Demand prediction method based on equipment maintenance equipment consumption rule

Abstract

The invention relates to a demand prediction method based on a consumption rule of equipment maintenance equipment, and the method comprises the steps: S1, collecting the normal working time X and defect working time Y of the equipment, and carrying out the statistics of the normal distribution N1 (mu1, sigma12) obeyed by the normal working time X of the equipment and the normal distribution N2 (mu2, sigma22) obeyed by the defect working time Y of the equipment; s2, constructing a first equipment consumption rule model based on a time maintenance mode within the [0, T] time, constructing a second equipment consumption rule model based on a state maintenance mode at the T moment, and then constructing an equipment maintenance consumption average value within the [0, T] time; and S3, combining the average value of the equipment maintenance consumption with the number of the equipment serving the unit equipment and the safety inventory level of the equipment to obtain an equipment replenishment prediction model. According to the method, the consumption rule of equipment maintenance equipment is deeply excavated, so that the accuracy and reliability of equipment demand prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and particularly relates to a demand prediction method based on the consumption law of equipment maintenance materials. Background Art

[0002] Equipment maintenance materials are important guarantee resources required in the process of equipment maintenance and are the main factors constituting the costs of equipment use and maintenance support. Scientifically analyzing the consumption law of equipment maintenance materials and reasonably predicting and determining the demand for equipment maintenance support materials can not only reduce the equipment support costs, but also effectively shorten the equipment maintenance time and improve the equipment combat readiness rate.

[0003] Currently, there are many prediction technologies for equipment maintenance materials from the aspects of equipment use time or equipment technical status respectively. For example, for the time-based maintenance method, influencing factors such as the service life of components and equipment use time are analyzed, and an equipment spare parts prediction model based on the service life of components is established; a joint strategy for maintenance and spare parts ordering based on remaining life prediction, and the model is optimized using the renewal reward theory; using the discrete event simulation method, a spare parts prediction model based on condition-based maintenance is established, which can better solve the problems of evaluation accuracy and agility.

[0004] Generally speaking, most equipment maintenance materials prediction technologies use a series of models, algorithms and methods, etc. to obtain the demand for equipment maintenance materials. However, considering the common consumption law of equipment maintenance materials under the two types of maintenance methods based on time and status within the task cycle, and on this basis, constructing a demand prediction model for maintenance materials, and thus calculating and predicting the demand for maintenance materials is still blank. Therefore, it is necessary to adopt more advanced algorithms and models to deeply study the consumption law of equipment maintenance materials for the two types of maintenance methods based on time and status, and scientifically predict the demand for maintenance materials configuration of equipment maintenance support institutions to improve the accuracy and reliability of prediction. Summary of the Invention

[0005] An embodiment of the present invention provides a demand prediction method based on the consumption law of equipment maintenance materials, aiming to scientifically and reasonably predict the demand for maintenance consumption materials by deeply exploring the common law of maintenance consumption materials under the two types of maintenance methods based on time and status, and provide a scientific basis for the procurement and configuration of equipment maintenance materials.

[0006] To achieve the above object, the embodiment of the present invention provides the following technical solutions:

[0007] A demand prediction method based on the consumption law of equipment maintenance materials includes the following steps:

[0008] S1. Collect the normal working time X and defective working time Y of the materials, and count the normal distribution N that the normal working time X of the materials follows1 (μ 1 ,σ 1 2 ), and the normal distribution N 2 (μ 2 ,σ 2 2 ) that the working time Y in the defect stage follows.

[0009] S2. Construct the first equipment consumption law model based on time-based maintenance within the time period [0, T] according to the normal distribution N 1 (μ 1 ,σ 1 2 ), and construct the second equipment consumption law model based on condition-based maintenance at time T by combining the remaining life threshold of the equipment according to the normal distribution N 2 (μ 2 ,σ 2 2 ). Then, construct the average equipment maintenance consumption within the time period [0, T] according to the first equipment consumption law model and the second equipment consumption law model;

[0010] Among them, the first equipment consumption law model is

[0011]

[0012] The second equipment consumption law model is

[0013]

[0014] The average equipment maintenance consumption is

[0015]

[0016] S3. Combine the average equipment maintenance consumption with the quantity of equipment serving a unit and the safety inventory level of the equipment to obtain an equipment replenishment prediction model;

[0017]

[0018] In the formula:

[0019] Q is the predicted quantity of equipment demand;

[0020] S is the annual safety inventory level of the maintenance support agency;

[0021] q is the annual consumption of maintenance equipment.

