A complex system and a method for evaluating task reliability of the complex system, and a computer program product

Abstract

This invention relates to a complex system and its task reliability assessment method and computer program product, belonging to the field of reliability modeling technology. The method divides the target task into several stages; decomposes the complex system to be modeled into several modules with independent operating functions, and determines the modules participating in each stage task; determines the redundancy strategy for each stage task according to the execution logic of the modules executing the target task within the system; constructs a reliability model for the corresponding stage task based on the redundancy strategy, obtaining the reliability of each stage task; and obtains the overall task reliability model of the system based on the reliability models of all stage tasks. The overall task reliability model is then used to assess the task reliability of the system. This method cleverly decomposes a loosely coupled complex system with dynamic tasks into independent functional modules, improving the accuracy of task reliability.

Description

Technical Field

[0001] This invention relates to a method for assessing the reliability of complex systems and their tasks, as well as a computer program product, belonging to the field of reliability model technology. Background Technology

[0002] Complex systems are systems composed of multiple interconnected and interacting components (or subsystems), characterized by nonlinearity, dynamism, and adaptability. Loose coupling refers to a system where the dependencies between its components (or subsystems) are low, allowing them to operate independently or be replaced. For example, a logistics system is a typical loosely coupled complex system. The tasks of a logistics system dynamically adjust according to changes in the type and quantity of goods, and each component (e.g., elevators, loaders, transport vehicles, stacker trucks, dispatch computers, etc.) can operate independently or be replaced, exhibiting loose coupling characteristics (i.e., low dependencies between system components, allowing them to operate or be replaced individually). A logistics system should ensure reliable operation even under dynamically changing tasks.

[0003] In existing technologies, task reliability models for complex systems under dynamic tasks often employ a typical task reliability model—one involving the largest number of components and the longest task duration—as the calculation model for task reliability. Here, a dynamic task refers to a task state where the system input adjusts with changes in events, time, and environment. Clearly, using a typical task reliability model as the task reliability model for complex systems under dynamic tasks fails to consider the changing states of the task reliability model under dynamic tasks. The typical task reliability model has a single state, which does not reflect reality. Furthermore, traditional system task reliability models do not consider the fault characteristics of loosely coupled systems and do not model fault tolerance mechanisms. Additionally, traditional reliability models have the limitations of static models, often based on fixed task scenarios, making them difficult to adapt to real-time changing loads and environments. Scheduling algorithms are also disconnected from reliability; traditional multi-task scheduling algorithms prioritize efficiency and do not deeply integrate reliability assessment mechanisms.

[0004] Chinese invention patent application CN115563705A discloses a network for dynamically reconfiguring system functions, comprising the following steps: 1) hierarchical analysis of the dynamically reconfigurable system, wherein the hierarchical relationships between the main task, sub-tasks, functions, sub-functions, and operational resources of the dynamically reconfigurable system are analyzed, and a functional hierarchy diagram is established; 2) defining the nodes and edges of the system functional network model to be constructed, wherein the nodes include: functional nodes, sub-functional nodes, hardware module nodes, and software module nodes; 3) functional layer modeling; 4) hardware and software layer modeling; 5) system functional model: based on the constructed functional layer model and the hardware and software layer model, combined with the functional hierarchy diagram, a system functional network model is obtained, thereby supporting the system simulation and reliability modeling and evaluation requirements based on the functional model. Although hierarchical analysis is performed on the dynamically reconfigurable system, the functional hierarchy diagram obtained from the analysis is used to match the nodes and edges of the system functional network model to be constructed. As the system tasks expand and the operational data is updated and iterated, the number of nodes in the system functional network model increases during construction, making the generated system functional network model complex and unfavorable for reliability modeling and evaluation. Summary of the Invention

[0005] The purpose of this invention is to provide a method and computer program product for evaluating the reliability of complex systems and their tasks, in order to solve the problem that the reliability of loosely coupled complex systems under dynamic tasks cannot be accurately obtained.

[0006] To achieve the above objectives, this invention proposes a method for evaluating the reliability of complex systems, comprising the following steps:

[0007] 1) Divide the target task into several phase tasks; decompose the complex system to be modeled into several modules with independent operating functions, and determine the modules participating in each phase task.

