Spare part demand prediction method and device, storage medium and electronic equipment

Abstract

The invention discloses a spare part demand prediction method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining the working time length of any part in operation, predicting the current fault rate of the part based on the working time length, carrying out the iterative prediction based on the current fault rate of the part, and obtaining a spare part demand prediction result. Determining the expected working duration and the expected failure rate of the part at the prediction moment; determining discrete distribution data of the type of parts based on the expected failure rate of each part of any type at the prediction moment; and on the basis of spare part service level data and the discrete distribution data of the type of parts, determining demand inventory data under the condition that the type of parts meet the spare part service level data. Under the condition of ensuring that the service level data of the spare parts are met, the problems of waste caused by excessive spare parts and too high management and maintenance cost of the spare parts are avoided, and the accuracy of the spare parts is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent industrial technology, and in particular to a spare parts demand forecasting method, device, storage medium and electronic equipment. Background Art

[0002] In the field of engineering machinery, spare parts inventory management is an important research area. Spare parts inventory is different from commodity inventory. Its demand comes from the current usage of equipment parts. For important equipment, managers often need to estimate the current status of its parts, so as to manage spare parts inventory more scientifically and make better decisions on its replacement.

[0003] In the process of implementing the present invention, it was found that there are at least the following technical problems in the prior art: too low spare parts inventory level is likely to lead to a lower service level, resulting in out-of-stock costs caused by long downtime when parts fail and need to be replaced; too high spare parts inventory will also bring higher inventory management costs, and managing and maintaining a high level of spare parts inventory requires a lot of manpower and maintenance costs. Summary of the invention

[0004] The present invention provides a spare parts demand forecasting method, device, storage medium and electronic device to improve the accuracy of spare parts demand forecasting and avoid the problem of excessive spare parts process management and maintenance success while ensuring that the demand is met.

[0005] According to one aspect of the present invention, a spare parts demand forecasting method is provided, comprising:

[0006] For any running part, obtain the working time of the part, predict the current failure rate of the part based on the working time, perform iterative prediction based on the current failure rate of the part, and determine the expected working time and expected failure rate of the part at the prediction time;

[0007] Determining discrete distribution data for parts of any type based on the expected failure rate of each of the parts at the prediction time;

[0008] Based on the spare parts service level data and the discrete distribution data of the type of parts, demand inventory data when the type of parts meets the spare parts service level data is determined.

[0009] Optionally, predicting the current failure rate of the part based on the working time includes:

[0010] Inputting the working time of the part into a pre-created part life prediction function to obtain the predicted life data of the part;

[0011] The current failure rate of the part under the current working time is determined based on the predicted life data of the part.

[0012] Optionally, the method for creating the part life prediction function includes:

[0013] Creating an initial prediction function, wherein the initial prediction function includes a plurality of undetermined parameters;

[0014] The multiple pending parameters are iteratively updated, and in any iterative process, the multiple pending parameters are updated alternately until an update condition is met, and a part life prediction function is obtained based on the parameter distribution after the update of each pending parameter.

[0015] Optionally, in any iterative process, the plurality of pending parameters are updated alternately, including:

[0016] For the current update parameter, extracting candidate data values ​​in the first distribution function;

[0017] determining a probability of receipt of the candidate data value based on the current data value;

[0018] A decision threshold is extracted from the second distribution function, and whether the current update parameter is updated to the candidate data value is determined based on the decision threshold and the reception probability.

[0019] Optionally, the iterative prediction based on the current failure rate of the part to determine the expected working time and expected failure rate of the part at the prediction time includes:

[0020] At any time t, based on the working time of the part at the time t and the predicted working time after the time t, and the failure rate of the part at the time t, the expected working time at the time t+1 is determined;

[0021] The expected failure rate at the time t+1 is determined based on the expected working time of the part at the time t+1.

[0022] Optionally, the predicted time includes an ordering time point and an arrival time point, wherein the ordering time point is determined based on the ordering lead time and the procurement cycle duration; the arrival time point is determined based on the ordering lead time, the procurement cycle duration and the stocking time.