[0022] W is the quantity of equipment in service in the troops;

[0023] L is the quantity of equipment components for a unit of equipment.

[0024] Further, before constructing the second equipment consumption law model, the maintenance equipment consumption probabilities of the equipment in the normal state and the defective state at time T are established respectively.

[0025] The maintenance equipment consumption probability of the equipment in the normal working state is

[0026]

[0027] When the equipment is in the defective working state and the remaining life is greater than the life threshold of the equipment, the maintenance equipment consumption probability is

[0028]

[0029] When the equipment is in the defective working state and the remaining life is less than the life threshold of the equipment, the maintenance equipment consumption probability is

[0030]

[0031] Combining the maintenance equipment consumption probabilities of the equipment in the normal working state and the defective working state, construct the second equipment consumption law model, that is, q 2 = P 1 ×0 + P 2 ×0 + P 3 ×1.

[0032] The embodiments of the present invention have the following advantages:

[0033] A demand prediction method based on the consumption law of equipment maintenance materials proposed by the present invention first collects the consumption data of the maintenance materials of the configured equipment of a certain establishment unit, including the equipment model, the single-equipment quantity of a certain unit of the equipment, the normal working time and defective working time of the materials under the time-based maintenance method, the remaining life threshold under the condition-based maintenance method, and the safety inventory level of the maintenance support organization of a certain unit of the equipment, etc., laying a data foundation for analyzing and mining the consumption law of the maintenance materials of a certain unit of the equipment and demand prediction. Then, based on the above collected data, construct a mathematical model for predicting the demand for materials based on the consumption law of equipment maintenance materials within [0, T], including a mathematical model for the consumption law of materials under the time-based maintenance method, a mathematical model for the consumption of materials under the condition-based maintenance method, and a model for predicting the demand for maintenance materials.

[0034] This method improves the accuracy and reliability of equipment requirement prediction by deeply exploring the consumption law of equipment maintenance materials; this technology has broad application prospects and important practical significance, further promoting the efficiency of equipment support, providing a scientific basis for the procurement and allocation of equipment maintenance materials, providing strong support for equipment support work, and can also be continuously trained and iterated through more sample data in the future to continuously optimize the algorithm, providing a scientific basis and theoretical support for the formulation of equipment maintenance material supply plans and the demonstration of equipment maintenance support systems. Brief Description of the Drawings

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required for use in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description are only exemplary, and for those of ordinary skill in the art, without creative efforts, other implementation drawings can be obtained based on the provided drawings.

[0036] The structures, ratios, sizes, etc. depicted in this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the limiting conditions under which the present invention can be implemented. Therefore, they do not have technical essence. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that can be covered by the technical content disclosed in the present invention.

[0037] Figure 1 For the deterioration process of equipment over training time

[0038] Figure 2 This is the method flow chart of a demand prediction method based on the consumption law of equipment maintenance materials provided in Embodiment 1 of the present invention;

[0039] Figure 3 For the equipment requirement prediction result based on the consumption law of equipment maintenance materials. Detailed Embodiments

[0040] The following specific embodiments illustrate the implementation manners of the present invention. Those familiar with this technology can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope protected by the present invention.

[0041] In this technology, the deterioration of equipment from "normal" to "fault" does not occur overnight, but shows a gradual deterioration process with the increase of training time, that is, from the normal working stage to the defective working stage, and then to the functional fault stage. The occurrence of equipment defect points and fault points has a certain randomness. Therefore, the normal working stage and the defective working stage are not fixed values but have randomness. After multiple cyclic iterations of the process of equipment deteriorating from "normal" to "fault", certain laws can be abstracted from it. These laws can guide the prediction and decision-making of equipment maintenance materials. Figure 1 It is the deterioration process of equipment with the increase of training time.