[0008] 2) Determine the redundancy strategy for each stage task according to the execution logic of the module of each stage task when executing the target task in the system; based on the redundancy strategy of each stage task, construct the reliability model of the corresponding stage task to obtain the reliability of each stage task.

[0009] 3) Based on the reliability models of all stages of the task, obtain the overall task reliability model of the system, and use the overall task reliability model to evaluate the task reliability of the system.

[0010] Furthermore, the overall reliability model of the mission is expressed by the following formula:

[0011] ;

[0012] in, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

[0013] Furthermore, when the execution logic of the modules participating in the phase task is serial logic when executing the target task within the system, the redundancy strategy of the phase task is determined to be a serial strategy.

[0014] When the execution logic of the modules participating in a phase task is parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be a parallel strategy.

[0015] When the execution logic of the modules participating in a phase task is a hybrid serial-parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be the k / n strategy, where k is the number of modules required for the system to execute the target task in that phase task, and n is the total number of modules participating in that phase task.

[0016] Furthermore, it also includes: when constructing the reliability model of the phase task, using the communication link status between modules in the phase task as an influencing factor, and using the influencing shadow to correct the reliability model of the phase.

[0017] Furthermore, each stage task contains several independent task paths for executing that stage task. The method also includes: setting weight factors for all task paths according to the type and number of modules in each task path; and correcting the overall task reliability model based on the weight factors of all paths.

[0018] Furthermore, the weighting factor is expressed by the following formula:

[0019] ;

[0020] in, , where is the weight factor for path p; For the reliability of path p; Let P be the cost of path p; P be the set of paths; and p be the path index.

[0021] Furthermore, it also includes: when constructing the reliability model of the phase task, the communication link status between modules in the phase task will be used as an influencing factor.

[0022] According to the type and number of modules in each task path, a weight factor is set for each task path. Each stage task contains several independent task paths for executing that stage task.

[0023] The overall reliability model of the mission is modified using the aforementioned influence shadow and the aforementioned weighting factors.

[0024] Furthermore, it also includes: based on the module status of all modules and the communication link status between modules, calling the Monte Carlo simulation method to simulate all task paths of each stage task to obtain the optimal task path for executing the target task.

[0025] On the other hand, the present invention also proposes a complex system including a processor for performing the following method steps:

[0026] 1) Divide the target task into several phase tasks; decompose the complex system to be modeled into several modules with independent operating functions, and determine the modules participating in each phase task.

[0027] 2) Determine the redundancy strategy for each stage task according to the execution logic of the module of each stage task when executing the target task in the system; based on the redundancy strategy of each stage task, construct the reliability model of the corresponding stage task to obtain the reliability of each stage task.

[0028] 3) Based on the reliability models of all stages of the task, obtain the overall task reliability model of the system, and use the overall task reliability model to evaluate the task reliability of the system.

[0029] Furthermore, the overall reliability model of the mission is expressed by the following formula:

[0030] ;

[0031] in, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

[0032] Furthermore, when the execution logic of the modules participating in the phase task is serial logic when executing the target task within the system, the redundancy strategy of the phase task is determined to be a serial strategy.

[0033] When the execution logic of the modules participating in a phase task is parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be a parallel strategy.

[0034] When the execution logic of the modules participating in a phase task is a hybrid serial-parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be the k / n strategy, where k is the number of modules required for the system to execute the target task in that phase task, and n is the total number of modules participating in that phase task.

[0035] Furthermore, it also includes: when constructing the reliability model of the phase task, using the communication link status between modules in the phase task as an influencing factor, and using the influencing shadow to correct the reliability model of the phase.

[0036] Furthermore, each stage task contains several independent task paths for executing that stage task. The method also includes: setting weight factors for all task paths according to the type and number of modules in each task path; and correcting the overall task reliability model based on the weight factors of all paths.

[0037] Furthermore, the weighting factor is expressed by the following formula:

[0038] ;

[0039] in, , where is the weight factor for path p; For the reliability of path p; Let P be the cost of path p; P be the set of paths; and p be the path index.

