transportation mode t, time tik Cijk (t) represents the transportation cost from i to k using transportation mode t, time

[0022] When i = R:

[0023] Where T represents the type of transportation mode, x kijt Cijk (t) represents the transportation cost from i to k using transportation mode t, time j Cijk (t) represents the transportation cost from i to k using transportation mode t, time

[0024] Cijk (t) represents the transportation cost from i to k using transportation mode t, time

[0025] When i ≠ R:

[0026] Where T represents the type of transportation mode, x ikjt Cijk (t) represents the transportation cost from i to k using transportation mode t, time ij Cijk (t) represents the transportation cost from i to k using transportation mode t, time

[0027] Where J represents the type of resource; set z ikt Cijk (t) represents the transportation cost from i to k using transportation mode t, time ikt is 1, otherwise 0, to calculate the scheduling cost; new_capacity ij = capacity ij -state ij (6);

[0028] Where new_capacity ij represents the latest number of j resources in i, state ij represents the occupation status of j resources in i; the above formula (6) represents the latest resource number of the node, the maintenance resource exists the occupied situation, and the latest resource number of the node is the resource number minus the occupied resource number;

[0029] If the scheme with the lowest cost under time constraints is selected, the following constraints are added:

[0030] Where time_max represents the maximum time constraint; if the solution with the lowest time under the cost constraint is selected, the following constraint is added:

[0031] Where cost_max represents the maximum cost constraint; by solving the hybrid integer programming model constructed above, the optimal objective is obtained.

[0032] Furthermore, the calculation of the fault repair time, repair cost, and repair probability for a certain maintenance activity specifically includes the following sub-steps:

[0033] The probability of repair during maintenance activities is calculated based on the failure rate predicted by reliability and the failure rate statistically analyzed from cases.

[0034] Where α and β are weighting coefficients, α+β=1; λ represents the failure rate predicted by the reliability of the component during the design phase, c_n represents the number of cases that have occurred during the operation and maintenance phase of the component, and c_s represents the total number of cases of the product during the operation and maintenance phase.

[0035] Let there be n types of maintenance activities a1, a2, ..., a n The repair probability for each maintenance activity is p1 to p2. n The execution time is t1~t n The maintenance cost is w1 to w n The set of all possible maintenance activities is SEQ{a(n)}=seq1,seq2,...,seq N If the maintenance activity sequence seq1 is a1, a2, ..., a n The average repair time for this sorting is: T m =t1+(1-p1)(t1+t2)+...+(1-p1)(1-p2)...(1-p n-1 (t1+t2+...+t) n (10);

[0036] Similarly, the average repair cost is: W m =w1+(1-p1)(w1+w2)+...+(1-p1)(1-p2)...(1-p n-1 (w1+w2+...+w) n (11).

[0037] Furthermore, the step of determining the objective of maintenance decision-making and outputting the optimal maintenance plan specifically includes the following sub-steps:

[0038] Let there be n types of maintenance activities a1, a2, ..., a n seq order mFor input, under the constraint condition of cost not more than cost_max, output the optimal maintenance activity sequence optimal_seq with the shortest time guarantee: W m ≤cost_max(12); fitness=Min(T m )(13);

[0039] Wherein, fitness is fitness function.

[0040] Further, when n is not greater than 5, the exhaustive method is used to traverse all the lists, and when n is greater than 5, the search algorithm is used for optimization.

[0041] The beneficial effects of the present application include:

[0042] The present application can assist the technical personnel in decision-making in the electronic product maintenance process, reduce the waste of time and resources caused by inconsistency and uncertainty of maintenance decision-making, guide the maintenance activities of remote operation and maintenance, reduce the maintenance time and reduce the operation and maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0044] Fig. 1 is a flowchart of a cross-domain maintenance decision-making method for maintenance network point and activity and resource linkage optimization provided in an embodiment of the present application;

[0045] Fig. 2 is a schematic diagram of a resource scheduling algorithm in an embodiment of the present application. DETAILED DESCRIPTION

[0046] All features disclosed in all embodiments in the present specification, or all steps in the implicitly disclosed methods or processes, can be combined and / or extended, replaced, or substituted, in any manner, except for mutually exclusive features and / or steps.