[0023] Optionally, the discrete distribution data of the type of parts include probability data of the type of parts under different demand quantities;

[0024] The determining, based on the spare parts service level data and the discrete distribution data of the type of parts, the demand inventory data when the type of parts meets the spare parts service level data comprises:

[0025] Based on the probability data of the type of parts under different demand quantities, determine the probability and function under different demand quantities;

[0026] Determine a minimum demand quantity when the probability and function is greater than or equal to the spare parts service level data, and determine the minimum demand quantity as the demand inventory data.

[0027] According to another aspect of the present invention, there is provided a spare parts demand forecasting device, comprising:

[0028] A failure rate prediction module is used to obtain the working time of any running part and predict the current failure rate of the part based on the working time;

[0029] An expected failure rate prediction module, used to perform iterative prediction based on the current failure rate of the part to determine the expected working time and expected failure rate of the part at the prediction time;

[0030] A discrete distribution data determination module, for determining discrete distribution data of parts of any type based on the expected failure rate of each of the parts at the prediction moment;

[0031] The demand inventory data determination module is used to determine the demand inventory data when the type of parts meets the spare parts service level data based on the spare parts service level data and the discrete distribution data of the type of parts.

[0032] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0033] at least one processor; and

[0034] a memory communicatively connected to the at least one processor; wherein,

[0035] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the spare parts demand forecasting method described in any embodiment of the present invention.

[0036] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the spare parts demand forecasting method described in any embodiment of the present invention when executed.

[0037] The technical solution of the embodiment of the present invention can obtain the expected failure rate of the parts at the predicted time in the future by predicting the failure rate of the parts in operation, further reflecting the spare parts demand of the parts at the predicted time, and obtain the minimum spare parts demand data that meets the spare parts service level data by performing data analysis, that is, the demand inventory data. While ensuring that the spare parts service level data is met, the waste caused by too many spare parts and the problem of excessively high spare parts management and maintenance costs are avoided, thereby improving the accuracy of spare parts.

[0038] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 is a flow chart of a spare parts demand forecasting method provided by an embodiment of the present invention;

[0041] Figure 2 This is a flowchart of a spare parts demand forecasting method provided by an embodiment of the present invention.

[0042] Figure 3 It is a schematic diagram of the process of modeling a parts life prediction function provided by an embodiment of the present invention;

[0043] Figure 4 It is a structural schematic diagram of a spare parts demand prediction device provided by an embodiment of the present invention;

[0044] Figure 5 It is a schematic diagram of the structure of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0047] The parts demand forecasting and inventory decision-making methods for construction machinery are as follows:

[0048] 1. Empirical forecasting method: the operators and managers of each production line often make forecasts on the current status of each part based on their long-term experience, and often make manual adjustments based on simple averages and historical information of the same period to obtain the final forecast data and make inventory plans based on it. This method relies heavily on manual experience and has high labor costs. When faced with complex data, it often uses a relatively extensive forecasting method, which cannot capture relatively complex features and has a low accuracy rate.

[0049] 2. Time series forecasting method: Based on the historical data of the order quantity of each part, the demand is constructed into a time series, and the statistical time series model is used to model and forecast it. Common methods include: simple exponential smoothing method, ETS model, ARIMA model, etc. After obtaining a point estimate of future demand, the inventory replenishment point and replenishment quantity can be determined based on the demand situation during the lead time and inventory counting cycle.

[0050] Traditional time series models are based on simple linear prediction structures such as autoregression and moving average, which only use historical demand as input. However, the demand for spare parts is actually known to supply chain managers. If the demand for spare parts is modeled using a demand-based time series model, the factors that bring about the demand will be largely lost, and the demand for spare parts cannot be linked to the expected life and failure rate of the current limited number of equipment parts. At the same time, such demand often shows strong discontinuity, and general time series models are often unable to handle such characteristics. In addition, without the introduction of other data, the prediction results often show a large bias.

[0051] 3. Parts status monitoring and modeling. For large equipment in construction machinery, sensors are often used to monitor the current status of parts in multiple dimensions, such as temperature, humidity, vibration, delay, etc. These data will be transmitted to the server in real time for data analysis to determine the current status of the parts and determine the probability of damage based on the modeling results of its degradation process. This probability can be used to derive the demand for spare parts and establish an inventory model accordingly.