[0040] Furthermore, it also includes: when constructing the reliability model of the phase task, the communication link status between modules in the phase task will be used as an influencing factor.

[0041] According to the type and number of modules in each task path, a weight factor is set for each task path. Each stage task contains several independent task paths for executing that stage task.

[0042] The overall reliability model of the mission is modified using the aforementioned influence shadow and the aforementioned weighting factors.

[0043] Furthermore, it also includes: based on the module status of all modules and the communication link status between modules, calling the Monte Carlo simulation method to simulate all task paths of each stage task to obtain the optimal task path for executing the target task.

[0044] On the other hand, the present invention also proposes a computer program product, including a computer program / instructions, which, when executed by a processor, implements the following method steps:

[0045] 1) Divide the target task into several phase tasks; decompose the complex system to be modeled into several modules with independent operating functions, and determine the modules participating in each phase task.

[0046] 2) Determine the redundancy strategy for each stage task according to the execution logic of the module of each stage task when executing the target task in the system; based on the redundancy strategy of each stage task, construct the reliability model of the corresponding stage task to obtain the reliability of each stage task.

[0047] 3) Based on the reliability models of all stages of the task, obtain the overall task reliability model of the system, and use the overall task reliability model to evaluate the task reliability of the system.

[0048] Furthermore, the overall reliability model of the mission is expressed by the following formula:

[0049] ;

[0050] in, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

[0051] Furthermore, when the execution logic of the modules participating in the phase task is serial logic when executing the target task within the system, the redundancy strategy of the phase task is determined to be a serial strategy.

[0052] When the execution logic of the modules participating in a phase task is parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be a parallel strategy.

[0053] When the execution logic of the modules participating in a phase task is a hybrid serial-parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be the k / n strategy, where k is the number of modules required for the system to execute the target task in that phase task, and n is the total number of modules participating in that phase task.

[0054] Furthermore, it also includes: when constructing the reliability model of the phase task, using the communication link status between modules in the phase task as an influencing factor, and using the influencing shadow to correct the reliability model of the phase.

[0055] Furthermore, each stage task contains several independent task paths for executing that stage task. The method also includes: setting weight factors for all task paths according to the type and number of modules in each task path; and correcting the overall task reliability model based on the weight factors of all paths.

[0056] Furthermore, the weighting factor is expressed by the following formula:

[0057] ;

[0058] in, , where is the weight factor for path p; For the reliability of path p; Let P be the cost of path p; P be the set of paths; and p be the path index.

[0059] Furthermore, it also includes: when constructing the reliability model of the phase task, the communication link status between modules in the phase task will be used as an influencing factor.

[0060] According to the type and number of modules in each task path, a weight factor is set for each task path. Each stage task contains several independent task paths for executing that stage task.

[0061] The overall reliability model of the mission is modified using the aforementioned influence shadow and the aforementioned weighting factors.

[0062] Furthermore, it also includes: based on the module status of all modules and the communication link status between modules, calling the Monte Carlo simulation method to simulate all task paths of each stage task to obtain the optimal task path for executing the target task.

[0063] The beneficial effects of this invention are as follows: The target task is divided into several phase tasks; the complex system to be modeled is decomposed into several modules with independent operating functions, and the modules participating in each phase task are determined; the redundancy strategy for each phase task is determined according to the execution logic of the modules executing the target task within the system; based on the redundancy strategy of each phase task, a reliability model for the corresponding phase task is constructed to obtain the reliability of each phase task; based on the reliability models of all phase tasks, the overall task reliability model of the system is obtained, and the overall task reliability model is used to evaluate the task reliability of the system, cleverly decomposing the loosely coupled complex system of dynamic tasks into independent functional modules; and the task is divided into multiple phases, reliability modeling is performed according to each task phase, and redundancy strategies are considered when constructing the reliability models of each task phase to improve the accuracy of task reliability, using highly accurate phase reliability models to generate the overall task reliability model. Attached Figure Description

[0064] Figure 1 This is a flowchart of a task reliability assessment method for complex systems proposed in this invention;

[0065] Figure 2 This is a schematic diagram of the task partitioning steps in a practical application scenario of the task reliability assessment method for complex systems proposed in this invention.