[0047] The specific implementation process of the present application is as follows:

[0048] In a preferred embodiment, as shown in Figs. 1 and 2, a cross-domain maintenance decision-making method for maintenance network point and activity and resource linkage optimization is specifically provided, including the following steps:

[0049] S1, obtaining a recommended set of multiple maintenance activities from an operation and maintenance platform according to the fault phenomenon; in a further implementation, according to the fault phenomenon, the fault cause and its binding troubleshooting activities and required spare parts or instrument tooling are obtained by searching or knowledge reasoning.

[0050] S2, judging whether the required resources of the fault product at the current network point are missing, marking the insufficient maintenance activities, and jumping to step S3 if there is a resource missing condition; if there is no resource missing condition, jumping to step S4.

[0051] S3, calculating the resource scheduling time and cost required for each resource-lacking maintenance activity; in a further implementation, taking the resource demand type and quantity selected by the user or the maintenance decision model as input, based on the resource distribution and use state in the support network, a resource scheduling scheme is provided according to the selected resource scheduling strategy.

[0052] First, a mathematical model is established according to the problem, and a resource scheduling objective function is constructed:

[0053] where w1 and w2 are the weight proportions of cost and time, respectively, if w1 = 1 and w2 = 0, the resource scheduling cost is minimized, if w1 = 0 and w2 = 1, the scheduling time is minimized, min represents the minimum value function, i represents the network point, k represents the network point, t represents the transportation mode, I represents the number of network points, z ikt represents whether to call resources from i network point to k network point by transportation mode t; cost tik represents the transportation cost from i network point to k network point under transportation mode t, time tik represents the scheduling time from i network point to k network point under transportation mode t.

[0054] When i = R:

[0055] where T represents the type of transportation mode, x kijt represents the number of j resources called from k network point to i network point under transportation mode t, demand j represents the demand of j resources. The above formula indicates that the resources called from all network points must meet the demand of the demand network point, where R represents the network point position requiring resources.

[0056] The above formula indicates that no resources are called from the demand network point, where R represents the network point position requiring resources.

[0057] When i ≠ R:

[0058] where x ikjtdenotes the number of resource j called from i to k by transportation mode t, capacity ij denotes the number of resource j in i. The above equation denotes that the number of resource j called from i is less than or equal to the capacity of resource j in i and the number of resource j called from other sites.

[0059] where J denotes the type of resource. Since the resource scheduling cost of a site is irrelevant to the number of called resources, z ikt denotes whether resource is called from i to k. If called, z ikt is 1, otherwise 0, which is used to calculate the scheduling cost. new_capacity ij = capacity ij -state ij ;

[0060] where new_capacity ij denotes the latest number of resource j in i, state ij denotes the occupation state of resource j in i. The above equation denotes that the latest resource number of a site is the resource number minus the occupied resource number, since the maintenance resource exists in the occupied state.

[0061] If the scheme with the lowest cost under time constraint is selected, the following constraint is added:

[0062] where time_max denotes the maximum time constraint; if the scheme with the lowest time under cost constraint is selected, the following constraint is added:

[0063] where cost_max denotes the maximum cost constraint. The above mixed integer programming model is solved to obtain the optimal objective.

[0064] S4, based on statistics, the failure repair time, repair cost and repair probability of the maintenance activity are obtained; in a further implementation, the repair probability of the maintenance activity is calculated based on the failure rate of reliability prediction and the case statistical failure rate:

[0065] where a and β are weighting coefficients, a + β = 1; λ denotes the failure rate of a component obtained by reliability prediction in the design stage, c_n denotes the number of cases of the component in the operation and maintenance stage, and c_s denotes the total number of cases of the product in the operation and maintenance stage.

[0066] Suppose there are n kinds of maintenance activities a1, a2,..., a n , and the repair probability of each maintenance activity is p1~p n, the execution time is t1~t n , the maintenance cost is w1~w n , all possible maintenance activity sequencing set is SEQ{a(n)} = seq1, seq2,..., seq N , if the maintenance activity sequencing seq1 is a1, a2,..., a n , then the average maintenance time of the sequencing is: T m = t1+ (1-p1) (t1+t2)+...+(1-p1) (1-p2)...(1-p n-1 ) (t1+t2+...+t n ).