[0042] Since most current equipment requires both corrective maintenance after failure and preventive maintenance, this technology fully considers and analyzes the consumption law of maintenance materials for maintenance methods based on time and condition.

[0043] Among them, the maintenance method based on time mainly means that the equipment experiences the normal working stage and the defective working stage with the increase of training time, and is naturally replaced after reaching the functional fault stage, which is also called corrective maintenance after failure.

[0044] The maintenance method based on condition mainly uses detection instruments or external monitoring equipment to take replacement or repair maintenance methods after the equipment enters the defective working stage and before it evolves into a functional fault, so as to avoid the occurrence of functional faults, which is also called preventive maintenance or condition-based maintenance.

[0045] Whether it is the maintenance method based on time or the maintenance method based on condition, they both conform to the actual situation of equipment maintenance. According to their maintenance history records, analyzing the consumption of maintenance materials is traceable and there are certain abstract laws. On this basis, this method constructs a prediction algorithm for equipment maintenance materials based on the consumption law mining algorithm, providing a scientific basis and theoretical support for the procurement and allocation of equipment maintenance materials.

[0046] However, for equipment to deteriorate from the "normal" state to the "fault" state, it needs to go through the normal working stage and the defective working stage in sequence. The failure rates of these two stages are different, independent, and follow a certain distribution. Among them, the normal working failure rate of the equipment is lower than that of the defective stage. Therefore, generally speaking, the normal working time of the equipment is longer than that of the defective stage. According to the corrective maintenance method and preventive maintenance method after equipment failure, the two branches of maintenance methods based on time and condition are distinguished for analysis, and the common laws of the consumption of equipment maintenance materials are studied to provide theoretical support for the demand prediction of equipment maintenance materials.

[0047] Based on the analysis of the consumption quantity of equipment over time, within the time period [0, T], a time-based maintenance method is mainly adopted, that is, maintenance after failure. According to renewal theory, after the first piece of equipment reaches the functional failure stage through the normal working stage and the defective working stage, the second piece of equipment takes over the first one to continue working. The lifetimes of each unit are independent of each other. The consumption quantity of maintenance equipment generated within the time period [0, T] is the renewal times of the graded equipment.

[0048] Based on the analysis of the consumption quantity of equipment based on its state, the consumption of equipment at time T depends on the functional state of the equipment. Here, two cases are analyzed respectively when the inspection point at time T is in the normal working stage and the defective working stage of the equipment. When the equipment identified at the inspection point at time T is in the normal working state, no corresponding maintenance measures are taken; when the equipment at the detection point at time T is in the defective working state and is less than the service life threshold, it is replaced. If it is greater than the service life threshold, it is not replaced.

[0049] Based on the exploration and prediction of the consumption law of equipment over time and state, according to the consumption quantity of equipment maintenance materials in different training cycles, the (R, S, M) strategy is adopted, that is, after each training cycle R, based on the annual consumption quantity of maintenance materials, if the inventory level is lower than S, the inventory level is replenished to S. Through the analysis of the laws of equipment consumption and inventory replenishment quantity over the years, the demand M for equipment maintenance materials is predicted, providing a scientific basis and algorithm support for the procurement and allocation of maintenance materials by the equipment department.

[0050] Embodiment 1

[0051] Based on the above theory, as Figure 2 shown, a demand prediction method based on the consumption law of equipment maintenance materials includes the following steps:

[0052] S1. Collect the normal working time X and the defective working time Y of the equipment, and statistically obtain the normal distribution N 1 (μ 1 , σ 1 2 ) that the normal working time X of the equipment follows, and the normal distribution N 2 (μ 2 , σ 2 2 ) that the defective stage working time Y follows.

[0053] S2. Construct the first equipment consumption law model based on the time-based maintenance method within the time period [0, T] according to the normal distribution N 1 (μ 1 , σ 1 2 ), and according to the normal distribution N 2 (μ 2 , σ2 2 )Construct the second equipment consumption law model based on the condition-based maintenance mode at time T by combining the remaining life threshold of the equipment, and then construct the average equipment maintenance consumption within the time period [0, T] according to the first equipment consumption law model and the second equipment consumption law model;

[0054] Among them, the first equipment consumption law model is

[0055]

[0056] The second equipment consumption law model is

[0057]

[0058] The average equipment maintenance consumption is

[0059]

[0060] S3. Combine the average equipment maintenance consumption with the quantity of equipment for a unit of equipment and the equipment safety inventory level to obtain an equipment replenishment prediction model;

[0061]

[0062] In the formula:

[0063] Q is the predicted quantity of equipment demand;

[0064] S is the annual safety inventory level of the maintenance support agency;

[0065] q is the annual consumption of maintenance equipment.