[0066] Figure 3 This is a flowchart of a task reliability assessment method for complex systems proposed in this invention in a practical application scenario;

[0067] Figure 4 This is a schematic diagram of the system structure of a complex system proposed in this invention when calculating the reliability of a task in a practical application scenario. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0069] The inventive concept of this invention is as follows: Addressing the dynamic and modular characteristics of task reliability in loosely coupled complex systems, the complex system is decomposed into several independent functional modules. Furthermore, the dynamic task is decomposed into stages according to the objective. A redundancy strategy is introduced to perform path redundancy analysis on each stage of the decomposed dynamic task. In addition to the above, the impact of communication link reliability on the overall system reliability is also considered. Through a task-resource-environment dynamic mapping mechanism, real-time reliability assessment and adaptive optimization are achieved, solving the failure problem of traditional models in dynamic scenarios.

[0070] It should be noted that, in this invention, the dynamic task refers to a task state in which the system input changes with events, time, and environment; the loose coupling refers to the low dependency between the components, which can operate or be replaced independently; the complex system refers to a system composed of multiple interconnected and interacting components (or subsystems); and the loosely coupled complex system based on dynamic tasks refers to a system that runs in a dynamic task environment and also possesses both loose coupling and complex system characteristics.

[0071] Detailed implementation method 1:

[0072] like Figure 1 The diagram shows a flowchart of a task reliability assessment method for complex systems proposed in this invention, which includes steps S11, S12, and S13, specifically:

[0073] Step S11: Divide the target task into several phase tasks; decompose the complex system to be modeled into several modules with independent operating functions, and determine the modules participating in each phase task; here, the target task is a dynamic task that adjusts the task status in real time. When dividing the target task, the division operation can be carried out according to the task flow, task objectives, or task characteristics. For example, when the target task is "to complete the sorting and delivery of N kinds of goods from the warehouse to the user", the target is divided into three phases: "sorting, transportation, and delivery" according to the task flow.

[0074] Step S12: Determine the redundancy strategy for each stage task according to the execution logic of the modules executing the target task within the system. Based on the redundancy strategy of each stage task, construct the reliability model of the corresponding stage task to obtain the reliability of each stage task. Here, the execution logic includes, but is not limited to, serial logic, parallel logic, and hybrid serial-parallel logic. When the execution logic of the modules participating in the stage task is serial logic, the redundancy strategy for that stage task is determined to be a serial strategy. When the execution logic of the modules participating in the stage task is parallel logic, the redundancy strategy for that stage task is determined to be a parallel strategy. When the execution logic of the modules participating in the stage task is hybrid serial-parallel logic, the redundancy strategy for that stage task is determined to be a k / n strategy, where k is the number of modules required to participate in the system's execution of the target task in that stage task, and n is the total number of modules participating in that stage task. Correspondingly:

[0075] The reliability model corresponding to the cascade strategy is expressed by the following formula:

[0076] ;

[0077] in, For the stage Reliability model; R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules.

[0078] The reliability model corresponding to the parallel strategy is expressed by the following formula:

[0079] ;

[0080] in, For the stage Reliability model; R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules.

[0081] The reliability model corresponding to the k / n strategy is expressed by the following formula:

[0082] ;

[0083] in, For the stage Reliability model; R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules in the system; k is the number of modules required for the system to run; j is the cumulative index.

[0084] Step S13: Based on the reliability models of all stages of the tasks, obtain the overall task reliability model of the system, and use the overall task reliability model to evaluate the system's task reliability; here, the overall task reliability model is expressed by the following formula:

[0085] ;

[0086] in, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

[0087] Through the above steps S11-S13, it is proposed to decompose complex systems into independent functional modules, and divide the phase tasks and functional modules participating in each phase task according to the task flow, etc.; it is proposed to perform reliability modeling according to each phase task to calculate the task reliability of each phase task; and to consider redundancy strategies according to the characteristics of each phase task and incorporate them into the reliability modeling, which can improve the accuracy of task reliability.

[0088] Method Detailed Implementation 2:

[0089] The following explanation, in conjunction with practical application scenarios, details a task reliability assessment method for complex systems proposed in this invention, including steps 1, 2, 3, 4, and 5.

