A time-sharing leasing solution system for IT equipment
Through real-time data analysis and simulation-optimized IT equipment time-sharing leasing solutions, the problems of insufficient resource allocation and inflexible pricing strategies in the changes in the market environment are solved, accurate prediction of market demand and efficient allocation of resources are achieved, and service quality is improved and customer satisfaction.
Patent Information
- Application Number
- CN202410940302.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-15
AI Technical Summary
Existing IT equipment time-sharing leasing solutions lack market adaptability and response speed in a rapidly changing market environment, resulting in insufficient or over-allocation of resources, inflexible pricing strategies, low failure response efficiency, and affect customer experience and corporate reputation.
Adopt user cognitive analysis module, market demand analysis module, pricing strategy optimization module, event monitoring and processing module, behavior trend prediction module, resource allocation adjustment module and fault response optimization module, and generate optimized pricing through real-time data analysis and simulation optimization. Plans, resource adjustment plans and fault response strategies to achieve accurate prediction of market demand and efficient allocation of resources.
It improves the sensitivity and response speed of market pricing, optimizes resource allocation, significantly reduces operating costs, and improves service quality and customer satisfaction.
Smart Images

Figure CN118469681B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of time-sharing leasing, and in particular to a time-sharing leasing solution system for IT equipment. Background Art
[0002] The field of time-sharing technology refers to systems and methods for sharing or leasing assets within a specific period of time. It is widely used in different asset types, such as real estate, cars, equipment, etc., to improve resource utilization and reduce costs. In the context of IT equipment time-sharing, it specifically refers to the use of software systems or platforms to enable multiple users to lease and use the same or multiple IT equipment, such as servers, storage devices, network resources, etc., in different time periods.
[0003] Among them, the IT equipment time-sharing leasing solution system is a system designed to enable enterprises or individuals to lease IT equipment when needed, with the aim of providing a flexible, cost-effective way to meet temporary or periodic IT resource needs, avoiding the burden of long-term purchase and maintenance costs. By implementing this solution, users can lease IT equipment according to actual needs, thereby achieving the effect of optimizing resource allocation, reducing operating costs and improving business flexibility.
[0004] Existing technologies usually lack sufficient market adaptability and response speed, which may lead to insufficient or excessive resource allocation in a rapidly changing market environment. At the same time, the lack of real-time analysis of market demand elasticity and flexible adjustment of pricing strategies often makes it impossible for enterprises to effectively respond to market fluctuations, resulting in reduced economic benefits. In addition, fault responses in traditional solutions often rely too much on preset processes and lack optimized processing for actual monitoring data, resulting in delayed responses and inefficient repairs, which in turn affects customer experience and corporate reputation, making it difficult to meet customers' expectations for efficient and reliable services. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an IT equipment time-sharing leasing solution system.
[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: An IT equipment time-sharing leasing solution system comprises:
[0007] The user cognitive analysis module analyzes user preferences and sensitivities based on user interaction data, identifies differentiated user groups, and generates user cognitive bias records;
[0008] The market demand analysis module analyzes the current market capacity and potential growth space based on the user cognitive bias record, evaluates the market demand elasticity and the feasibility of the pricing strategy, and generates a market demand analysis record;
[0009] The pricing strategy optimization module adjusts the pricing strategy based on the market demand analysis record, simulates the impact of the differentiated pricing scheme on the market response, and generates an optimized pricing scheme;
[0010] The event monitoring and processing module captures events in real time based on online user behavior and market dynamics, analyzes the impact of rental demand and user behavior, and generates event response records;
[0011] The behavior trend prediction module classifies and analyzes the user behavior data based on the event response records to generate a behavior trend prediction result;
[0012] The resource allocation adjustment module optimizes the allocation of available IT equipment rental resources based on the behavior trend prediction results and generates a resource adjustment plan;
[0013] The fault response optimization module optimizes the fault handling process and response strategy based on equipment monitoring data and historical fault records, and generates a fault response optimization strategy;
[0014] The full life cycle management module comprehensively evaluates the equipment performance based on the optimized pricing plan, resource adjustment plan, fault response optimization strategy and user feedback, and generates a full life cycle management plan.