[0067] Similarly, the average maintenance cost is: W m = w1+ (1-p1) (w1+w2)+...+(1-p1) (1-p2)...(1-p n-1 ) (w1+w2+...+w n ).

[0068] S5, comprehensive cost and time factors, determine the target of maintenance decision, set the objective function, use optimization algorithm to traverse all possible maintenance activity sequencing, output the optimal maintenance scheme. Specifically, the order seq n of n maintenance activities a1, a2,..., a m is input, the cost is not more than cost_max as a constraint condition, in the case of ensuring the shortest time, output the optimal maintenance activity sequencing optimal_seq: W m ≤ cost_max fitness = Min(T m )

[0069] Wherein, fitness is the fitness function of this model, it is recommended to use exhaustive method to traverse all when n is not greater than 5, when n is greater than 5, use intelligent search algorithm for optimization.

[0070] The cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization provided by the application can be applied to multiple fields. First, a recommended set of multiple maintenance activities is obtained from an operation and maintenance platform according to a fault phenomenon; then, it is judged whether the required resources of the fault product at the current network point are missing, and the maintenance activities that are insufficient are marked; if there is a resource missing condition, the resource scheduling time and cost required by each maintenance activity lacking resources are calculated respectively using a mixed integer programming model; then, the fault repair time, repair cost and repair probability of the maintenance activity are obtained based on statistics; finally, the target of the maintenance decision is determined, the objective function is set, the optimization algorithm is used to sort and traverse all possible maintenance activities, and the optimal maintenance scheme is output. The application can assist technicians in decision-making in the process of electronic product maintenance, reduce the waste of time and resources caused by inconsistency and uncertainty of maintenance decision-making, guide the maintenance activities of remote operation and maintenance, reduce the maintenance time and reduce the operation and maintenance cost.

[0071] It should be noted that the following embodiments can be combined and / or extended, replaced, in any logical manner within the scope of protection defined in the claims of the application, for example, disclosed technical principles, disclosed technical features or implied disclosed technical features.

[0072] In other embodiments, including but not limited to the following embodiments:

[0073] Embodiment 1

[0074] A cross-domain maintenance decision method for maintenance network point and activity and resource linkage optimization includes the following steps:

[0075] According to the fault phenomenon, a set of multiple maintenance activities is obtained;

[0076] It is judged whether the required resources of the fault product at the current network point are missing, and the maintenance activities that are insufficient are marked; if there is a resource missing condition, the resource scheduling time and cost required by each maintenance activity lacking resources are calculated respectively using a mixed integer programming model; then, the fault repair time, repair cost and repair probability of the maintenance activity are obtained based on statistics; finally, the target of the maintenance decision is determined, the objective function is set, the optimization algorithm is used to sort and traverse all possible maintenance activities, and the optimal maintenance scheme is output. The application can assist technicians in decision-making in the process of electronic product maintenance, reduce the waste of time and resources caused by inconsistency and uncertainty of maintenance decision-making, guide the maintenance activities of remote operation and maintenance, reduce the maintenance time and reduce the operation and maintenance cost.

[0077] The fault repair time, repair cost and repair probability of a certain maintenance activity are calculated.

[0078] The target of the maintenance decision is determined, and the optimal maintenance scheme is output.

[0079] Embodiment 2

[0080] On the basis of embodiment 1, the set of multiple maintenance activities is obtained from an operation and maintenance platform.

[0081] Embodiment 3

[0082] On the basis of embodiment 1, the resource scheduling time and cost required by each resource-lacking maintenance activity are calculated respectively, specifically including: using a mixed integer programming model to calculate the resource scheduling time and cost required by each resource-lacking maintenance activity respectively.

[0083] Embodiment 4

[0084] On the basis of embodiment 1, the target of the maintenance decision is determined, and the optimal maintenance scheme is output, specifically including: determining the target of the maintenance decision by comprehensively considering the cost and time factors, setting the objective function, using an optimization algorithm to traverse all possible maintenance activity sequences, and outputting the optimal maintenance scheme.

[0085] Embodiment 5

[0086] On the basis of embodiment 1, the set of multiple maintenance activities is obtained according to the fault phenomenon, specifically including the sub-steps of: obtaining the fault cause and its bound troubleshooting activities and the required spare parts or instrument tooling by searching or knowledge reasoning according to the fault phenomenon.