[0090] Step 1: Decompose the loosely coupled complex system to be modeled into several modules with independent functions (for example, decompose into module 1, module 2, ..., module n, where n is a positive integer greater than or equal to 2); divide the task into several stages (stage 1, stage 2, ..., stage m, where m is a positive integer greater than or equal to 2) according to task characteristics / task process / task objectives.

[0091] Step 2 involves performing reliability modeling for the divided tasks across several stages. This includes modeling the impact of each module involved in a particular stage on the overall system performance (see [link to relevant documentation]). Figure 2 The redundancy strategy for each stage of the task is determined, and based on the redundancy strategy for each stage of the task, the reliability model for the corresponding stage of the task is determined.

[0092] Step 3: Based on all stage tasks and their reliability models, obtain the overall system reliability model. Preferably, perform a series calculation on the reliability models of each stage task (i.e., multiply the reliability models together) to obtain the overall system reliability model.

[0093] As system uptime and environment change, the communication link status between modules within the system becomes a significant factor affecting reliability. Therefore, step 4 integrates communication link reliability into each stage of the task, revising the reliability model for each stage. This involves incorporating communication link status to optimize the reliability model and improve reliability. Specifically, when constructing the reliability model for a stage task, the communication link status between the first and second modules within that stage is used as an influencing factor, and this influencing factor is used to revise the reliability model for that stage.

[0094] Step 5: Based on methods such as Monte Carlo simulation, the reliability model parameters of each stage of the task are dynamically updated in conjunction with the real-time status data of the system to generate the optimal task path. Here, Monte Carlo simulation refers to random sampling or statistical simulation, which solves the problem through a large number of simple repeated samplings and calculations. That is, in response to the real-time status data of the system, the reliability model parameters of each stage of the task are dynamically updated. The optimal task path is obtained through a large number of simulations. The optimal task path can be the path with the highest probability of success or the path with the lowest cost.

[0095] Through steps 1-5 above, the reliability model is constructed in stages, taking into account the reliability of the communication link and incorporating the effects of signal delay, packet loss rate, etc., into the reliability model, thereby improving the accuracy of task reliability. Furthermore, the dynamic task reliability results for various paths are applied, and methods such as Monte Carlo simulation are proposed for optimal selection.

[0096] Method Detailed Implementation 3:

[0097] Following the specific embodiments of the present invention described above, the specific steps for establishing the task reliability model of a loosely coupled complex system based on dynamic tasks according to steps 1-5 are as follows:

[0098] Based on step 1, the system consists of N independent modules M1, M2, M3, M4, ..., M NComposition: The modules are linked through information; Task characteristics: The task consists of K stages (such as preparation stage, execution stage, confirmation stage, etc.), and each stage can be completed through multiple paths; Dynamic task: The status of modules (such as model integrity, module load, etc.) and the status of communication links (whether they are available, whether there is a delay, etc.) change in real time.

[0099] Based on step 2, the task consists of K phases. Reliability modeling is performed in a layered manner, with layered calculations (i.e., modeling is divided into K layers, and reliability calculations are performed within each layer). Specifically:

[0100] First, take phase task S k For example, stage S k The path set is: Stage Task S k Let the corresponding set of independent paths be P. k Therefore, the total task path P is the Cartesian product of the K stage paths, and the total task path P is obtained by the following formula:

[0101] ;

[0102] Where P is the total task path; P1 is the set of independent paths for task S1 in phase 1; P2 is the set of independent paths for task S2 in phase 2; P k For the k-th stage task S k There is an independent set of paths. Here, the complete task execution path is composed of the paths selected at each stage, and all possible combinations constitute the complete solution space of the task. In practical applications, stage reliability synthesis avoids enumerating all combinations; in practical applications, based on the obtained set of paths, the modules and their number involved in the task can be determined, and redundancy strategies can also be decided.