[0015] As a further solution of the present invention, the user cognitive bias record includes the user's anchoring effect analysis, risk preference analysis and time preference inconsistency analysis, the market demand analysis record includes the supply and demand balance status, price elasticity analysis and market saturation index, the optimized pricing plan includes the pricing strategy for differentiated time periods and its expected impact on user acceptance and system utilization, the event response record includes the event type, impact scope and potential demand changes, the behavior trend prediction results include future user rental behavior trend prediction and potential market demand change analysis, the resource adjustment plan includes the resource optimized allocation plan, expected inventory level and adjusted response strategy, the fault response optimization strategy includes the accuracy of fault detection, the efficiency of the diagnostic process and the improvement of response measures, and the full life cycle management plan includes equipment maintenance plan, upgrade recommendations and reuse strategy.
[0016] As a further solution of the present invention, the pricing strategy optimization module includes:
[0017] The pricing model building submodule builds a pricing model for the rental period based on the market demand analysis record, solves the price setting for maximizing the revenue, and generates a preliminary pricing strategy;
[0018] The user acceptance evaluation submodule performs regression analysis on user purchase data at different price levels based on the preliminary pricing strategy, calculates price elasticity, evaluates user acceptance, and generates user price sensitivity analysis results;
[0019] The pricing optimization submodule performs price optimization based on the user price sensitivity analysis results, iteratively optimizes the preliminary pricing strategy, and generates an optimized pricing plan.
[0020] As a further solution of the present invention, the event monitoring processing module includes:
[0021] The event capture submodule filters and converts data based on online user behavior and market dynamics, retains key event data, including login anomalies, trading operations or major market news, and obtains key event data streams;
[0022] The event analysis submodule configures a sliding event window based on the key event data stream to aggregate and analyze time series data, identify abnormal behaviors and important market changes, evaluate potential impacts, and generate event impact analysis results;
[0023] The event record submodule analyzes the event details, including event type, time, affected users and market reactions, based on the event impact analysis results, optimizes data access and query efficiency, and establishes event response records.
[0024] As a further solution of the present invention, the behavior trend prediction module includes:
[0025] The data classification submodule groups the user behavior data based on the event response records, distinguishes the user behavior patterns, and generates user behavior grouping results;
[0026] The trend analysis submodule analyzes the change trends of different user behavior groups based on the user behavior group division results and generates trend analysis results;
[0027] The trend prediction submodule adopts the ARIMA model based on the trend analysis results, performs model training according to historical user behavior trend data, predicts future user behavior and rental demand trends, and generates behavior trend prediction results.
[0028] As a further solution of the present invention, the ARIMA model is according to the formula:
[0029] ;
[0030] Calculate future user behavior and rental demand trends , generate behavioral trend prediction results, where is a constant term, , is the coefficient of the autoregressive term, , represent the difference values of the time series in the previous period and the previous two periods, , is the coefficient of the moving average term, , are random error terms representing the time series in the previous period and the previous two periods, is the random error term at the current time point, is the seasonal influence coefficient, is the weight coefficient of seasonal impact, is the industry dynamics index, is the weight coefficient of the impact of industry dynamics, For specific events, is the weight coefficient of the impact of a specific event, is the technological progress adjustment coefficient, is the weight coefficient of the impact of technological progress.
[0031] As a further solution of the present invention, the resource allocation adjustment module includes:
[0032] The resource assessment submodule analyzes the usage, maintenance history and performance indicators of the equipment based on the behavior trend prediction results, identifies resources with high maintenance requirements and frequent use, and generates comprehensive resource assessment results;
[0033] The inventory management submodule dynamically calculates inventory levels and replenishment points based on the comprehensive resource evaluation results, analyzes the difference between forecasted demand and existing inventory, adjusts inventory strategies, and generates inventory optimization plans;
[0034] The dynamic adjustment submodule of the scheme uses a linear programming method based on the inventory optimization scheme to optimize and adjust resource allocation, formulate a resource allocation plan, balance resource utilization and cost-effectiveness, and generate a resource adjustment scheme.