[0087] Embodiment 6

[0088] On the basis of embodiment 1, whether the required resources of the fault product at the current network point are missing is judged, specifically including the sub-steps of: if there is no resource missing, then the fault repair time, repair cost and repair probability of a certain maintenance activity are calculated.

[0089] Embodiment 7

[0090] On the basis of embodiment 1 or embodiment 6, the fault repair time, repair cost and repair probability of a certain maintenance activity are calculated, specifically including the sub-steps of:

[0091] Taking the maintenance decision model or the resource demand type and quantity selected by the user as input, based on the resource distribution and use state in the support network, according to the selected resource scheduling strategy, a resource scheduling scheme is provided, specifically including:

[0092] First, a mathematical model of the problem is established, and a resource scheduling objective function is constructed:

[0093] Where w1 and w2 are the weight proportions of cost and time respectively, if w1 = 1, w2 = 0, the resource scheduling cost is the lowest, if w1 = 0, w2 = 1, the scheduling time is the lowest, min represents the minimum value function, i represents the network point, k represents the network point, t represents the transportation mode, I represents the number of network points, z ikt represents whether to call resources from i network point to k network point by transportation mode t; cost tik represents the transportation cost from i network point to k network point under transportation mode t, timetik denotes the dispatch time from i to k using transportation mode t;

[0094] When i = R:

[0095] where T denotes the transportation mode type, x kijt denotes the number of j resources called from k to i using transportation mode t, demand j denotes the demand of j resources; the above equation (2) denotes that the resources called from all the nodes must satisfy the demand of the demand nodes, where R denotes the node position requiring resources;

[0096] The above equation (3) denotes that no resources are called from the demand nodes, where R denotes the node position requiring resources;

[0097] When i ≠ R:

[0098] where T denotes the transportation mode type, x ikjt denotes the number of j resources called from i to k using transportation mode t, capacity ij denotes the number of j resources in i; the above equation (4) denotes that the number of j resources called from node i is less than or equal to the capacity of j resources in node i and the sum of the number of j resources called from other nodes;

[0099] where J denotes the type of resources; z ikt denotes whether resources are called from node i to node k, and if called, z ikt is 1, otherwise 0, which is used to calculate the dispatch cost; new_capacity ij = capacity ij -state ij (6);

[0100] where new_capacity ij denotes the latest number of j resources in i, state ij denotes the occupation state of j resources in i; the above equation (6) denotes that the latest resource number of the node, the maintenance resources exist the occupied situation, the latest resource number of the node is the resource number minus the occupied resource number;

[0101] If the scheme with the lowest cost under the time constraint is selected, the following constraint is added:

[0102] where time_max denotes the maximum time constraint; if the scheme with the lowest time under the cost constraint is selected, the following constraint is added:

[0103] wherein, cost_max represents the maximum cost constraint; solving the above constructed mixed integer programming model to obtain the optimal objective.

[0104] Embodiment 8

[0105] On the basis of embodiment 7, the calculating the failure repair time, repair cost and repair probability of a certain maintenance activity specifically comprises the following sub-steps:

[0106] The repair probability of the maintenance activity is calculated based on the failure rate of reliability prediction and the case statistical failure rate:

[0107] wherein, a and β are weighting coefficients, a + β = 1; λ represents the failure rate of the component obtained in the design stage of reliability prediction, c_n represents the number of cases of the component in the operation and maintenance stage, and c_s represents the total number of cases of the product in the operation and maintenance stage;

[0108] Suppose there are n maintenance activities a1, a2,..., a n , the repair probability of each maintenance activity is p1~p n , the execution time is t1~t n , the maintenance cost is w1~w n , and all possible maintenance activity sorting sets are SEQ{a(n)} = seq1, seq2,..., seq N , if the maintenance activity sorting seq1 is a1, a2,..., a n , then the average maintenance time of the sorting is: T m = t1 + (1-p1)(t1+t2) +... + (1-p1)(1-p2)...(1-p n-1 )(t1+t2+...+t n )(10);

[0109] Similarly, the average maintenance cost is: W m = w1 + (1-p1)(w1+w2) +... + (1-p1)(1-p2)...(1-p n-1 )(w1+w2+...+w n )(11).