[0103] Next, the redundancy strategy for each stage is determined: based on the impact of the modules participating in each stage's task on the system task, the redundancy strategy for each stage is determined. This redundancy strategy includes serial, parallel, and k / n strategies. The serial strategy is suitable when modules are connected sequentially within a stage, and the system function requires all modules to be functioning normally. The parallel strategy is used when multiple modules work in parallel, and the system function only needs to know that one module is functioning normally. The k / n strategy is suitable when at least k out of n modules are functioning normally, and the system can operate. The reliability models corresponding to different strategies are as follows:

[0104] a) Reliability model under the cascade strategy: Where R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules.

[0105] b) Reliability model under parallel strategy: Where R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules.

[0106] c) Reliability model under k / n strategy: Where R represents the stage. A unified identifier for the reliability of each module; n represents the stage. The number of modules in the system; k is the number of modules required for the system to run; j is the cumulative index.

[0107] Finally, the reliability mathematical model is determined:

[0108] ;

[0109] Among them, among them, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

[0110] Furthermore, based on the reliability mathematical model, dynamic weight optimization is introduced: Since each path (where a path refers to several independent task paths within each stage task for executing that stage task) involves different types and numbers of modules, their cost differences and priorities also differ, and weighting factors can be introduced. ( Under the premise of ensuring reliability, the lowest-cost and highest-efficiency path is prioritized. The weighting factor is expressed by the following formula:

[0111] ;

[0112] in, , where is the weight factor for path p; For the reliability of path p; Let P be the cost of path p; P be the set of paths; and p be the path index.

[0113] Furthermore, the decision to include communication links and weighting factors is based on the specific task characteristics of the actual system. For example, when the communication link is simple and highly reliable, its reliability can be ignored; when the task is averaged, the weighting factor can be disregarded. Therefore, different mathematical models for overall reliability will be obtained for different considerations:

[0114] When considering communication link reliability and weighting factors, the mathematical model for overall mission reliability is expressed by the following formula:

[0115] ;

[0116] Among them, R 总 For the overall reliability mathematical model; Let be the weight factor of path p; P is the set of paths; p is a path in the set of paths P. The reliability factor of the communication link between module i and module j; For stage S K The reliability; K is the number of stages.

[0117] When communication link reliability is not considered, but weighting factors are taken into account, the mathematical model for overall mission reliability is expressed by the following formula:

[0118] ;

[0119] Among them, R 总 For the overall reliability mathematical model; Let be the weight factor of path p; P is the set of paths; p is a path in the set of paths P. For stage S K The reliability; K is the number of stages.

[0120] When communication link reliability and weighting factors are not considered, the mathematical model for overall mission reliability is expressed by the following formula:

[0121] ;

[0122] Among them, R 总 For the overall reliability mathematical model; For stage S K The reliability; K is the number of stages.

[0123] Method Detailed Implementation 4:

[0124] Following the specific embodiments described above, the loosely coupled complex system is preferably a logistics system, which consists of n1 dispatch centers, n2 warehouses (warehouses and sorting machines, with warehouses not participating in reliability modeling), n3 vehicles, n4 elevators, etc. The current task is to complete the sorting and delivery of N types of goods from the warehouses to the users.

[0125] The logistics system has the following characteristics: module independence and dynamic path change. Module independence: the failure probability of the modules such as the dispatch center, sorting machine, vehicles, and elevators is calculated independently, reflecting the low dependency of a loosely coupled system. Dynamic path change: the completion path of the system task, that is, the combination of participating equipment, is dynamically adjusted according to the changes in the types of goods required by the user. Each type of goods may also have multiple implementation paths.

[0126] The first step is to divide the current tasks of the logistics system into three stages: sorting, transportation, and delivery.

[0127] The second step is to determine the redundancy strategy for each stage: the redundancy strategy is determined based on the impact of the participating task modules in each stage on the current task. The result is that the redundancy strategy for all three stages is a series strategy (redundancy strategies include series strategy, parallel strategy, k / n strategy, etc.).

[0128] The third step is to perform reliability modeling for the three stages:

[0129] 31) The equipment involved in the sorting stage includes sorting machines and vehicles. The reliability mathematical model is as follows:

[0130] ;

[0131] In the formula, For the reliability of a single sorting machine; Redundancy strategy for sorting machines in the sorting stage; For the reliability of a single vehicle; Redundancy strategy for vehicles in the sorting stage.