[0035] As a further solution of the present invention, the linear programming method is according to the formula:
[0036] ;
[0037] Calculate the optimal allocation of each resource , generate a resource adjustment plan, where For the The unit cost of the class resource, For the The amount of class resources allocated, For the Risk adjustment factor for class resources, For the The demand volatility of the resource class, is the weight coefficient for balancing cost and resource utilization, For the The resource requirements of the activity, For the Class resources for The contribution of a type of activity, n represents the type of resource, and m represents the type of activity.
[0038] Compared with the prior art, the advantages and positive effects of the present invention are:
[0039] In the present invention, through in-depth analysis of market demand and real-time feedback mechanism, more accurate resource allocation and demand forecasting are achieved. The assessment of market capacity and growth space makes leasing pricing more market-sensitive and can be adjusted in real time to adapt to market changes, thereby improving the effectiveness of pricing strategies. At the same time, the classification and trend analysis of user behavior data can predict future resource needs and optimize resource allocation, which not only promotes the accurate capture of leasing demand, but also significantly improves service quality and system reliability by optimizing fault response processes, significantly reduces operating costs, and improves business flexibility and customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 is a system flow chart of the present invention;
[0041] Figure 2 It is a schematic diagram of the system framework of the present invention;
[0042] Figure 3 This is a flow chart of the user cognitive analysis module of the present invention;
[0043] Figure 4 It is a flow chart of the market demand analysis module of the present invention;
[0044] Figure 5 This is a flowchart of the pricing strategy optimization module of the present invention;
[0045] Figure 6 It is a flow chart of the event monitoring processing module of the present invention;
[0046] Figure 7 It is a flow chart of the behavior trend prediction module of the present invention;
[0047] Figure 8 It is a flow chart of the resource allocation adjustment module of the present invention;
[0048] Fig. 9 This is a flow chart of the fault response optimization module of the present invention;
[0049] Fig.10 This is a flow chart of the full life cycle management module of the present invention. DETAILED DESCRIPTION
[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0051] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0052] Example 1: Please refer to Figure 1 and Figure 2 , an IT equipment time-sharing leasing solution system includes:
[0053] The user cognitive analysis module analyzes user preferences and sensitivities based on user interaction data, identifies differentiated user groups, and generates user cognitive bias records;
[0054] The market demand analysis module analyzes the current market capacity and potential growth space based on the user cognitive bias record, evaluates the market demand elasticity and the feasibility of the pricing strategy, and generates a market demand analysis record;
[0055] The pricing strategy optimization module adjusts the pricing strategy based on the market demand analysis record, simulates the impact of the differentiated pricing scheme on the market response, and generates an optimized pricing scheme;
[0056] The event monitoring and processing module captures events in real time based on online user behavior and market dynamics, analyzes the impact of rental demand and user behavior, and generates event response records;
[0057] The behavior trend prediction module classifies and analyzes the user behavior data based on the event response records to generate a behavior trend prediction result;
[0058] The resource allocation adjustment module optimizes the allocation of available IT equipment rental resources based on the behavior trend prediction results and generates a resource adjustment plan;
[0059] The fault response optimization module optimizes the fault handling process and response strategy based on equipment monitoring data and historical fault records, and generates a fault response optimization strategy;
[0060] The full life cycle management module comprehensively evaluates the equipment performance based on the optimized pricing plan, resource adjustment plan, fault response optimization strategy and user feedback, and generates a full life cycle management plan.
[0061] User cognitive bias records include user anchoring effect analysis, risk preference analysis and time preference inconsistency analysis. Market demand analysis records include supply and demand balance status, price elasticity analysis and market saturation indicators. The optimized pricing plan includes the pricing strategy for differentiated time periods and its expected impact on user acceptance and system utilization. Event response records include event type, impact scope and potential demand changes. Behavioral trend prediction results include future user rental behavior trend prediction and potential market demand change analysis. Resource adjustment plan includes resource optimized allocation plan, expected inventory level and adjusted response strategy. Fault response optimization strategy includes the accuracy of fault detection, efficiency of diagnostic process and improvement of response measures. The full life cycle management plan includes equipment maintenance plan, upgrade recommendations and reuse strategy.