[0052] When monitoring the status of parts, it is often necessary to deploy more sensors to collect multi-dimensional data, and this data often needs to be transmitted and calculated in real time. When faced with fine-grained SKU-level predictions of parts, multi-dimensional data input is often not enough, and it is difficult to deploy multiple sensors in a general production environment. At the same time, multi-SKU and multi-dimensional data also cause great difficulties for data modeling. The model often needs to process data of different types and frequencies, which puts a lot of computing pressure on it. In addition, it is found that the prediction effect is not ideal in the actual data set.

[0053] In view of the above technical problems, an embodiment of the present invention provides a method for predicting spare parts demand. Figure 1 As shown, Figure 1 This is a flowchart of a spare parts demand prediction method provided by an embodiment of the present invention. This embodiment can be applied to the case where the spare parts demand data is determined by predicting the usage status of the parts. The method can be executed by a spare parts demand prediction device, which can be implemented in the form of hardware and / or software. The spare parts demand prediction device can be configured in electronic devices such as computers, mobile phones, servers, etc. Figure 1 As shown, the method includes:

[0054] S110. For any running part, obtain the working time of the part, predict the current failure rate of the part based on the working time, perform iterative prediction based on the current failure rate of the part, and determine the expected working time and expected failure rate of the part at the prediction time.

[0055] S120. Determine discrete distribution data of the parts of any type based on the expected failure rate of each of the parts of the type at the prediction moment.

[0056] S130. Determine, based on the spare parts service level data and the discrete distribution data of the type of parts, demand inventory data when the type of parts meets the spare parts service level data.

[0057] In this embodiment, a parts work list is maintained in advance for the parts in operation on the equipment, and the parts work list includes but is not limited to the part number of each part in operation, the configured equipment number, the replacement date of the part, and the used time of the part. The corresponding working time, i.e., the used time, can be read from the above parts work list according to the part number.

[0058] Optionally, the spare parts demand is predicted based on a preset time interval, and accordingly, the working hours of each part can be read from the parts work list according to the preset time interval to predict the current failure rate of each part. Exemplarily, the time period for prediction of different devices can be different, and accordingly, the corresponding time period can be determined according to the device type or device model, and the working hours of each part on the corresponding device can be read based on the time period.

[0059] For example, among the running parts, the set of part types can be marked as Φ = {Φ1, Φ2 ..., Φ m}, where the part type can be distinguished by the part model. The number of parts of each type can be K = {k1, k2…, k m}, the working time of the part can be marked as

[0060] For any part, determine the current failure rate of the part, where the failure rate of the part is used to characterize the demand rate of the part, that is, when a part fails, it needs to be replaced and there is a demand for spare parts. The failure rate of a part is related to the working time of the part, where the longer the working time, the greater the failure rate of the part.

[0061] In some embodiments, a failure rate calculation formula can be pre-set, and the working time of the part is input into the above failure rate calculation formula to obtain the current failure rate.

[0062] In some embodiments, predicting the current failure rate of the part based on the working time includes: inputting the working time of the part into a pre-created part life prediction function to obtain predicted life data of the part; and determining the current failure rate of the part at the current working time based on the predicted life data of the part.

[0063] A part life prediction function is created in advance to predict the life of the part through the part life prediction function, and the part life prediction function is further converted into a part failure rate prediction function to achieve the failure rate prediction of the part, so as to further determine whether there is a spare part demand for the part. Among them, the current failure rate of the part can be the ratio of the life data of the current working time to the life data greater than the current working time, wherein the life data of the current working time can be determined based on the life prediction data of the current working time, that is, The life data greater than the current working time can be determined based on the life prediction data corresponding to the working time from the current working time to infinity, that is, Among them, f(τ; θ) is the predicted life data of the part, where τ is the working time of the part, θ is the parameter in the part life prediction function, and f() is the part life prediction function.

