An inventory optimization method for analyzing the replacement cycle and failure rate of spare parts of a charging facility

Abstract

The application discloses a kind of inventory optimization methods of charging facility spare parts replacement cycle and failure rate analysis, and the core process of this method includes building data layer, multidimensional data is cleaned and optimized, to realize standardized storage;Based on the LSTM algorithm, a multidimensional linkage prediction model is built, the loss effect of design defects and environmental factors is quantified to optimize the model, and the accurate spare parts replacement cycle threshold and equipment failure rate are output;According to the flow properties of spare parts, a differentiated inventory management mode is designed, and the optimal management mode of slow-moving parts is selected;A full-cost quantification model is built to calculate the total cost and service level under each mode;Finally, with the goal of minimizing cost, an intelligent inventory control model is built, which dynamically outputs inventory warning thresholds, replenishment quantities and times;Through the algorithmization and parameter quantification of the whole process, the application realizes fine and intelligent management of inventory, significantly reduces operating costs and stockout risk, and improves the stability of charging facilities.

Description

[0001] This invention relates to the field of power facility operation and maintenance management and intelligent inventory optimization technology, and in particular to an inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the layout density and operational scale of charging facilities continue to expand. Spare parts are the core support for ensuring the continuous and stable operation of charging facilities. The scientific nature of their inventory management directly affects operation and maintenance efficiency and cost control. The existing spare parts inventory management of charging facilities relies heavily on experience reserves, which has problems such as inaccurate prediction of spare parts replacement cycle, insufficient analysis of the correlation between failure rate and inventory demand, weak ability to cope with extreme scenarios, and incomplete full cost quantification. This can easily lead to spare parts backlog or shortage, resulting in adverse consequences such as increased operation and maintenance costs and prolonged facility downtime.

[0003] Currently, some existing technologies attempt to predict spare parts demand using a single algorithm, but lack analysis of spare parts replacement cycles, failure rates, and multiple influencing factors. Furthermore, they fail to achieve dynamic matching between inventory strategies and overall costs, making it difficult to adapt to the diverse operating scenarios and complex environmental impacts of charging facilities. Therefore, there is an urgent need for an inventory optimization method that integrates multiple algorithms and comprehensive analysis to analyze the replacement cycles and failure rates of charging facility spare parts, enabling accurate analysis of spare parts replacement cycles and failure rates, as well as intelligent optimization of inventory strategies. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide an inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts, thereby solving the technical problems of low prediction accuracy, incomplete cost quantification, and insufficient response to extreme scenarios in the existing inventory management of charging facility spare parts.

[0005] An inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts includes the following steps:

[0006] S1. Data Layer Construction: Input the operating status of all charging facilities in the input area, spare parts wear and tear, existing spare parts inventory data, historical data on environmental impact, and replacement data of charging facilities to build a digital ledger system for charging facilities. Integrate the above data and clean and optimize it to achieve standardized storage of equipment information.

[0007] S2. Multi-dimensional linkage analysis model construction: Define five core indicators: spare parts replacement cycle, equipment failure rate, number of devices associated with spare parts, scope of failure impact, and extreme scenario risks, and construct a predictive model for spare parts replacement cycle and equipment failure rate.

[0008] S3. Model Optimization and Result Output: Quantify the impact of design defects, environmental factors, etc. on spare parts wear, optimize the prediction model, and output the prediction results of spare parts replacement cycle threshold and equipment failure rate.

[0009] S4. Construction of a distinctive inventory management module: Based on the spare parts replacement cycle and equipment failure rate prediction results output by S3, and combined with the fast and slow flow attributes of spare parts, a differentiated inventory management model is designed. For fast-flowing parts, a dynamic turnover inventory management model is adopted, and for slow-flowing parts, three management models are configured: joint inventory, spare parts pool sharing, and supplier emergency supply agreement.

[0010] S5. Optimal Inventory Management Model Selection: Based on spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk, a decision tree model is constructed, the weight of each parameter is quantified, the adaptation score of various management models is calculated, and the optimal decision is selected.

[0011] S6. Construction of a full cost quantification model: Input real-time supply chain quotations, industry implicit cost benchmarks, and scenario parameters to construct a full cost quantification model, and simulate and calculate the total cost distribution, service level achievement rate, and stockout risk probability under various management modes;

[0012] S7. Intelligent Inventory Management Model Construction: Based on the existing inventory quantity and the predicted demand for spare parts, with the goal of minimizing inventory costs, set an inventory early warning threshold and output the spare parts replenishment quantity and spare parts replenishment time.

[0013] 2. Further: The specific steps for constructing the data layer in step S1 are as follows:

[0014] S1.1: Comprehensively collect basic data on charging facilities in the target area, covering core information such as equipment, spare parts, inventory, environment and operating status;

[0015] S1.2: Subsequently, a digital ledger system was built based on a relational database to establish the association mapping between equipment, spare parts, environment, and inventory, realize the structured storage of data, and adopt a combination algorithm of outlier detection and min-max normalization to remove invalid outlier data, unify the data format and eliminate the influence of units, output standardized high-quality data, and finally store the cleaned and optimized data back into the ledger, which is dynamically updated every 24 hours.

[0016] Furthermore: the specific steps for constructing the multi-dimensional linkage analysis model in step S2 are as follows:

[0017] S2.1: Define five core indicators and clarify quantitative standards, covering spare parts replacement cycle, number of associated equipment, equipment operating status, degree of failure impact and extreme scenario risks, to achieve standardized quantification of multi-dimensional influencing factors;

[0018] S2.2: A prediction model is built using a Long Short-Term Memory Neural Network (LSTM). Leveraging its advantage in capturing long-term dependencies in time series data, it is adapted to the temporal characteristics of parts replacement and failure occurrence to improve prediction accuracy. The model is set with fixed structural parameters. The input layer corresponds to five core indicators, and the hidden layer and output layer are responsible for feature extraction and predicted value output, respectively. The ReLU activation function, Adam optimizer, and preset learning rate ensure the model's computational efficiency.

[0019] S2.3: By using two independent prediction formulas and substituting the corresponding initial weights, the spare parts replacement cycle and equipment failure rate are predicted synchronously. The initial weights are iteratively optimized through the training dataset to ensure that the prediction results fit the actual operating scenario.