[0062] See also Figure 2 and Figure 3 , the user cognitive analysis module includes:
[0063] The data collection submodule removes null values and duplicate records based on user interaction data, replaces inconsistent format data in a unified manner, performs data standardization, converts data formats, and generates user original behavior data sets;
[0064] First, set the standardized target format for each data entry, for example, convert the date from "day / month / year" to "year-month-day" format. For text data, lowercase letters are used uniformly and all non-alphabetic characters are removed. Numerical data is normalized to ensure that all data is compared on the same scale. Next, use a one-by-one inspection method to traverse each record in the dataset. For each field, check whether its data type conforms to the predetermined format, and convert it if it does not conform. After completing this process, apply a deduplication algorithm to the dataset to identify and delete duplicates based on the joint hash value of all fields. Finally, the processed data is output as a standardized user original behavior dataset.
[0065] The user feature extraction submodule extracts key user behavior feature vectors based on the user's original behavior data set, applies the K-means clustering algorithm to identify user group characteristics, and generates a user feature vector set;
[0066] First, define key behavioral features according to business needs, such as page visit frequency, purchase behavior type, etc., and extract these features from the data set to form a feature vector. Use normalization to ensure that each feature is in the same numerical range to avoid a certain feature dominating the distance calculation. Next, set the number of clusters for the K-means algorithm and the selection strategy for the initial center point, such as random selection or using the K-means++ method to optimize the selection of the initial center. During the iteration process, calculate the distance from each data point to each center point based on the Euclidean distance, assign it to the nearest cluster, and then update the center point of the cluster to the mean position of all members of the cluster. Repeat this process until the movement of the cluster center is less than the predetermined threshold or the maximum number of iterations is reached, and finally output the cluster assignment results of each data point to form a user feature vector set.
[0067] The cognitive bias analysis submodule analyzes the selection bias in differentiated scenarios based on the user feature vector set, divides user groups, identifies the preferences and sensitivities of different user groups to rental periods, and generates user cognitive bias records;
[0068] First, based on the feature vector set, differentiated scenarios are defined, such as user behavior during peak and off-peak hours. Decision tree analysis is used to divide user groups according to user behavior feature vectors, identifying user groups with significantly different preferences and sensitivities in different scenarios. On this basis, the specific preferences of each group are further analyzed. For example, some groups may prefer to rent on weekdays, while others may prefer weekends. By setting preference strength indicators and sensitivity thresholds, the specific preferences of each group are calculated and calibrated. Finally, all analysis results are integrated to generate a cognitive bias record containing user group classification, preference and sensitivity indicators for subsequent strategy formulation and optimization.
[0069] See also Figure 2 and Figure 4 ,The market demand analysis module includes the supply and demand data analysis submodule, the market saturation evaluation submodule, and the demand prediction submodule;
[0070] The supply and demand data analysis submodule conducts supply and demand curve analysis based on user cognitive bias records, analyzes the equilibrium price and quantity obtained through the intersection of the curves, and generates supply and demand status analysis results based on the changes in supply and demand relationships under differential price levels;
[0071] First, a supply and demand curve model is constructed, in which the supply function is set to S(p)=αp+β, and the demand function is set to D(p)=γ-δp, where p represents the price, and α, β, γ, and δ are parameters estimated from the data. Through the user cognitive bias record, the parameters α and δ are adjusted to reflect the differences in price sensitivity among different user groups. Next, the intersection of the supply and demand curves is calculated, that is, the equation αp+β=γ-δp is solved to obtain the equilibrium price p*=(γ-β) / (α+δ). Further analyze the changes in supply and demand at different price levels, and use the equilibrium price as a benchmark to explore the impact of price fluctuations on the supply and demand relationship. Finally, the curve analysis results and user data on price sensitivity are combined to generate analysis results describing the current market supply and demand status and price sensitivity distribution.