[0066] W is the quantity of equipment in service in the troops;

[0067] L is the quantity of equipment components for a unit of equipment.

[0068] Embodiment 2

[0069] Since there are not only equipment in the normal working state but also equipment in the defective working state at time T, Embodiment 2 improves the construction of the second equipment consumption law model on the basis of Embodiment 1.

[0070] The equipment consumption at time T depends on the functional state of the equipment. Here, it is analyzed from two situations: the normal working stage and the defective working stage of the equipment at the inspection point T. When the equipment identified at the inspection point T is in the normal working state, no corresponding maintenance measures are taken; when the equipment at the detection point T is in the defective working state and is less than the service life threshold, it is replaced, and if it is greater than the service life threshold, it is not replaced.

[0071] Before constructing the second equipment consumption law model, first establish the maintenance equipment consumption probabilities of the equipment in the normal state and the defective state at time T respectively;

[0072] The maintenance equipment consumption probability when the equipment is in the normal working state is

[0073]

[0074] When the equipment is in the defective working state and the remaining life is greater than the life threshold of the equipment, the maintenance equipment consumption probability is

[0075]

[0076] When the equipment is in the defective working state and the remaining life is less than the life threshold of the equipment, the maintenance equipment consumption probability is

[0077]

[0078] Combining the maintenance equipment consumption probabilities of the equipment in the normal working state and the defective working state, construct the second equipment consumption law model, that is

[0079] The following introduces a specific example of the practice of this method:

[0080] Taking the demand prediction of tank equipment maintenance materials of a certain unit as an example, using the equipment demand prediction technology based on the consumption law mining algorithm proposed by the present invention, deeply analyze the consumption data of vehicle equipment maintenance materials, excavate the common laws of maintenance equipment consumption in two types of maintenance methods based on time and status, and construct a maintenance equipment demand prediction algorithm. The implementation process of the demand prediction algorithm based on the consumption law of equipment maintenance materials is mainly divided into 5 steps: data collection, law mining, instance substitution, algorithm solution, and demand prediction. The mathematical symbols and their meanings involved in each algorithm during the implementation process are shown in Table 1.

[0081] Table 1 Mathematical symbols and their meanings involved in the algorithm

[0082]

[0083] 1. Data collection

[0084] Collect 8 units of a certain type of tank equipment equipped by a certain military unit. For each tank equipment, a combination of post-failure maintenance and condition-based maintenance is adopted. Through data collection and preprocessing of a certain unit of this type of tank equipment, it is preliminarily statistically found that the normal working time X of a certain unit of this type of equipment follows a normal distribution of N(1000, 100^2), and the working time Y during the defect stage follows a normal distribution of N(100, 10^2). The number of single installations of this type of unit is 5. During condition-based maintenance, if its remaining life is less than 100h, it will be replaced, and the inventory level of the maintenance support agency for a certain unit of this equipment remains unchanged at 100 pieces.

[0085] 2. Rule Mining

[0086] Analyze and mine the consumption rules of maintenance materials for the unit within the time range of [0, 1500]h, determine the quantity of maintenance materials that the maintenance support agency of this type of equipment needs to store at the 1500h moment, and analyze the consumption rules and standards of maintenance materials under different annual trainings, so as to predict the quantity of maintenance materials that need to be applied to the superior department, thereby ensuring the safety inventory level of maintenance materials.