[0020] Furthermore: the specific steps for model optimization and result output in step S3 are as follows:

[0021] S3.1: Design defects and environment are the two core factors affecting spare parts wear. Corresponding influence coefficients and quantitative standards are set for each. The degree of influence of design defects is reflected by combining the level quantitative value with the influence coefficient. Environmental factors are first weighted and summed for sub-factors such as temperature, humidity, dust concentration, and extreme weather to obtain a comprehensive quantitative value, and then combined with the environmental influence coefficient to quantify the effect.

[0022] S3.2: Construct a loss increment function to integrate the quantitative impact of the two types of factors into a unified loss increment. Substitute it into the original prediction result output by S2, and adjust the spare parts replacement cycle and equipment failure rate prediction values ​​through the multiplication correction formula to offset the interference of loss factors on prediction accuracy.

[0023] S3.3: Based on the optimized predicted value, a 10% redundancy is reserved to determine the spare parts replacement cycle threshold, and the optimized equipment failure rate is output simultaneously to form a core parameter that combines accuracy and safety.

[0024] 5. Further: The specific steps for constructing the featured inventory management module in step S4 are as follows:

[0025] S4.1: Based on monthly turnover rate as the core classification criterion, a threshold of 3 times / month is set to divide spare parts into fast-moving and slow-moving parts. The attributes are quantified through the turnover rate formula to provide a classification standard for mode adaptation.

[0026] S4.2: For fast-moving components, taking into account their high demand and high turnover characteristics, the Economic Order Quantity (EOQ) algorithm is used to construct a dynamic turnover inventory model. By integrating the annual demand forecast, single order cost and unit annual holding cost through the core formula, the optimal inventory quantity is accurately calculated. At the same time, the annual demand forecast is updated monthly to dynamically adjust the inventory size, maximize inventory turnover efficiency and reduce the risk of backlog.

[0027] S4.3: For slow-moving parts, considering their low demand and high backlog risk, three targeted management modes are configured: joint inventory management allocates shared inventory according to the number of sites through formulas to achieve inventory concentration; spare parts pool shared management sets the upper limit of call volume and cost rules to balance the flexibility and fairness of sharing; and supplier emergency supply agreement clarifies the minimum supply quantity standard and relies on response time to ensure emergency needs.

[0028] Furthermore: the specific steps for selecting the optimal inventory management mode in step S5 are as follows:

[0029] S5.1: Taking spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk as core decision nodes, construct a decision tree model and set judgment thresholds for each node, build a systematic pattern screening framework, and clarify the screening dimensions and judgment criteria.

[0030] S5.2: The Analytic Hierarchy Process (AHP) is used to quantify the weights of each decision parameter. An importance ratio judgment matrix is ​​constructed between parameters, and the initial weights are obtained by substituting them into the weight calculation formula. The weight allocation is then verified through a consistency check. If it is not qualified, the matrix is ​​adjusted until it meets the standard, ensuring that the weights are in line with the priorities of the actual scenario.

[0031] S5.3: Calculate the adaptation score of each management mode using a weighted summation formula, multiply the weight of each parameter by the adaptation score of the corresponding mode, and sum them to quantify the mode adaptation capability.

[0032] S5.4: Optimal models are selected by sorting according to their suitability scores, with the highest score being the first choice. In case of a tie, the lowest risk of stockout loss is used as a supplementary criterion to ensure that the selection results take into account both suitability and supply security, providing a definite management model sample for subsequent full-cost quantitative analysis.

[0033] Furthermore: the specific steps for constructing the full-cost quantification model in step S6 are as follows:

[0034] S6.1: Collect three types of core input parameters, including real-time supply chain quotations, industry implicit cost benchmarks and replenishment lead time, and stockout loss amount, to ensure that the parameters cover cost composition and scenario characteristics;

[0035] S6.2: Integrate four categories of costs: procurement, inventory holding, stockout losses, and hidden costs. Clarify the core formulas for each sub-item cost and the total cost. Clarify the meaning and calculation logic of each variable through parameter definition. The average inventory quantity is differentiated according to the characteristics of fast and slow flow items. The stockout probability is quantified by combining existing inventory and predicted demand. This achieves accurate decomposition and aggregation of the entire cost and builds a scientific full-cost quantification model.

[0036] S6.3: Based on the Monte Carlo simulation algorithm, the optimal mode selected by S5 and other alternative modes are substituted to perform multi-scenario simulation calculations on the total cost distribution, service level achievement rate and stockout risk probability under various modes, and output standardized result reports.

[0037] Furthermore: the specific steps for constructing the intelligent inventory management model in step S7 are as follows:

[0038] S7.1: Taking the total cost quantified in S6 as the core, set the objective function of minimizing the total inventory cost, integrate procurement, holding, stockout losses and hidden costs, and add inventory constraints to ensure that the replenishment quantity is not less than the difference between the predicted demand and the existing inventory.

[0039] S7.2: Based on the equipment failure rate and spare parts replacement cycle prediction value optimized by S3, and superimposed with a redundancy coefficient of 1.2, calculate the inventory warning threshold. When the existing inventory is lower than the threshold, a replenishment reminder is automatically triggered.

[0040] S7.3: The replenishment parameters are accurately derived through two core formulas. The replenishment quantity is calculated by combining the predicted demand and the replenishment lead time to ensure that the inventory demand is fully covered within the lead time. The replenishment time is calculated by working backward from the existing inventory consumption rate to reserve sufficient replenishment cycle to avoid stockouts.

[0041] S7.4: Establish a dynamic adjustment mechanism to update forecasted demand, existing inventory, and cost parameters every 7 days, and simultaneously iterate the early warning threshold, replenishment quantity, and timing.