[0072] The market saturation assessment submodule aggregates market data based on the supply and demand status analysis results, evaluates the current market share and potential growth space of all market segments, analyzes the saturation of all market segments, and generates market saturation assessment results;
[0073] First, define market saturation as the ratio of current market share to total potential market demand, i.e., saturation = current market share / total potential demand. Use the supply and demand balance data from the supply and demand status analysis results to estimate the current market share of each market segment. Then, combine industry reports and market research data to estimate the total potential demand of each market segment. By comparing the current market share and potential demand, evaluate the saturation status of each market segment and identify market segments with greater growth potential. Finally, summarize the saturation assessment of each market segment to form a detailed market saturation report to provide a basis for strategic decision-making.
[0074] The demand forecasting submodule obtains historical IT equipment leasing data, analyzes future demand trends based on market saturation assessment results, combined with seasonal factors and market dynamics, and generates market demand analysis records;
[0075] First, a baseline demand model is constructed using historical data, and time series analysis is introduced into the model to identify historical trends and seasonal fluctuations. Combined with the market saturation assessment results, the growth rate parameters in the model are adjusted to reflect the impact of different saturations of each market segment on demand. Further analysis of external market dynamics, such as economic indicators and industry policy changes, is used to adjust model forecasts. Finally, all factors are combined to predict market demand in the next few quarters, generating a market demand analysis record that comprehensively reflects market trends, seasonal effects, and external dynamics.
[0076] See also Figure 2 and Figure 5 , the pricing strategy optimization module includes:
[0077] The pricing model building submodule builds a pricing model for the rental period based on market demand analysis records, solves the price setting for maximizing revenue, and generates a preliminary pricing strategy;
[0078] First, define the pricing model as the revenue function R(p)=p×Q(p), where p is the price and Q(p) is the expected sales volume at price p. The function form of Q(p) is estimated based on the market demand analysis records. The quadratic function Q(p)=a-bp+cp is used. 2 To simulate the impact of price on sales, the parameters a, b, and c are obtained by fitting historical data. By solving the derivative of the revenue function and setting it to zero, that is, dR / dp=Q(p)+p*dQ / dp=0, find the price p* that maximizes revenue. Solve this equation with a derivative equal to zero to obtain possible price points, and further determine which of these price points maximize revenue through a quadratic derivative test. Finally, integrate these analysis and calculation steps to generate a preliminary pricing strategy that describes how to adjust prices to maximize revenue.
[0079] The user acceptance evaluation submodule performs regression analysis on user purchase data at different price levels based on the preliminary pricing strategy, calculates price elasticity, evaluates user acceptance, and generates user price sensitivity analysis results;
[0080] First, extract the data points of price and sales from the purchase data, and use the linear regression model Ln(Q)=Ln(Q0)-ε*Ln(P), where Q is sales, P is price, ε is the price elasticity coefficient, and Q0 is the benchmark sales. The model parameters ε and Q0 are estimated through regression analysis. The price elasticity ε represents the percentage change in sales when the price changes by 1%. Estimate the price elasticity for different user groups and different product categories to reflect the differences in the market. According to the regression results, analyze the impact of price changes on sales, so as to evaluate the user's sensitivity to price changes. Finally, summarize the analysis results to form a user price sensitivity analysis report to provide a basis for adjusting the price strategy.
[0081] The pricing optimization submodule performs price optimization based on the results of user price sensitivity analysis, iteratively optimizes the initial pricing strategy, and generates an optimized pricing plan;
[0082] First, according to the results of the user price sensitivity analysis, adjust the parameters in the original pricing model, especially among the user groups who perceive higher sensitivity, appropriately lower the price to increase acceptance. Set multiple price test points, and use simulation testing methods to predict the market response of each test price point, including expected sales and revenue. By comparing the expected results of each test point, select the price point that maximizes revenue and has high user acceptance as the new selling price. In addition, considering seasonal factors and market competition, segment the market for differentiated pricing. During the iteration process, continuously collect market feedback and adjust the pricing model parameters until the balance point with optimal revenue and highest user acceptance is found. Finally, generate an optimized pricing plan that integrates current market conditions and user behavior.