[0132] 32) The equipment involved in the transportation phase includes elevators and vehicles. Dynamic selection is performed based on the transportation route, and the reliability mathematical model is as follows:

[0133] ;

[0134] in, The reliability of a single elevator; Redundancy strategy for elevators during the transportation phase; For the reliability of a single vehicle; A vehicle redundancy strategy for the transportation phase, from warehouse to elevator; A vehicle redundancy strategy for the transportation phase, from elevator to target floor.

[0135] 33) The equipment involved in the delivery phase is a vehicle, and the reliability mathematical model is:

[0136] ;

[0137] in, For the reliability of a single vehicle; Redundancy strategy for vehicles during the delivery phase.

[0138] Step 4, Dynamic Weighting Factor: This involves assigning a task proportion P to each type of goods. i As a dynamic weighting factor.

[0139] Step 5, Mathematical Model for Overall Mission Reliability: Here, we prefer to disregard communication link reliability and only consider dynamic weighting factors.

[0140] ;

[0141] Where n is the total number of goods types; For the task reliability of the i-th type of goods, .

[0142] Method Detailed Implementation 5:

[0143] Below, in conjunction with the attached diagram... Figure 3 The method for assessing the reliability of complex systems proposed in this invention will be explained.

[0144] like Figure 3 The diagram shows a flowchart of a task reliability assessment method for complex systems proposed in this invention in a practical application scenario. First, the loosely coupled complex system to be modeled is decomposed into several modules; the dynamic task is divided into several task stages; then, reliability modeling is performed for each task stage based on the divided task stages, and a redundancy strategy is introduced during the modeling process; then, based on the redundancy strategy for each task stage, the impact of each module failure on the system task is considered to determine all available system task implementation paths, ultimately obtaining the overall system reliability model under all paths.

[0145] Based on the overall system reliability model, the determination of whether to introduce communication links and weighting factors is made according to the actual system's task characteristics, ultimately resulting in three types of optimized overall reliability models: 1. A mathematical model of overall reliability considering communication link reliability and weighting factors (i.e., expressed by the following formula: Among them, R 总 For the overall reliability mathematical model; Let be the weight factor of path p; P is the set of paths; p is a path in the set of paths P. The reliability factor of the communication link between module i and module j; For stage S K The reliability of the communication link is K, which is the number of stages. The overall reliability mathematical model, which does not consider the reliability of the communication link but considers the weighting factor, is expressed by the following formula: Among them, R 总 For the overall reliability mathematical model; Let be the weight factor of path p; P is the set of paths; p is a path in the set of paths P. For stage S K The reliability of the communication link (K is the number of stages); the overall reliability mathematical model does not consider the reliability of the communication link or the weighting factor (i.e., it is expressed by the following formula: Among them, R 总 For the overall reliability mathematical model; For stage S K Reliability; K is the number of stages).

[0146] System implementation details:

[0147] On the other hand, the present invention also proposes a complex system including a processor for executing the steps of the task reliability assessment method for the complex system as described above.

[0148] like Figure 4 The diagram shows a system structure for calculating the reliability of a complex system in a real-world application scenario, as proposed in this invention. For several missions running in the complex system, each mission is divided into several stages, and a reliability model is constructed for each stage of each mission to obtain the reliability of each stage. In calculating the reliability of each stage of each mission, within a certain stage (stage n in the diagram), the reliability of that stage is obtained based on the task reliability of each module under the redundancy strategy. Thus, the task reliability of each mission is obtained, and finally, the total system reliability is obtained according to the task ratio and task reliability of each mission.

[0149] In addition, for specific implementation methods of the system, please refer to Specific Implementation Methods 1-5, which will not be repeated here.

[0150] On the other hand, the present invention also proposes a computer program product, including a computer program / instructions, which, when executed by a processor, implements the steps of the task reliability assessment method for the complex system described above. Specific embodiments of the computer program product can be found in Method Embodiments 1-5 and the System Embodiment, and will not be repeated here.