[0064] The failure rate data reflects the probability of spare parts demand occurring for a single running component. In actual usage scenarios, the failure rate data needs to be aggregated and converted into overall spare parts demand. According to general demand planning scenarios, this demand value should be a point estimate, that is, the expected value aggregated from the demand probability.

[0065] Accordingly, for any type of parts, the spare parts demand data d of the parts can be determined by the current failure rates of multiple parts of the same type, for example: d i Spare parts demand data for multiple parts of type i, is the failure rate data of part j, and the working time of part j is

[0066] Through the above-mentioned prediction method, a demand sequence for each type of parts can be formed, and the demand sequence includes spare parts demand data corresponding to multiple prediction time points. Among them, the prediction time point can be a specific time point of each day, such as zero o'clock every day. The prediction time point can be determined according to the prediction time interval, which is not limited here. Exemplarily, the corresponding spare parts demand data is formed into a demand sequence according to the time sequence of the prediction time points. The demand sequence of each type of parts can be used as a reference for purchasing personnel to accurately purchase spare parts.

[0067] Since the spare parts procurement process takes a certain amount of time, for example, the time required for the spare parts procurement process includes the order lead time, the procurement cycle time, and the stocking time, it is necessary to predict the spare parts demand data in the future in advance to avoid the situation where the procurement process consumes time and cannot accurately select spare parts.

[0068] Specifically, for the prediction time, the prediction time can be one or more of the order time point and the arrival time point, wherein the order time point is determined based on the order lead time and the procurement cycle time; the arrival time point is determined based on the order lead time, the procurement cycle time and the preparation time. In this embodiment, the RSS (reorder point safety stock) strategy is adopted to manage the spare parts inventory, specifically: inventory management is performed through an interval inspection mechanism, if the inventory is higher than s, no measures are taken, and when the inventory is lower than s, an order is issued to restore the nominal inventory to the inventory S. Correspondingly, for the case where the prediction time is the order time point, the prediction time is determined based on the sum of the order lead time and the procurement cycle time, and for the case where the prediction time is the arrival time point, the prediction time is determined based on the sum of the order lead time, the procurement cycle time and the preparation time.

[0069] The spare parts demand data at the prediction time is determined based on the expected failure rate of each part at the prediction time. Accordingly, the expected failure rate of each part at the prediction time is determined, and further, the spare parts demand data of each type of part at the prediction time is determined. In this embodiment, for any part, the expected failure rate at each future moment is recursively determined based on the current failure rate, wherein the future moment may include the prediction moment.

[0070] Optionally, an iterative prediction is performed based on the current failure rate of the part to determine the expected working time and the expected failure rate of the part at the prediction time, including: at any time t, based on the working time of the part at time t and the predicted working time after time t, and the failure rate of the part at time t, determining the expected working time at time t+1; determining the expected failure rate at time t+1 based on the expected working time of the part at time t+1.

[0071] At any time t of the iterative prediction, an iterative prediction is performed based on the working time at time t and the failure rate at time t to determine the expected working time at time t+1, and the failure rate is predicted based on the expected working time at time t+1 to determine the expected failure rate at time t+1. This time t can be each prediction time point, for example, it can be the number of prediction days.

[0072] Specifically, the expected working time at time t+1 is determined based on the working time at time t and the probability of no failure rate within the predicted working time after time t, as well as the probability of failure occurring during the predicted working time after time t. Specifically, the sum of the working time at time t τ(t) and the predicted working time u(t) after time t is determined as the first time. is the failure rate of the part at time t, and accordingly, the probability of no failure rate occurring within the working time at time t and the predicted working time after time t is the first time length and The probability of failure after the predicted working time u(t) can be the product of the predicted working time u(t) and the failure rate Correspondingly, the sum of the above two products can be used as the expected working time τ(t+1) at time t+1.

[0073] Optionally, the predicted working time after time t can be obtained by counting the service life of each part. For example, the predicted working time of the part after time t can be obtained by counting the difference between the service life of each type of parts and the working time of the part at time t. Optionally, the predicted working time after time t can be obtained by a prediction model, that is, the type of part and the working time up to time t are input into the above prediction model to obtain the predicted working time after time t. The prediction model can be a machine learning model, such as a neural network model, etc. Optionally, the predicted working time after time t can be the time between time t and time t+1.