[0083] See also Figure 2 and Figure 6 , the event monitoring processing module includes:
[0084] The event capture submodule filters and converts data based on online user behavior and market dynamics, retains key event data, including login anomalies, trading operations or major market news, and obtains key event data streams;
[0085] First, implement data filtering strategies to capture qualified events by setting specific keywords and behavior patterns, such as "too many failed logins", "large transactions", and "market news keywords". Next, convert the captured data to a unified timestamp format to ensure consistency and accuracy in event descriptions. For each type of event captured, further filter out events with significant impact potential, such as login anomalies that may indicate security risks, large transactions that may indicate market trends, and major market news that may affect the direction of the entire industry, to build a continuously updated key event data stream to provide real-time data support for subsequent analysis.
[0086] The event analysis submodule configures a sliding event window to aggregate and analyze time series data based on key event data streams, identifies abnormal behaviors and important market changes, assesses potential impacts, and generates event impact analysis results;
[0087] First, set sliding window parameters, such as window size and step size, to accommodate the time sensitivity of different types of events. Within each sliding window, aggregate and analyze event data, using statistical methods and pattern recognition techniques to identify abnormal patterns and trend changes in the data. For example, by comparing the short-term login failure rate with the long-term average, potential security risks can be identified. At the same time, for market news, analyze the correlation between news events and market price fluctuations, evaluate the market influence of news events, form an impact assessment of events within each window period, and integrate them into event impact analysis results.
[0088] The event record submodule analyzes event details based on the event impact analysis results, including event type, time, affected users and market reactions, optimizes data access and query efficiency, and establishes event response records;
[0089] First, extract key information from the event impact analysis results, such as the type of event, the specific time of occurrence, the affected user groups and their reactions, and design an efficient data storage solution, such as using indexes to optimize query speed and applying data compression technology to reduce storage requirements. For each type of event, establish a standardized record format to ensure that all event records can be quickly retrieved and uniformly presented. In addition, implement data security measures to ensure the confidentiality and integrity of event data, and build a comprehensive event response record system so that detailed information on historical events can be quickly accessed at any time to support rapid response and decision-making.
[0090] See also Figure 2 and Figure 7 , the behavior trend prediction module includes:
[0091] The data classification submodule groups user behavior data based on event response records, distinguishes user behavior patterns, and generates user behavior grouping results;
[0092] First, a set of feature vectors is defined based on the detailed user behavior data in the event response records, such as login frequency, transaction type, and interaction preference. Using the feature vectors, a hierarchical clustering method is applied to group the user data, and the distance between the feature vectors is calculated to construct a similarity dendrogram of user behavior. An appropriate distance threshold is selected and the dendrogram is cut to form clearly distinguishable groups. Statistical analysis is performed on each generated group to determine its main behavioral characteristics and group size, which can clearly distinguish user groups with different behavioral patterns, providing a basis for further behavioral analysis and personalized marketing strategies.
[0093] The trend analysis submodule analyzes the changing trends of different user behavior groups based on the user behavior group division results and generates trend analysis results;
[0094] First, we conduct time series analysis on the historical data of each user group to identify the key time nodes and trends of each group's behavior changes. We use sliding window technology to calculate the behavioral parameters of each group in different time periods, such as average transaction volume and activity index. Through correlation calculations, we reveal the behavioral dynamics of each group during the observation period. We compare and analyze the behavioral trends between different groups, identify potential behavioral pattern changes and similarities, and provide targeted insights for marketing and product development.