[0042] Furthermore: In step S2, the spare parts replacement cycle and equipment failure rate prediction model adopts the Long Short-Term Memory (LSTM) neural network algorithm. The model input is historical data of five core indicators, and the output is the predicted value. The core function formula is as follows:

[0043] (1) Formula for predicting spare parts replacement cycle:

[0044]

[0045] in, This represents the predicted spare parts replacement cycle value at time t+k. The initial weights for the five core indicators, Let t be the actual value of the spare parts replacement cycle at time t. Let t be the actual equipment failure rate. Let be the number of spare parts associated with the device at time t. Let be the extent of the fault's impact at time t. Let be the extreme scenario risk value at time t;

[0046] (2) Equipment failure rate prediction formula:

[0047]

[0048] in This represents the predicted equipment failure rate at time t+k. The initial weights for the five core indicators are defined, with values ​​ranging from (0,1). , ;

[0049] The functional formula for the impact of quantitative design defects and environmental factors on spare parts wear in S3 is as follows:

[0050]

[0051] in, To account for the increase in spare parts wear and tear, The design defect impact coefficient has a value range of [0.1, 0.5]. To quantify the design defect level, The comprehensive impact coefficient of environmental factors ranges from [0.2, 0.6], and E is the comprehensive quantitative value of environmental factors, which is calculated by weighting sub-factors such as temperature, humidity, and dust concentration.

[0052]

[0053] in, , , , These are sub-factors such as temperature, humidity, and dust concentration;

[0054] The optimized prediction model formula is as follows:

[0055]

[0056]

[0057] in, , These are the optimized spare parts replacement cycle and the predicted equipment failure rate. .

[0058] Furthermore: The core formula of the fast-moving components dynamic turnover inventory management mode in S4 is:

[0059]

[0060] in, For dynamic turnover inventory batches, This is the projected annual demand for spare parts. Cost per order, Annual inventory holding cost per unit of spare parts;

[0061] The formula for shared inventory in joint inventory management of slow-moving components is:

[0062]

[0063] in, K represents the number of charging facility sites participating in the joint inventory, which is the shared inventory quantity.

[0064] The parameter weight quantification of the decision tree model in S5 adopts the Analytic Hierarchy Process (AHP), and the weight calculation function is:

[0065]

[0066] in, The weight of the i-th parameter, This represents the importance ratio of the i-th parameter relative to the k-th parameter. This represents the total number of parameters.

[0067] The formula for the fit score is:

[0068]

[0069] in, The adaptation score for the m-th management model is... Score the fit of the i-th parameter with the m-th management mode;

[0070] The total cost formula for the full-cost quantification model in S6 is:

[0071]

[0072] in, For spare parts procurement costs, For inventory holding costs, To avoid losses due to stockouts, Hidden costs;

[0073] The formulas for each item of cost are as follows:

[0074]

[0075] Where Q is the quantity purchased and P is the unit price of spare parts;

[0076]

[0077] in, This represents the average inventory quantity. Annual holding cost per unit of spare parts;

[0078]

[0079] in, This represents the probability of being out of stock. This refers to the amount of loss due to a single stockout.

[0080]

[0081] This is an implicit cost coefficient, with a value range of [0.05, 0.15], determined based on industry benchmarks.

[0082] The present invention has the following beneficial effects:

[0083] 1. Accurate and reliable data support lays the foundation for scientific decision-making: This invention constructs a digital ledger system for charging facilities, integrating multi-dimensional data such as equipment operation, spare parts wear and tear, and environmental impact. It employs a combination algorithm of 3σ criterion outlier detection and min-max normalization to complete data cleaning and optimization, achieving standardized data storage and 24-hour dynamic updates. This effectively eliminates invalid and interfering data and eliminates dimensional differences, providing high-quality, real-time data support for subsequent prediction models and inventory management. It significantly reduces inventory decision-making bias caused by data errors, ensuring the scientific and reliable nature of the entire process optimization.

[0084] 2. Precise and efficient predictive model, enhancing the foresight of inventory planning: Employing a Long Short-Term Memory (LSTM) neural network to construct a multi-dimensional, interconnected predictive model accurately captures the temporal dependency between spare parts replacement cycles and equipment failure rates. Simultaneously, it innovatively quantifies the impact of design flaws and environmental factors on losses, correcting the prediction results through a loss increment function, and outputting accurate thresholds with redundant reserves. Compared to traditional experience-based predictions, this model offers higher prediction accuracy, enabling early prediction of spare parts demand patterns and failure risks. This shifts inventory planning from reactive response to proactive prediction, avoiding blind stockpiling or insufficient stockpiling.

[0085] 3. Refined inventory management, balancing efficiency and cost: Based on spare parts turnover rate, fast and slow-moving parts are classified, and targeted differentiated management models are designed: For fast-moving parts, the EOQ algorithm is used to dynamically adjust the inventory batch size to maximize turnover efficiency. For slow-moving parts, three exclusive modes are configured, and the optimal solution is selected through AHP weight quantification and decision tree model. This breaks the limitations of traditional one-size-fits-all inventory management, reduces the capital occupation and backlog losses of fast-moving parts, and reduces the risk of idle inventory of slow-moving parts. It achieves precise adaptation and control of spare parts with different characteristics, and takes into account both inventory efficiency and cost optimization.

[0086] 4. Full-cost quantitative accounting for ultimate cost control: Construct a full-cost quantitative model covering procurement, inventory holding, stockout losses, and hidden costs. Combined with Monte Carlo simulation algorithms, accurately calculate the total cost distribution under various management models. Break through the limitations of traditional models that only focus on explicit costs, incorporate industry hidden costs into the quantitative system. At the same time, set constraints with the goal of minimizing total cost to ensure that inventory decisions achieve optimal cost while meeting demand, significantly reducing various cost wastes and losses in the operation process.

[0087] 5. Significantly reduced stockout risk, ensuring stable facility operation: By setting a 1.2 times redundant inventory warning threshold through optimized forecast data, and combining replenishment lead time and inventory consumption rate to accurately deduce replenishment time, a dual stockout prevention and control mechanism is constructed. This effectively shortens the response time to inventory gaps, ensures timely and sufficient replenishment, significantly reduces the risk of charging facility downtime due to spare parts shortages, improves service level achievement rate, ensures the continuous and stable operation of charging facilities, and enhances the user experience. Attached Figure Description

[0088] Figure 1 This is a flowchart of an inventory optimization method for analyzing the replacement cycle and failure rate of spare parts for charging facilities. Detailed Implementation

[0089] The present invention will be further described in detail below with reference to specific embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0090] Please see the appendix Figure 1 The present invention provides an embodiment 1: an inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts, comprising the following specific implementation steps:

[0091] S1. Data Layer Construction: Input the operating status of all charging facilities in the input area, spare parts wear and tear, existing spare parts inventory data, historical data on environmental impact, and replacement data of charging facilities to build a digital ledger system for charging facilities. Integrate the above data and clean and optimize it to achieve standardized storage of equipment information.