[0074] It should be noted that, in the case of replacement of a part, the working time of the part becomes 0.

[0075] At time t+1, the expected failure rate at time t+1 based on the expected working time at time t+1 can be determined by the following formula: ij (t+1) = FR(τ ij (t+1)), that is, the current failure rate of the part can be the expected working time τ at time t+1 ij (t+1) life data (i.e. With life span data greater than the current working time (i.e. The ratio of ij (t+1) is the expected failure rate of part j of type i at time t+1.

[0076] It is understandable that for any part of any type, the failure rate at any time can be predicted by the above method. Specifically, for any part of any type, the expected working time and expected failure rate at the next moment can be predicted by the working time and current failure rate at the current moment, and the next round of iterative prediction is performed by the expected working time and expected failure rate at the next moment until the expected working time and expected failure rate at the prediction moment are obtained.

[0077] The expected failure rate at the prediction time can be used to characterize the demand state of parts at the prediction time, where the expected failure rate of parts is the demand data of parts. For any type of parts, the discrete distribution data can be determined through the demand data of each part, that is, the probability data of this type of parts under different demand quantities.

[0078] For example, the probability that the demand quantity is zero in the time period before the prediction time can be: In the time period before the forecast moment, the probability that the demand quantity is 1 can be: Similarly, in the period before the forecast time, the demand quantity is k i The probability can be:

[0079] Obtain spare parts service level data, that is, the level at which the spare parts warehouse can meet spare parts demand. For example, the spare parts service level data can be 80%, indicating that the spare parts warehouse can meet 80% of the replacement parts demand. The spare parts service level data is set according to the spare parts warehouse demand. Different types of parts may correspond to different spare parts service level data. For example, different spare parts service level data may be set according to the importance level of the parts, and the spare parts service level data is positively correlated with the importance level of the parts.

[0080] Optionally, based on the spare parts service level data and the discrete distribution data of the type of parts, the demand inventory data of the type of parts is determined when the spare parts service level data is met, including: based on the probability data of the type of parts under different demand quantities, determining the probabilities and functions under different demand quantities; determining the minimum demand quantity when the probability and function is greater than or equal to the spare parts service level data, and determining the minimum demand quantity as the demand inventory data.

[0081] For example, the demand inventory data can be obtained by analyzing the following formula: in, is the probability and function under different demand quantities, and k is the spare parts service level data.

[0082] The minimum demand quantity that satisfies the above formula is determined as the demand inventory data, and the demand inventory data that meets the spare parts service level data is obtained. While ensuring that the spare parts service level data is met, the waste caused by too many spare parts and the problem of excessively high costs for spare parts management and maintenance are avoided, thereby improving spare parts accuracy.

[0083] The technical solution of this embodiment can obtain the expected failure rate of the parts at the predicted time in the future by predicting the failure rate of the parts in operation, and further reflect the spare parts demand of the parts at the predicted time. By performing data analysis, the minimum spare parts demand data that meets the spare parts service level data, that is, the demand inventory data, is obtained. While ensuring that the spare parts service level data is met, the waste caused by too many spare parts and the problem of excessively high spare parts management and maintenance costs are avoided, thereby improving the accuracy of spare parts.

[0084] On the basis of the above embodiment, the part life prediction function is created in advance, for example, it can be updated regularly according to demand, and the method for creating the part life prediction function includes: creating an initial prediction function, the initial prediction function includes multiple undetermined parameters; iteratively updating the multiple undetermined parameters, and in any iterative process, the multiple undetermined parameters are updated alternately until the update condition is met, and the part life prediction function is obtained based on the parameter distribution after the update of each of the undetermined parameters. Among them, the part life prediction function can be a Gamma function or a Weibull function.

[0085] The pending parameters in the initial prediction function need to undergo multiple rounds of iterative updates, and in each round of iterative updates, each pending parameter can be updated alternately, that is, the pending parameters can be updated in sequence in a specific order, and in the process of updating any pending parameter, the parameter values ​​of other pending parameters are kept unchanged until multiple pending parameters have completed a round of iterations, and then the next round of pending parameters are updated.