[0151] In summary, the key points and protection points of the complex system and its task reliability assessment method, and computer program product proposed in this invention include: 1) proposing to decompose the loosely coupled complex system with dynamic tasks into independent functional modules, and dividing the tasks into stages and functional modules participating in each stage according to the task flow; 2) proposing to perform reliability modeling according to each stage task to calculate the task reliability of each stage task; 3) proposing to consider redundancy strategies based on the characteristics of each stage task, including serial strategies, parallel strategies, k / n strategies, etc., and incorporate them into the reliability modeling to improve the accuracy of task reliability; 4) proposing to introduce weighting factors, with different weighting factors determined according to the different types and numbers of modules involved in each path, cost differences, and priorities, to improve the accuracy of task reliability; 5) proposing to consider the reliability of communication links, incorporating the effects of signal delay, packet loss rate, etc. into the reliability model to improve the accuracy of task reliability.

[0152] By employing a phased model and redundancy in dynamic paths, the reliability prediction accuracy of loosely coupled systems is improved. Furthermore, the system can dynamically adjust task paths based on equipment status data to adapt to complex environmental changes. In practical applications, it can also be applied to loosely coupled systems in logistics systems, collaborative intelligent manufacturing, power transmission, and internet information systems.

Claims

1. A method for assessing the reliability of a complex system, characterized in that, Includes the following steps: 1) Divide the target task into several phase tasks; decompose the complex system to be modeled into several modules with independent operating functions, and determine the modules participating in each phase task. 2) Determine the redundancy strategy for each stage task according to the execution logic of the module of each stage task when executing the target task in the system; based on the redundancy strategy of each stage task, construct the reliability model of the corresponding stage task to obtain the reliability of each stage task. 3) Based on the reliability models of all stages of the task, obtain the overall task reliability model of the system, and use the overall task reliability model to evaluate the task reliability of the system.

2. The method for assessing the reliability of complex systems according to claim 1, characterized in that, The overall reliability model of the mission is expressed by the following formula: ； in, Here is the overall reliability model for the task; m is the total number of tasks in the phase; i is the index of the task in the phase; k represents the reliability of modules 1 to n in phase task i; i This is the redundancy strategy for stage i.

3. The method for assessing the reliability of complex systems according to claim 1, characterized in that, When the execution logic of the modules involved in a phase task is serial logic when they execute the target task within the system, the redundancy strategy for that phase task is determined to be a serial strategy. When the execution logic of the modules participating in a phase task is parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be a parallel strategy. When the execution logic of the modules participating in a phase task is a hybrid serial-parallel logic when executing the target task within the system, the redundancy strategy for that phase task is determined to be the k / n strategy, where k is the number of modules required for the system to execute the target task in that phase task, and n is the total number of modules participating in that phase task.

4. The method for assessing the reliability of complex systems according to claim 1, characterized in that, Also includes: When constructing the reliability model of a phase task, the communication link status between modules in that phase task is used as an influencing factor, and the influencing shadow is used to correct the reliability model of that phase.

5. The method for assessing the reliability of a complex system according to claim 1, characterized in that, Each stage of the task contains several independent task paths for executing that stage. The method further includes: setting weight factors for all task paths according to the type and number of modules in each task path; and revising the overall task reliability model based on the weight factors of all paths.

6. The method for assessing the task reliability of a complex system according to claim 5, characterized in that, The weighting factor is expressed by the following formula: ； in, , where is the weight factor for path p; For the reliability of path p; Let P be the cost of path p; P be the set of paths; and p be the path index.

7. The method for assessing the reliability of a complex system according to claim 1, characterized in that, Also includes: When constructing a reliability model for a phase task, the communication link status between modules in that phase task will be used as an influencing factor. According to the type and number of modules in each task path, a weight factor is set for each task path. Each stage task contains several independent task paths for executing that stage task. The overall reliability model of the mission is modified using the aforementioned influence shadow and the aforementioned weighting factors.

8. The method for assessing the reliability of a complex system according to claim 1, characterized in that, Also includes: Based on the module status of all modules and the communication link status between modules, the Monte Carlo simulation method is used to simulate all task paths of each stage task to obtain the optimal task path for executing the target task.

9. A complex system, characterized in that, Includes a processor for performing the method steps as described in any one of claims 1-8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method steps as described in any one of claims 1-8.

Images