[0092] The specific steps for constructing the data layer in step S1 are as follows:

[0093] S1.1: Comprehensively collect basic data on charging facilities in the target area, covering core information such as equipment, spare parts, inventory, environment and operating status;

[0094] S1.2: Subsequently, a digital ledger system was built based on a relational database to establish the association mapping between equipment, spare parts, environment, and inventory, realize the structured storage of data, and use a combination algorithm of outlier detection and min-max normalization to remove invalid outlier data, unify the data format and eliminate the influence of units, output standardized high-quality data, and finally store the cleaned and optimized data back into the ledger and update it dynamically at a frequency of 24 hours each time.

[0095] The data cleaning and optimization algorithm in S1 employs a combination of outlier detection and data standardization. Outlier detection is achieved using the 3σ criterion, and data standardization uses the min-max normalization function, as detailed below:

[0096] (1) Outlier detection formula: For dataset Calculate the mean with standard deviation ,when When, determine x i Outliers were identified and removed.

[0097]

[0098]

[0099] Where X is a dataset of a single type of data (such as a dataset of the wear rate of a certain type of spare parts). Let n be the i-th data point in the dataset, and n be the sample size of the dataset. The mean of the dataset. The standard deviation of the dataset;

[0100] (2) min-max normalization formula:

[0101]

[0102] Where, x' i For standardized data, Let X be the minimum value of the dataset. For dataset The maximum value;

[0103] S2. Multi-dimensional linkage analysis model construction: Define five core indicators: spare parts replacement cycle, equipment failure rate, number of spare parts associated with equipment, scope of failure impact, and extreme scenario risks, and construct a predictive model for spare parts replacement cycle and equipment failure rate;

[0104] Step S2, the specific steps for constructing the multi-dimensional linkage analysis model, are as follows:

[0105] S2.1: Define five core indicators and clarify quantitative standards, covering spare parts replacement cycle, number of associated equipment, equipment operating status, degree of failure impact and extreme scenario risks, to achieve standardized quantification of multi-dimensional influencing factors;

[0106] S2.2: A prediction model is built using a Long Short-Term Memory Neural Network (LSTM). Leveraging its advantage in capturing long-term dependencies in time series data, it is adapted to the temporal characteristics of parts replacement and failure occurrence to improve prediction accuracy. The model is set with fixed structural parameters. The input layer corresponds to five core indicators, and the hidden layer and output layer are responsible for feature extraction and predicted value output, respectively. The ReLU activation function, Adam optimizer, and preset learning rate ensure the model's computational efficiency.

[0107] S2.3: By using two independent prediction formulas and substituting the corresponding initial weights, the spare parts replacement cycle and equipment failure rate are predicted simultaneously. The initial weights are iteratively optimized through the training dataset to ensure that the prediction results fit the actual operating scenario.

[0108] In step S2, the spare parts replacement cycle and equipment failure rate prediction model uses the Long Short-Term Memory (LSTM) neural network algorithm. The model input is historical data of five core indicators, and the output is the predicted value. The core function formula is as follows:

[0109] (1) Formula for predicting spare parts replacement cycle:

[0110]

[0111] in, This represents the predicted spare parts replacement cycle value at time t+k. The initial weights for the five core indicators, Let t be the actual value of the spare parts replacement cycle at time t. Let t be the actual equipment failure rate. Let be the number of spare parts associated with the device at time t. Let be the extent of the fault's impact at time t. Let be the extreme scenario risk value at time t;

[0112] (3) Equipment failure rate prediction formula:

[0113]

[0114] in, This represents the predicted equipment failure rate at time t+k. The initial weights for the five core indicators are defined, with values ​​ranging from (0,1). , ;

[0115] S3. Model Optimization and Result Output: Quantify the impact of design defects, environmental factors, etc. on spare parts wear, optimize the prediction model, and output the prediction results of spare parts replacement cycle threshold and equipment failure rate.

[0116] The specific steps for model optimization and result output in step S3 are as follows:

[0117] S3.1: Design defects and environment are the two core factors affecting spare parts wear. Corresponding influence coefficients and quantitative standards are set for each. The degree of influence of design defects is reflected by combining the level quantitative value with the influence coefficient. Environmental factors are first weighted and summed for sub-factors such as temperature, humidity, dust concentration, and extreme weather to obtain a comprehensive quantitative value, and then combined with the environmental influence coefficient to quantify the effect.

[0118] S3.2: Construct a loss increment function to integrate the quantitative impact of the two types of factors into a unified loss increment. Substitute it into the original prediction result output by S2, and adjust the spare parts replacement cycle and equipment failure rate prediction values ​​through the multiplication correction formula to offset the interference of loss factors on prediction accuracy.

[0119] S3.3: Based on the optimized predicted value, reserve 10% redundancy to determine the spare parts replacement cycle threshold, and output the optimized equipment failure rate simultaneously.

[0120] The functional formula for the impact of quantitative design defects and environmental factors on spare parts wear in S3 is as follows:

[0121]

[0122] in, To account for the increase in spare parts wear and tear, The design defect impact coefficient has a value range of [0.1, 0.5]. To quantify the design defect level, The comprehensive impact coefficient of environmental factors ranges from [0.2, 0.6], and E is the comprehensive quantitative value of environmental factors, which is calculated by weighting sub-factors such as temperature, humidity, and dust concentration.