[0086] Optionally, in any iterative processing, multiple pending parameters are updated alternately, including: for the current update parameter, extracting a candidate data value in the first distribution function; determining the reception probability of the candidate data value based on the current data value; extracting a decision threshold in the second distribution function, and determining whether the current update parameter is updated to the candidate data value based on the decision threshold and the reception probability.

[0087] For the Weibull distribution, θ * =(k * ,λ * ). Accordingly, k and λ can be the undetermined parameters in the Weibull distribution, any of the undetermined parameters can be the current update parameter, the undetermined parameters other than the current update parameter are kept unchanged, and the parameter value of the current update parameter is updated.

[0088] Assuming j is the current sampling round number, then θ i j is the value of the current candidate point set, then the relationship for each update is: * =θ1 j +sZ,θ2 * =θ2j , Z~N(0,1), s is a constant. f(θ * |θ j ) is the proposed density function, used to j-1 Generate θ * , f(·) is the mean value of θ i j , a normal distribution with standard deviation s, which represents θ * The sampling target. θ1 and θ2 are two unknown parameters in the initial prediction function, θ1 * and θ2 * Can be a candidate data value.

[0089] Based on the preset reception probability calculation formula, the reception probability of the candidate data value is calculated, specifically: The data may be the historical replacement data of the parts collected, and the historical replacement data of the parts may include the equipment number, part number, replacement extraction and used time of the parts. In some embodiments, the data may be the used time of each part.

[0090] is the ratio of the posterior distribution of the candidate data value to the current data value, so that the parameter moves to a value with a higher posterior probability; is the ratio of the proposed density function of the candidate data value to the current data value, which moves the parameter to a larger value of the proposed density function. Among them, the proposed density function is symmetric, that is, f(θ j-1 |θ * )=f(θ * |θ j-1 ), corresponding Can be ignored.

[0091] The decision threshold is used to determine whether to accept the candidate data value. The decision threshold can be a random number, and the decision threshold satisfies the second distribution function, wherein the second distribution function can be a uniform distribution function. The decision threshold is compared with the receiving probability. If the decision threshold u≤receiving probability r, the candidate data value is accepted, that is, θ j =θ * ; If the decision threshold u> acceptance probability r, the candidate data value is rejected, that is, θ j =θ j-1 It should be noted that the initial value of each pending parameter can be preset before updating.

[0092] Repeat the above three steps for each element of θ. According to the basic assumptions of the Markov chain, the process must first be transformed into a stable state. After the above sampling is performed a sufficient number of times, the parameter sequence generated by the algorithm can be considered as a sample from the posterior distribution. The algorithm before this transitions the parameter values ​​from the proposed density function to the values ​​of the posterior distribution and forms a cluster. After the transition to the stable process, the same operation is continued for the parameters to be determined. At this time, the data can be directly used as data for parameter estimation, and the obtained data points can be used as parameter distribution estimates of the parameters to be determined. After the parameter estimates are obtained, the estimated values ​​will be used in combination with the original probability distribution model to obtain the final part life prediction function.

[0093] Based on the above embodiments, see Figure 2 , Figure 2 This is a flow chart of a spare parts demand prediction method provided by an embodiment of the present invention. The historical replacement data of the parts is collected in advance, and the parameters of the created initial prediction function are updated based on the historical replacement data, that is, the part life prediction function is modeled to obtain the part life prediction function. Figure 3 , Figure 3 It is a schematic diagram of the process of modeling the part life prediction function provided by an embodiment of the present invention. Figure 3 In the example of Weibull function, the undetermined parameters k and λ in the Weibull function are iteratively updated. The undetermined parameters are initialized and the initial values ​​of the undetermined parameters are determined. In any iteration process, the undetermined parameters k and λ are updated alternately. Specifically, in the normal distribution (k (j-1) ,s1 2 ) to extract candidate data value k * , calculate the reception probability r, which is:

[0094]

[0095] The decision threshold u is extracted from the uniform distribution function (0,1). If the decision threshold u> acceptance probability r, the candidate data value is rejected, that is, k j =k j-1 If the decision threshold u≤acceptance probability r, the candidate data value is accepted, that is, k j =k * .