[0110] Embodiment 9

[0111] On the basis of embodiment 8, the determining the target of maintenance decision and outputting the optimal maintenance scheme specifically comprises the following sub-steps:

[0112] In the order seq n of the n maintenance activities a1, a2,..., a mFor input, with the constraint of cost not more than cost_max, output the optimal maintenance activity sequence optimal_seq in the case of ensuring the shortest time: W m ≤ cost_max (12); fitness = Min(T m )(13).

[0113] Wherein, fitness is fitness function.

[0114] Embodiment 10

[0115] On the basis of embodiment 9, when n is not greater than 5, the exhaustive method is used to traverse all the lists, and when n is greater than 5, the search algorithm is used for optimization.

[0116] The units described in the embodiments of the present application can be implemented in the form of software, or can be implemented in the form of hardware, and the described units can also be arranged in a processor. In some cases, the name of the unit does not constitute a limitation on the unit itself.

[0117] According to an aspect of the embodiments of the present application, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in the various optional implementation manners described above.

[0118] As another aspect, the embodiments of the present application also provide a computer readable medium, which can be included in the electronic device described in the above embodiments, or can exist separately without being assembled into the electronic device. The above computer readable medium carries one or more programs, which, when executed by the electronic device, enable the electronic device to implement the method described in the above embodiments.

Claims

1. A cross-domain maintenance decision-making method for optimizing the linkage between maintenance outlets, activities, and resources, characterized in that, Includes the following steps: Based on the fault symptoms, obtain a set of various maintenance activities; Determine whether the required resources for the faulty product are lacking at the current site. If resources are lacking, calculate the resource scheduling time and cost required for each type of repair activity where resources are lacking. Calculate the fault repair time, repair cost, and repair probability for a certain maintenance activity; Define the objectives of maintenance decisions and output the optimal maintenance solution.

2. The cross-domain maintenance decision-making method for optimizing maintenance outlets, activities, and resources according to claim 1, characterized in that, The acquisition of the collection of various maintenance activities includes acquisition from the operation and maintenance platform.

3. The cross-domain maintenance decision-making method for optimizing maintenance outlets, activities, and resources according to claim 1, characterized in that, The step of calculating the resource scheduling time and cost required for each type of maintenance activity lacking resources specifically includes: using a mixed integer programming model to calculate the resource scheduling time and cost required for each type of maintenance activity lacking resources.

4. The cross-domain maintenance decision-making method for optimizing maintenance network, activities, and resources according to claim 1, characterized in that, The process of determining the objective of maintenance decisions and outputting the optimal maintenance plan specifically includes: considering cost and time factors, determining the objective of maintenance decisions, setting an objective function, using an optimization algorithm to traverse all possible maintenance activities in a sorted manner, and outputting the optimal maintenance plan.

5. The cross-domain maintenance decision-making method for optimizing maintenance outlets, activities, and resources according to claim 1, characterized in that, The process of obtaining a set of various maintenance activities based on the fault phenomenon includes the following sub-steps: based on the fault phenomenon, using retrieval or knowledge reasoning, obtaining the fault cause and its associated troubleshooting activities, as well as the required spare parts or instrument tooling.

6. The cross-domain maintenance decision-making method for optimizing maintenance network, activities, and resources according to claim 1, characterized in that, The step of determining whether the required resources for the faulty product at the current site are missing includes the following sub-steps: if there is no resource shortage, then the calculation of the fault repair time, repair cost and repair probability for a certain repair activity is performed.