[0123]

[0124] in, , , , These are sub-factors such as temperature, humidity, and dust concentration;

[0125] The optimized prediction model formula is as follows:

[0126]

[0127]

[0128] in, , These are the optimized spare parts replacement cycle and the predicted equipment failure rate. ;

[0129] S4. Construction of a distinctive inventory management module: Based on the spare parts replacement cycle and equipment failure rate prediction results output by S3, and combined with the fast and slow flow attributes of spare parts, a differentiated inventory management model is designed. For fast-flowing parts, a dynamic turnover inventory management model is adopted, and for slow-flowing parts, three management models are configured: joint inventory, spare parts pool sharing, and supplier emergency supply agreement.

[0130] The specific steps for building the featured inventory management module in step S4 are as follows:

[0131] S4.1: Based on monthly turnover rate as the core classification criterion, a threshold of 3 times / month is set to divide spare parts into fast-moving and slow-moving parts. The attributes are quantified through the turnover rate formula to provide a classification standard for mode adaptation.

[0132] S4.2: For fast-moving components, taking into account their high demand and high turnover characteristics, the Economic Order Quantity (EOQ) algorithm is used to construct a dynamic turnover inventory model. By integrating the annual demand forecast, single order cost and unit annual holding cost through the core formula, the optimal inventory quantity is accurately calculated. At the same time, the annual demand forecast is updated monthly to dynamically adjust the inventory size, maximize inventory turnover efficiency and reduce the risk of backlog.

[0133] S4.3: For slow-moving parts, considering their low demand and high backlog risk, three targeted management modes are configured: joint inventory management allocates shared inventory according to the number of sites through formula to achieve inventory intensification; spare parts pool shared management sets the upper limit of call volume and cost rules to balance the flexibility and fairness of sharing; and supplier emergency supply agreement clarifies the minimum supply quantity standard and relies on response time to ensure emergency needs.

[0134] The core formula of the S4 fast-moving components dynamic turnover inventory management model is:

[0135]

[0136] in, For dynamic turnover inventory batches, This is the projected annual demand for spare parts. Cost per order, Annual inventory holding cost per unit of spare parts;

[0137] The formula for shared inventory in joint inventory management of slow-moving components is:

[0138]

[0139] in, K represents the number of charging facility sites participating in the joint inventory, which is the shared inventory quantity.

[0140] S5. Optimal Inventory Management Model Selection: Based on spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk, a decision tree model is constructed, the weight of each parameter is quantified, the adaptation score of various management models is calculated, and the optimal decision is selected.

[0141] The specific steps for selecting the optimal inventory management model in step S5 are as follows:

[0142] S5.1: Taking spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk as core decision nodes, construct a decision tree model and set judgment thresholds for each node, build a systematic pattern screening framework, and clarify the screening dimensions and judgment criteria.

[0143] S5.2: The Analytic Hierarchy Process (AHP) is used to quantify the weights of each decision parameter. An importance ratio judgment matrix is ​​constructed between parameters, and the initial weights are obtained by substituting them into the weight calculation formula. The weight allocation is then verified through a consistency check. If it is not qualified, the matrix is ​​adjusted until it meets the standard, ensuring that the weights are in line with the priorities of the actual scenario.

[0144] S5.3: Calculate the adaptation score of each management mode using a weighted summation formula, multiply the weight of each parameter by the adaptation score of the corresponding mode, and sum them to quantify the mode adaptation capability.

[0145] S5.4: Filter the optimal model by sorting according to the fit score. The one with the highest score is the first choice. If there is a tie, the lowest risk of stockout loss is used as the supplementary criterion to ensure that the screening results take into account both fit and supply security, and provide a definite management model sample for subsequent full cost quantitative analysis.

[0146] In S5, the parameter weights of the decision tree model are quantified using the Analytic Hierarchy Process (AHP), and the weight calculation function is as follows:

[0147]

[0148] in, The weight of the i-th parameter, This represents the importance ratio of the i-th parameter relative to the k-th parameter. This represents the total number of parameters.

[0149] The formula for the fit score is:

[0150]

[0151] in, The adaptation score for the m-th management model is... Score the fit of the i-th parameter with the m-th management mode.

[0152] S6. Construction of a full cost quantification model: Input real-time supply chain quotations, industry implicit cost benchmarks, and scenario parameters to construct a full cost quantification model, and simulate and calculate the total cost distribution, service level achievement rate, and stockout risk probability under various management modes;

[0153] The specific steps for constructing the full-cost quantification model in step S6 are as follows:

[0154] S6.1: Collect three types of core input parameters, including real-time supply chain quotations, industry implicit cost benchmarks and replenishment lead time, and stockout loss amount, to ensure that the parameters cover cost composition and scenario characteristics;

[0155] S6.2: Integrate four categories of costs: procurement, inventory holding, stockout losses, and hidden costs. Clarify the core formulas for each sub-item cost and the total cost. Clarify the meaning and calculation logic of each variable through parameter definition. The average inventory quantity is differentiated according to the characteristics of fast and slow flow items. The stockout probability is quantified by combining existing inventory and predicted demand. This achieves accurate decomposition and aggregation of the entire cost and builds a scientific full-cost quantification model.

[0156] S6.3: Based on the Monte Carlo simulation algorithm, the optimal mode selected by S5 and other alternative modes are substituted into the algorithm to perform multi-scenario simulation calculations on the total cost distribution, service level achievement rate and stockout risk probability under various modes, and output standardized result reports.

[0157] The total cost formula for the full-cost quantification model in S6 is:

[0158]

[0159] in, For spare parts procurement costs, For inventory holding costs, To avoid losses due to stockouts, Hidden costs;

[0160] The formulas for each item of cost are as follows:

[0161]

[0162] Where Q is the quantity purchased and P is the unit price of spare parts;

[0163]

[0164] in, This represents the average inventory quantity. Annual holding cost per unit of spare parts;

[0165]

[0166] in, This represents the probability of being out of stock. This refers to the amount of loss due to a single stockout.

[0167]

[0168] This is an implicit cost coefficient, with a value range of [0.05, 0.15], determined based on industry benchmarks;

[0169] S7. Intelligent Inventory Management Model Construction: Based on the existing inventory quantity and the predicted demand for spare parts, with the goal of minimizing inventory costs, set an inventory early warning threshold and output the spare parts replenishment quantity and spare parts replenishment time.