[0087] 3. Example Substitution

[0088] As described above, it can be known that the failure probability functions of the normal working time and the working time during the defect stage of this type of unit are respectively

[0089]

[0090] The failure probability functions of the cumulative normal working time and the cumulative working time during the defect stage of the tank materials within the time range of [0, 1500]h are respectively

[0091]

[0092] The average value of the consumption quantity of maintenance materials for post-failure maintenance of tank materials within the time range of [0, 1500]h is

[0093]

[0094] The average value of the consumption quantity of maintenance materials for condition-based maintenance of tank materials at the 1500h moment is

[0095]

[0096] Then the average value of the consumption quantity of maintenance materials for tank materials within the time range of [0, 1500]h is

[0097]

[0098]

[0099] Finally, using the mining tool of the data analysis subsystem of the big data management and analysis for joint operation equipment maintenance support, programming calculations are carried out to obtain q≈1.37.

[0100] 4. Algorithm Solving

[0101] According to the data collection results, it is known that the inventory level of this type of tank equipment maintenance support agency for such equipment is 100 pieces. There are 8 sets of a certain type of tank equipment equipped by a certain establishment unit, and the single equipment quantity of this type of unit is 5. The predicted value of the remaining maintenance equipment quantity stored by the equipment maintenance support agency within the time of [0, 1500] h is

[0102]

[0103] Therefore, for the prediction of equipment maintenance materials based on the consumption law mining algorithm, after 1500 h, the quantity of maintenance materials that a certain type of tank equipment needs to apply to the superior department is

[0104] M≈S - Q≈100 - 45.2≈55.

[0105] After calculation, after 1500 h, a certain type of tank equipment needs to apply to the superior department for M≈55 unit maintenance materials to maintain its safety inventory level.

[0106] 5. Demand Prediction

[0107] Finally, the mining tool of the data analysis subsystem of the big data management and analysis for joint operation equipment maintenance support can also be used to program and predict the quantity of equipment maintenance materials that the equipment maintenance support agency needs to configure during different annual training times.

[0108] As Figure 3 shown, as the annual training time of tank equipment increases, the consumption quantity of unit maintenance materials also increases. The equipment maintenance support agency can scientifically predict the quantity of maintenance materials configuration based on the consumption law of equipment maintenance materials obtained by analysis and mining, combined with the inventory safety level.

[0109] Although the present invention has been described in detail with general descriptions and specific embodiments above, based on the present invention, some modifications or improvements can be made, which are obvious to those skilled in the art. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection required by the present invention.

Claims

1. A demand forecasting method based on equipment maintenance equipment consumption law, characterized in that: The following steps are involved: S1. Collect the normal working time X and defective working time Y of the equipment, and calculate the normal distribution N1 (μ1, σ1) that the normal working time X of the equipment obeys. 2 ), and the defective phase working time Y obeys the normal distribution N2(μ2,σ2 2 ); S2, according to the normal distribution N1(μ1,σ1 2 ) constructs the first equipment consumption law model based on the time maintenance mode within [0, T] time, and according to the normal distribution N2(μ2, σ2 2 ) constructing a second equipment consumption law model based on the state maintenance mode at time T in combination with the remaining life threshold of the equipment, and then constructing the average equipment maintenance consumption value within the time [0, T] according to the first equipment consumption law model and the second equipment consumption law model; Among them, the first equipment consumption law model is: The second equipment consumption law model is: The average maintenance consumption of the equipment is S3. The average value of equipment maintenance consumption is combined with the number of equipment serving the unit equipment and the equipment safety stock level to obtain the equipment replenishment prediction model; Where: Q is the forecast quantity of equipment demand; S is the annual safety stock level of the maintenance support organization; q is the annual consumption of maintenance equipment. W is the number of equipment equipped by the troops; L is the number of equipment components of the unit equipment.

2. A demand forecasting method based on equipment maintenance material consumption rules according to claim 1, characterized in that: Before constructing the second equipment consumption law model, the maintenance equipment consumption probability of the equipment in normal state and defective state at time T is established respectively; The probability of maintenance equipment consumption when the equipment is in normal working condition is When the equipment is in a defective working state and the remaining life is greater than the equipment life threshold, the probability of maintenance equipment consumption is When the equipment is in a defective working state and the remaining life is less than the equipment life threshold, the probability of maintenance equipment consumption is Combining the maintenance equipment consumption probability of the equipment in normal working state and defective working state, the second equipment consumption law model is constructed, that is,

Images