[0096] In the normal distribution (λ (j-1) ,s1 2 ) to extract candidate data value λ * , calculate the reception probability r, which is:

[0097]

[0098] The decision threshold u is extracted from the uniform distribution function (0,1). If the decision threshold u> acceptance probability r, the candidate data value is rejected, that is, λ j =λ j-1 If the decision threshold u≤ acceptance probability r, the candidate data value is accepted, that is, λ j =λ * .

[0099] The next round of pending parameters is updated until the iteration end condition is met, the parameter distribution of each pending parameter is obtained, and the part life prediction function is formed based on the parameter distribution of each pending parameter.

[0100] Figure 4 Schematic diagram of the structure of a spare parts demand forecasting device provided by an embodiment of the present invention. Figure 4 As shown, the device comprises:

[0101] The failure rate prediction module 210 is used to obtain the working time of any running part and predict the current failure rate of the part based on the working time;

[0102] An expected failure rate prediction module 220 is used to perform iterative prediction based on the current failure rate of the component to determine the expected working time and expected failure rate of the component at the prediction time;

[0103] A discrete distribution data determination module 230, for determining discrete distribution data of parts of any type based on the expected failure rate of each of the parts at the prediction time;

[0104] The demand inventory data determination module 240 is used to determine the demand inventory data when the type of parts meets the spare parts service level data based on the spare parts service level data and the discrete distribution data of the type of parts.

[0105] Based on the above embodiment, optionally, the failure rate prediction module 210 is used to:

[0106] Inputting the working time of the part into a pre-created part life prediction function to obtain the predicted life data of the part;

[0107] The current failure rate of the part under the current working time is determined based on the predicted life data of the part.

[0108] Based on the above embodiment, optionally, the device further includes:

[0109] A function creation module, used to create an initial prediction function, wherein the initial prediction function includes a plurality of parameters to be determined;

[0110] The parameter updating module is used to iteratively update the multiple pending parameters, and in any iterative process, the multiple pending parameters are updated alternately until the updating condition is met, and the part life prediction function is obtained based on the parameter distribution after the update of each pending parameter.

[0111] Optionally, the parameter update module is used to:

[0112] For the current update parameter, extracting candidate data values ​​in the first distribution function;

[0113] determining a probability of receipt of the candidate data value based on the current data value;

[0114] A decision threshold is extracted from the second distribution function, and whether the current update parameter is updated to the candidate data value is determined based on the decision threshold and the reception probability.

[0115] Based on the above embodiment, optionally, the expected failure rate prediction module 220 is used to:

[0116] At any time t, based on the working time of the part at the time t and the predicted working time after the time t, and the failure rate of the part at the time t, the expected working time at the time t+1 is determined;

[0117] The expected failure rate at the time t+1 is determined based on the expected working time of the part at the time t+1.

[0118] Optionally, the predicted time includes an ordering time point and an arrival time point, wherein the ordering time point is determined based on the ordering lead time and the procurement cycle duration; the arrival time point is determined based on the ordering lead time, the procurement cycle duration and the stocking time.

[0119] Based on the above embodiment, optionally, the discrete distribution data of the type of parts includes probability data of the type of parts under different demand quantities;

[0120] The demand inventory data determination module 240 is used to:

[0121] Based on the probability data of the type of parts under different demand quantities, determine the probability and function under different demand quantities;

[0122] Determine a minimum demand quantity when the probability and function is greater than or equal to the spare parts service level data, and determine the minimum demand quantity as the demand inventory data.

[0123] The spare parts demand prediction device provided in the embodiment of the present invention can execute the spare parts demand prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0124] Figure 5 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0125] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0126] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0127] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a spare parts demand forecasting method.

[0128] In some embodiments, the spare parts demand forecasting method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the spare parts demand forecasting method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the spare parts demand forecasting method in any other appropriate manner (e.g., by means of firmware).