7. The cross-domain maintenance decision-making method for optimizing maintenance network, activities, and resources according to any one of claims 1 or 6, characterized in that, The calculation of the fault repair time, repair cost, and repair probability for a certain maintenance activity specifically includes the following sub-steps: Taking the maintenance decision model or the types and quantities of resource requirements selected by the user as input, and based on the resource distribution and usage status within the network, a resource scheduling scheme is provided according to the selected resource scheduling strategy. Specifically, this includes: First, establish a mathematical model for the problem and construct the resource scheduling objective function: Where w1 and w2 are the weighted proportions of cost and time, respectively. If w1 = 1 and w2 = 0, the resource scheduling cost is minimized; if w1 = 0 and w2 = 1, the scheduling time is minimized. min represents the minimum value function, i represents the network point, k represents the network point, t represents the transportation mode, I represents the number of network points, and z represents the number of network points. ikt Indicates whether to use transportation method t to request resources from network point i to network point k; cost tik This represents the transportation cost from point i to point k under transportation mode t, where time is the time factor. tik This represents the scheduling time from point i to point k under transportation mode t. When i = R: Where T represents the mode of transport, x kijt This represents the number of resources (j) requested from network point (k) to network point (i) under transportation mode (t). j Let R represent the demand for resource j; Equation (2) above means that the resources transferred from all network points must meet the demand of the network points, where R represents the location of the network point that needs the resources; Equation (3) above means that resources are not transferred from the demanding network point, where R represents the location of the network point that needs resources; When i≠R: Where, x ikjt This represents the number of resources (capacity) that can be transferred from network point i to network point k using transportation method t. ij Let i represent the quantity of j resources in i network point; Equation (4) above indicates that the quantity of j resources transferred from network point i is less than or equal to the sum of the capacity of j resources in network point i and the quantity of j resources transferred from other network points; Where J represents the resource type; z is set ikt This indicates whether to request resources from network point i to network point k. If requested, then z... ikt A value of 1 indicates a change in the time limit, while a value of 0 indicates a change in the time limit, which is used to calculate the scheduling cost. new_capacity ij =capacity ij -state ij (6); Among them, new_capacity ij The state represents the number of resources j in the latest i-th network point. ij This indicates the occupancy status of resource j in network point i; Equation (6) above indicates the latest resource quantity of the network point. Maintenance resources may be occupied, and the latest resource quantity of the network point is the resource quantity minus the occupied resource quantity; If we choose the solution with the lowest cost under the time constraint, then we add the following constraint: Where time_max represents the maximum time constraint; if the solution with the lowest time under the cost constraint is selected, the following constraint is added: Where cost_max represents the maximum cost constraint; by solving the hybrid integer programming model constructed above, the optimal objective is obtained.

8. The cross-domain maintenance decision-making method for optimizing maintenance network, activities, and resources according to claim 7, characterized in that, The calculation of the fault repair time, repair cost, and repair probability for a certain maintenance activity specifically includes the following sub-steps: The probability of repair during maintenance activities is calculated based on the failure rate predicted by reliability and the failure rate statistically analyzed from cases. Where α and β are weighting coefficients, α+β=1; λ represents the failure rate predicted by the reliability of the component during the design phase, c_n represents the number of cases that have occurred during the operation and maintenance phase of the component, and c_s represents the total number of cases of the product during the operation and maintenance phase. Let there be n types of maintenance activities a1, a2, ..., a n The repair probability for each maintenance activity is p1 to p2. n The execution time is t1~t n The maintenance cost is w1 to w n The set of all possible maintenance activities is SEQ{a(n)}=seq1,seq2,...,seq N If the maintenance activity sequence seq1 is a1, a2, ..., a n Therefore, the average repair time for this sorting is: T m =t1+(1-p1)(t1+t2)+...+(1-p1)(1-p2)...(1-p n-1 )(t1+t2+...+t n )(10); Similarly, the average repair cost is: W m =w1+(1-p1)(w1+w2)+...+(1-p1)(1-p2)...(1-p n-1 )(w1+w2+...+w n )(11)。 9. The cross-domain maintenance decision-making method for optimizing maintenance outlets, activities, and resources according to claim 8, characterized in that, The objective of determining maintenance decisions and outputting the optimal maintenance plan specifically includes the following sub-steps: Let there be n types of maintenance activities a1, a2, ..., a n seq order m As input, with the constraint that the cost does not exceed cost_max, output the optimal maintenance activity sequence optimal_seq while ensuring the shortest possible time. W m ≤cost_max(12); fitness=Min(T m )(13); Where fitness is the fitness function.

10. The cross-domain maintenance decision-making method for optimizing maintenance network, activities, and resources according to claim 9, characterized in that, When n is not greater than 5, an exhaustive search method is used to traverse all sequences; when n is greater than 5, a search algorithm is used to find the optimal sequence.

Images