[0170] The specific steps for building the intelligent inventory management model in step S7 are as follows:

[0171] S7.1: Taking the total cost quantified in S6 as the core, set the objective function of minimizing the total inventory cost, integrate procurement, holding, stockout losses and hidden costs, and add inventory constraints to ensure that the replenishment quantity is not less than the difference between the predicted demand and the existing inventory.

[0172] S7.2: Based on the equipment failure rate and spare parts replacement cycle prediction value optimized by S3, and superimposed with a redundancy coefficient of 1.2, calculate the inventory warning threshold. When the existing inventory is lower than the threshold, a replenishment reminder is automatically triggered.

[0173] S7.3: The replenishment parameters are accurately derived through two core formulas. The replenishment quantity is calculated by combining the predicted demand and the replenishment lead time to ensure that the inventory demand is fully covered within the lead time. The replenishment time is calculated by working backward from the existing inventory consumption rate to reserve sufficient replenishment cycle to avoid stockouts.

[0174] S7.4: Establish a dynamic adjustment mechanism to update forecasted demand, existing inventory, and cost parameters every 7 days, and simultaneously iterate the early warning threshold, replenishment quantity, and timing.

[0175] The objective function of the intelligent inventory management model in S7 is:

[0176]

[0177] The constraints are:

[0178] in, To predict demand, For the current inventory quantity ;

[0179] The formula for the inventory warning threshold is:

[0180] Among them, when the existing inventory quantity ≤ Replenishment is triggered at certain times;

[0181] Replenishment quantity: Q=

[0182] in, To allow for lead time for replenishment;

[0183] Restocking time:

[0184] Where N is the number of associated devices.

[0185] An embodiment 2 of this invention is provided: This embodiment takes a new energy charging operation company in a certain city as the application object. The company manages 20 charging stations and 500 charging devices of different models, involving 10 types of core spare parts such as charging modules, contactors, and fuses. The method of this invention is applied to optimize inventory. The specific implementation process is as follows:

[0186] In the initial implementation phase, the data layer was constructed according to the S1 step: collecting equipment operation data (average daily operation of a single device is 8 hours, average load rate is 60%), spare parts loss records (average monthly loss of 5 charging modules and 2 fuses), current inventory data (current inventory of 10 charging modules and 15 fuses), environmental data (average annual humidity of 75% in coastal areas, 2 extreme rainstorms per year, salt spray concentration of 0.05mg / m³), and spare parts replacement records. Abnormal data such as erroneous daily loss of 10 charging modules were eliminated using the 3σ criterion. After standardization processing, the data was stored in a MySQL relational database with a 24-hour automatic update mechanism to ensure data real-time performance.

[0187] Following step S2, an LSTM prediction model is constructed, selecting five core indicators and training with historical data from the past two years. The hidden layer has 32 neurons, employing the ReLU activation function and Adam optimizer with a learning rate of 0.001. The initial prediction is a charging module replacement cycle of 33 days and a failure rate of 0.018 times / day. In step S3, considering the charging module's design defect level of 2, a design defect impact coefficient of 0.2 is determined. Based on the characteristics of the coastal high-humidity salt spray environment, a comprehensive environmental impact coefficient of 0.4 is set. After correction, the output charging module replacement cycle threshold is 36 days, with a failure rate of 0.02 times / day, and a 10% redundancy is reserved to cope with environmental fluctuations.

[0188] S4 steps are categorized by turnover rate: charging modules have a monthly turnover rate of 4 times and are classified as fast-moving components, adopting a dynamic turnover inventory model; fuses have a monthly turnover rate of 1.5 times and are classified as slow-moving components, configured with three management models: joint inventory, spare parts pool sharing, and supplier emergency supply agreement.

[0189] Step S5 uses the analytic hierarchy process to determine the weights of spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk as 0.3, 0.2, 0.2, and 0.3, respectively. The calculated fit score for the spare parts pool sharing mode is 8.2, which is the optimal solution.

[0190] The S6 step focuses on full cost quantification, inputting real-time supply chain quotes (charging module unit price 800 yuan, fuse unit price 50 yuan), industry implicit cost benchmarks (implicit cost coefficient 0.1), and scenario parameters (replenishment lead time 3 days, single out-of-stock loss amount for charging modules 2000 yuan, single out-of-stock loss amount for fuses 800 yuan).

[0191] When constructing the full cost model, the procurement cost, inventory holding cost (annual unit holding cost of 80 yuan for charging modules and 6 yuan for fuses), stockout loss cost and hidden cost are integrated. Based on Monte Carlo simulation, the total cost under the spare parts pool sharing model is reduced by 18% compared with the traditional model, including a 12% reduction in procurement cost, a 25% reduction in inventory holding cost, a 15% reduction in hidden cost, and a 95% service level achievement rate.

[0192] Step S7, based on the total cost quantified in S6, sets a target function to minimize total inventory cost. It also adds inventory constraints to ensure that replenishment quantity is not less than the difference between predicted demand and existing inventory, balancing cost optimization and supply assurance. Based on the optimized charging module failure rate of 0.02 times / day and replacement cycle of 36 days (S3), and with a redundancy coefficient of 1.2, the calculated charging module inventory warning threshold is 8 units. When the existing inventory falls below this value, the system automatically triggers a replenishment alert, establishing the first line of defense against stockouts.

[0193] Then, the replenishment parameters were precisely derived: combining the 3-day replenishment lead time of the charging module and the daily failure rate of 0.02 times / day, and the requirement to adapt to 500 devices, the demand within the lead time was predicted to be 3 units, the current inventory was 10 units, and the replenishment quantity was calculated to be 3 units, to ensure that the inventory demand within the lead time is fully covered.

[0194] Based on the existing inventory consumption rate of 0.17 units per day, the replenishment initiation time is calculated to be when the inventory is depleted to 10 units, allowing sufficient replenishment cycle to avoid stockouts.

[0195] Establish a dynamic adjustment mechanism to update forecasted demand (e.g., increase forecasted demand for charging modules by 10% during the summer high-temperature period), existing inventory and cost parameters (e.g., the unit price of charging modules in the supply chain fluctuates to 820 yuan) every 7 days, and simultaneously iterate inventory warning thresholds, replenishment quantities and times.