[0129] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0130] The computer program for implementing the spare parts demand forecasting method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed completely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or completely on a remote machine or server.

[0131] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a spare parts demand forecasting method, the method comprising:

[0132] For any running part, obtain the working time of the part, predict the current failure rate of the part based on the working time, perform iterative prediction based on the current failure rate of the part, and determine the expected working time and expected failure rate of the part at the prediction time;

[0133] Determining discrete distribution data for parts of any type based on the expected failure rate of each of the parts at the prediction time;

[0134] Based on the spare parts service level data and the discrete distribution data of the type of parts, demand inventory data when the type of parts meets the spare parts service level data is determined.

[0135] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0136] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0137] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0138] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0139] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0140] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A spare parts demand forecasting method, characterized in that: include: For any running part, obtain the working time of the part, predict the current failure rate of the part based on the working time, perform iterative prediction based on the current failure rate of the part, and determine the expected working time and expected failure rate of the part at the prediction time; Determining discrete distribution data for parts of any type based on the expected failure rate of each of the parts at the prediction time; Based on the spare parts service level data and the discrete distribution data of the type of parts, demand inventory data when the type of parts meets the spare parts service level data is determined.

2. The method according to claim 1, characterized in that The predicting the current failure rate of the part based on the working time includes: Inputting the working time of the part into a pre-created part life prediction function to obtain the predicted life data of the part; The current failure rate of the part under the current working time is determined based on the predicted life data of the part.

3. The method according to claim 2, characterized in that The method for creating the part life prediction function includes: Creating an initial prediction function, wherein the initial prediction function includes a plurality of undetermined parameters; The multiple pending parameters are iteratively updated, and in any iterative process, the multiple pending parameters are updated alternately until an update condition is met, and a part life prediction function is obtained based on the parameter distribution after the update of each pending parameter.

4. The method according to claim 3, characterized in that The multiple pending parameters are updated alternately in any iterative process, including: For the current update parameter, extracting candidate data values ​​in the first distribution function; determining a probability of receipt of the candidate data value based on the current data value; A decision threshold is extracted from the second distribution function, and whether the current update parameter is updated to the candidate data value is determined based on the decision threshold and the reception probability.

5. The method according to claim 1, characterized in that The iterative prediction based on the current failure rate of the part to determine the expected working time and expected failure rate of the part at the prediction time includes: At any time t, based on the working time of the part at the time t and the predicted working time after the time t, and the failure rate of the part at the time t, the expected working time at the time t+1 is determined; The expected failure rate at the time t+1 is determined based on the expected working time of the part at the time t+1.

6. The method according to claim 1 or 5, characterized in that: The predicted time includes an ordering time point and an arrival time point, wherein the ordering time point is determined based on the ordering lead time and the procurement cycle duration; the arrival time point is determined based on the ordering lead time, the procurement cycle duration and the stocking time.

7. The method according to claim 1, characterized in that Discrete distribution data of the type of parts, including probability data of the type of parts under different demand quantities; The determining, based on the spare parts service level data and the discrete distribution data of the type of parts, the demand inventory data when the type of parts meets the spare parts service level data comprises: Based on the probability data of the type of parts under different demand quantities, determine the probability and function under different demand quantities; Determine a minimum demand quantity when the probability and function is greater than or equal to the spare parts service level data, and determine the minimum demand quantity as the demand inventory data.

8. A spare parts demand forecasting device, characterized in that: include: A failure rate prediction module is used to obtain the working time of any running part and predict the current failure rate of the part based on the working time; An expected failure rate prediction module, used to perform iterative prediction based on the current failure rate of the part to determine the expected working time and expected failure rate of the part at the prediction time; A discrete distribution data determination module, for determining discrete distribution data of parts of any type based on the expected failure rate of each of the parts at the prediction moment; The demand inventory data determination module is used to determine the demand inventory data when the type of parts meets the spare parts service level data based on the spare parts service level data and the discrete distribution data of the type of parts.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the spare parts demand forecasting method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the spare parts demand forecasting method according to any one of claims 1 to 7 when executed.

Images