[0196] For example, after a certain update, the warning threshold for charging modules was adjusted to 9 units, the replenishment quantity increased to 4 units, and the replenishment initiation time was advanced by 1 day. This enabled the model parameters to be adapted to the actual scenario in real time, forming an intelligent management and control closed loop of target setting, threshold warning, precise replenishment, and dynamic optimization.

[0197] After implementation, the company's spare parts inventory backlog decreased by 23%, the out-of-stock rate dropped from 12% to 3%, annual operating costs decreased by 150,000 yuan, and the stable operation rate of charging facilities increased to 98%.

[0198] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts, characterized in that, Includes the following steps: S1. Data Layer Construction: Input the operating status of all charging facilities in the input area, spare parts wear and tear, existing spare parts inventory data, historical data on environmental impact, and replacement data of charging facilities to build a digital ledger system for charging facilities. Integrate the above data and clean and optimize it to achieve standardized storage of equipment information. S2. Multi-dimensional linkage analysis model construction: Define five core indicators: spare parts replacement cycle, equipment failure rate, number of spare parts associated with equipment, scope of failure impact, and extreme scenario risks, and construct a predictive model for spare parts replacement cycle and equipment failure rate; S3. Model Optimization and Result Output: Quantify the impact of design defects, environmental factors, etc. on spare parts wear, optimize the prediction model, and output the prediction results of spare parts replacement cycle threshold and equipment failure rate. S4. Construction of a distinctive inventory management module: Based on the spare parts replacement cycle and equipment failure rate prediction results output by S3, and combined with the fast and slow flow attributes of spare parts, a differentiated inventory management model is designed. For fast-flowing parts, a dynamic turnover inventory management model is adopted, and for slow-flowing parts, three management models are configured: joint inventory, spare parts pool sharing, and supplier emergency supply agreement. S5. Optimal Inventory Management Model Selection: Based on spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk, a decision tree model is constructed, the weight of each parameter is quantified, the adaptation score of various management models is calculated, and the optimal decision is selected. S6. Construction of a full cost quantification model: Input real-time supply chain quotations, industry implicit cost benchmarks, and scenario parameters to construct a full cost quantification model, and simulate and calculate the total cost distribution, service level achievement rate, and stockout risk probability under various management modes; S7. Intelligent Inventory Management Model Construction: Based on the existing inventory quantity and the predicted demand for spare parts, with the goal of minimizing inventory costs, set an inventory early warning threshold and output the spare parts replenishment quantity and spare parts replenishment time.

2. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for constructing the data layer in step S1 are as follows: S1.1: Comprehensively collect basic data on charging facilities in the target area, covering core information such as equipment, spare parts, inventory, environment and operating status; S1.2: Subsequently, a digital ledger system was built based on a relational database to establish the association mapping between equipment, spare parts, environment, and inventory, realize the structured storage of data, and adopt a combination algorithm of outlier detection and min-max normalization to remove invalid outlier data, unify the data format and eliminate the influence of units, output standardized high-quality data, and finally store the cleaned and optimized data back into the ledger, which is dynamically updated every 24 hours.

3. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for constructing the multi-dimensional linkage analysis model in step S2 are as follows: S2.1: Define five core indicators and clarify quantitative standards, covering spare parts replacement cycle, number of associated equipment, equipment operating status, degree of failure impact and extreme scenario risks, to achieve standardized quantification of multi-dimensional influencing factors; S2.2: A prediction model is built using a Long Short-Term Memory Neural Network (LSTM). Leveraging its advantage in capturing long-term dependencies in time series data, it is adapted to the temporal characteristics of parts replacement and failure occurrence to improve prediction accuracy. The model is set with fixed structural parameters. The input layer corresponds to five core indicators, and the hidden layer and output layer are responsible for feature extraction and predicted value output, respectively. The ReLU activation function, Adam optimizer, and preset learning rate are used to ensure the model's computational efficiency. S2.3: By using two independent prediction formulas and substituting the corresponding initial weights, the spare parts replacement cycle and equipment failure rate are predicted synchronously. The initial weights are iteratively optimized through the training dataset to ensure that the prediction results fit the actual operating scenario.

4. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for model optimization and result output in step S3 are as follows: S3.1: Design defects and environment are the two core factors affecting spare parts wear. Corresponding influence coefficients and quantitative standards are set for each. The degree of influence of design defects is reflected by combining the level quantitative value with the influence coefficient. Environmental factors are first weighted and summed for sub-factors such as temperature, humidity, dust concentration, and extreme weather to obtain a comprehensive quantitative value, and then combined with the environmental influence coefficient to quantify the effect. S3.2: Construct a loss increment function to integrate the quantitative impact of the two types of factors into a unified loss increment. Substitute it into the original prediction result output by S2, and adjust the spare parts replacement cycle and equipment failure rate prediction values ​​through the multiplication correction formula to offset the interference of loss factors on prediction accuracy. S3.3: Based on the optimized predicted value, reserve 10% redundancy to determine the spare parts replacement cycle threshold, and output the optimized equipment failure rate simultaneously.

5. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 4, characterized in that, The specific steps for constructing the featured inventory management module in step S4 are as follows: S4.1: Based on monthly turnover rate as the core classification criterion, a threshold of 3 times / month is set to divide spare parts into fast-moving and slow-moving parts. The attributes are quantified through the turnover rate formula to provide a classification standard for mode adaptation. S4.2: For fast-moving components, combined with their high demand and high turnover characteristics, the Economic Order Quantity (EOQ) algorithm is used to construct a dynamic turnover inventory model. The core formula integrates the annual demand forecast, the cost of one order and the unit annual holding cost to accurately calculate the optimal inventory quantity. At the same time, the annual demand forecast is updated monthly to dynamically adjust the inventory size, maximize inventory turnover efficiency and reduce the risk of backlog. S4.3: For slow-moving parts, considering their low demand and high backlog risk, three targeted management modes are configured: joint inventory management allocates shared inventory according to the number of sites through formulas to achieve inventory concentration; spare parts pool shared management sets the upper limit of call volume and cost rules to balance the flexibility and fairness of sharing; and supplier emergency supply agreement clarifies the minimum supply quantity standard and relies on response time to ensure emergency needs.

6. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for selecting the optimal inventory management mode in step S5 are as follows: S5.1: Taking spare parts characteristics, supply cycle, coordination difficulty, and stockout loss risk as the core decision nodes, construct a decision tree model and set judgment thresholds for each node, build a systematic pattern screening framework, and clarify the screening dimensions and judgment criteria. S5.2: The Analytic Hierarchy Process (AHP) is used to quantify the weights of each decision parameter. An importance ratio judgment matrix is ​​constructed between parameters, and the initial weights are obtained by substituting them into the weight calculation formula. Then, a consistency check is performed to verify the rationality of the weight allocation. If it is not qualified, the matrix is ​​adjusted until it meets the standard, ensuring that the weights are in line with the priority of the actual scenario. S5.3: Calculate the adaptation score of each management mode using a weighted summation formula, multiply the weight of each parameter by the adaptation score of the corresponding mode, and sum them to quantify the mode adaptation capability. S5.4: Optimal models are selected by sorting according to their suitability scores, with the highest score being the first choice. In case of a tie, the lowest risk of stockout loss is used as a supplementary criterion to ensure that the selection results take into account both suitability and supply security, providing a definite management model sample for subsequent full-cost quantitative analysis.

7. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for constructing the full-cost quantification model in step S6 are as follows: S6.1: Collect three types of core input parameters, including real-time supply chain quotations, industry implicit cost benchmarks and replenishment lead time, and stockout loss amount, to ensure that the parameters cover cost composition and scenario characteristics; S6.2: Integrate four categories of costs: procurement, inventory holding, stockout losses, and hidden costs. Clarify the core formulas for each sub-item cost and the total cost. Clarify the meaning and calculation logic of each variable through parameter definition. The average inventory quantity is differentiated according to the characteristics of fast and slow flow items. The stockout probability is quantified by combining existing inventory and predicted demand. This achieves accurate decomposition and aggregation of the entire cost and builds a scientific full-cost quantification model. S6.3: Based on the Monte Carlo simulation algorithm, the optimal mode selected by S5 and other alternative modes are substituted to perform multi-scenario simulation calculations on the total cost distribution, service level achievement rate and stockout risk probability under various modes, and output standardized result reports.

8. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The specific steps for constructing the intelligent inventory management model in step S7 are as follows: S7.1: Taking the total cost quantified in S6 as the core, set the objective function of minimizing the total inventory cost, integrate procurement, holding, stockout losses and hidden costs, and add inventory constraints to ensure that the replenishment quantity is not less than the difference between the predicted demand and the existing inventory. S7.2: Based on the equipment failure rate and spare parts replacement cycle prediction value optimized by S3, and superimposed with a redundancy coefficient of 1.2, calculate the inventory warning threshold. When the existing inventory is lower than the threshold, a replenishment reminder is automatically triggered. S7.3: The replenishment parameters are accurately derived through two core formulas. The replenishment quantity is calculated by combining the predicted demand and the replenishment lead time to ensure that the inventory demand is fully covered within the lead time. The replenishment time is calculated by working backward from the existing inventory consumption rate to reserve sufficient replenishment cycle to avoid stockouts. S7.4: Establish a dynamic adjustment mechanism to update forecasted demand, existing inventory, and cost parameters every 7 days, and simultaneously iterate the early warning threshold, replenishment quantity, and timing.

9. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, In step S2, the spare parts replacement cycle and equipment failure rate prediction model uses a Long Short-Term Memory (LSTM) neural network algorithm. The model input is historical data of five core indicators, and the output is the predicted value. The core function formula is as follows: (1) Formula for predicting spare parts replacement cycle: in, This represents the predicted spare parts replacement cycle value at time t+k. The initial weights for the five core indicators, Let t be the actual value of the spare parts replacement cycle at time t. Let t be the actual equipment failure rate. Let be the number of spare parts associated with the device at time t. Let be the extent of the fault's impact at time t. Let be the extreme scenario risk value at time t; (2) Equipment failure rate prediction formula: in This represents the predicted equipment failure rate at time t+k. The initial weights for the five core indicators are defined, with values ​​ranging from (0,1). , ; The functional formula for the impact of quantitative design defects and environmental factors on spare parts wear in S3 is as follows: in, To account for the increase in spare parts wear and tear, The design defect impact coefficient has a value range of [0.1, 0.5]. To quantify the design defect level, The comprehensive impact coefficient of environmental factors ranges from [0.2, 0.6], and E is the comprehensive quantitative value of environmental factors, which is calculated by weighting sub-factors such as temperature, humidity, and dust concentration. in, , , , These are sub-factors such as temperature, humidity, and dust concentration; The optimized prediction model formula is as follows: in, , These are the optimized spare parts replacement cycle and the predicted equipment failure rate. .

10. The inventory optimization method for analyzing the replacement cycle and failure rate of charging facility spare parts according to claim 1, characterized in that, The core formula of the fast-moving components dynamic turnover inventory management mode in S4 is: in, For dynamic turnover inventory batches, This is the projected annual demand for spare parts. Cost per order, Annual inventory holding cost per unit of spare parts; The formula for shared inventory in joint inventory management of slow-moving components is: in, K represents the number of charging facility sites participating in the joint inventory, which is the shared inventory quantity. The parameter weight quantification of the decision tree model in S5 adopts the Analytic Hierarchy Process (AHP), and the weight calculation function is: in, The weight of the i-th parameter, This represents the importance ratio of the i-th parameter relative to the k-th parameter. This represents the total number of parameters. The formula for the fit score is: in, The adaptation score for the m-th management model is... Score the fit of the i-th parameter with the m-th management mode; The total cost formula for the full-cost quantification model in S6 is: in, For spare parts procurement costs, For inventory holding costs, To avoid losses due to stockouts, Hidden costs; The formulas for each item of cost are as follows: Where Q is the quantity purchased and P is the unit price of spare parts; in, This represents the average inventory quantity. Annual holding cost per unit of spare parts; in, This represents the probability of being out of stock. This refers to the amount of loss due to a single stockout. This is an implicit cost coefficient, with a value range of [0.05, 0.15], determined based on industry benchmarks.

Images