[0095] The trend prediction submodule uses the ARIMA model based on the trend analysis results, performs model training based on historical user behavior trend data, predicts future user behavior and rental demand trends, and generates behavior trend prediction results;
[0096] ARIMA model, according to the formula:
[0097] ;
[0098] Calculate future user behavior and rental demand trends , represents the processed user behavior or rental demand trend data at the current time point, generating the behavior trend prediction result, where, is a constant term representing the bias or baseline level in the model, , is the coefficient of the autoregressive term, which indicates the impact of the previous one or two periods of the time series on the current value. , represent the difference values of the time series in the previous period and the previous two periods, , is the coefficient of the moving average term, which indicates the impact of the random error of the previous one or two periods on the current error. , are random error terms representing the time series in the previous period and the previous two periods, is the random error term at the current time point, is the seasonal impact coefficient, which reflects the impact of seasonal changes on user behavior or rental demand trends. is the weight coefficient of seasonal impact, It is an industry dynamic index that reflects the impact of changes in industry conditions or economic environment on trends. is the weight coefficient of the impact of industry dynamics, For the impact of specific events, consider the impact of unconventional factors such as major events or holidays on trends. is the weight coefficient of the impact of a specific event, The technology progress adjustment coefficient reflects the long-term impact of technology development on user behavior or rental demand trends. is the weight coefficient of the impact of technological progress.
[0099] The execution process is as follows:
[0100] Data preprocessing and difference: First, preprocess the original time series data, including cleaning and missing value processing, and then perform a difference to ensure the stability of the data and obtain ;
[0101] External Factors Data Collection: Collect and , , , Relevant data. For example, collect data on seasonal changes, industry reports, dates of major events and their possible impacts, technology development reports, etc.
[0102] Determination of weight coefficient: For , , , , and estimate their effects on to determine the degree of impact , , , This step may involve multivariate regression analysis, where As the dependent variable, , , , As independent variables, statistical tests (e.g., t-tests) and information criteria (e.g., AIC, BIC) are used to assess the goodness of fit of the model and select the most appropriate coefficient values;
[0103] Model integration and parameter estimation: All parameter values obtained , , , Substitute it into the improved ARIMA model, use historical data to fit this cross-section, and estimate the parameters.
[0104] Predict future trends: Based on the fitted model and parameters, predict future The improved ARIMA model can provide more accurate prediction results by considering the internal time series characteristics and the influence of external factors.
[0105] See also Figure 2 and Figure 8 , the resource allocation adjustment module includes:
[0106] The resource assessment submodule analyzes the equipment usage, maintenance history, and performance indicators based on the behavior trend prediction results, identifies resources with high maintenance requirements and frequent use, and generates comprehensive resource assessment results;
[0107] First, integrate the equipment usage data, including usage frequency, usage duration, and operating conditions, combined with the equipment maintenance history, such as the number of repairs and reasons for repairs. Evaluate equipment performance indicators, including equipment efficiency and failure rate. Apply cluster analysis to classify equipment according to frequency of use and maintenance needs, identify resources that require priority maintenance, and equipment that is frequently used and performs stably. Finally, generate a resource assessment report that comprehensively considers usage efficiency and maintenance needs, providing data support for resource optimization.
[0108] The inventory management submodule dynamically calculates inventory levels and replenishment points based on comprehensive resource evaluation results, analyzes the difference between forecasted demand and existing inventory, adjusts inventory strategies, and generates inventory optimization plans;
[0109] First, forecast resource demand in the short term based on resource assessment results and behavioral trend forecasts. Use the economic order quantity (EOQ) model to calculate the optimal order quantity and replenishment point. This model takes into account ordering costs and holding costs to find a cost-minimizing ordering strategy. Use a safety stock strategy to buffer the uncertainty of demand forecasts and determine the safety stock level for each resource. Dynamically adjust the inventory system and update the inventory strategy based on market demand and resource usage frequency. Finally, develop a detailed inventory optimization plan to reduce out-of-stock and overstock risks and optimize inventory costs and service levels.
[0110] The dynamic adjustment submodule uses linear programming methods based on the inventory optimization plan to optimize resource allocation, formulate resource allocation plans, balance resource utilization and cost-effectiveness, and generate resource adjustment plans;
[0111] Linear programming method, according to the formula:
[0112] ;
[0113] Calculate the optimal allocation of each resource ,in, For the The unit cost of the class resource, For the The amount of class resources allocated, For the Risk adjustment factor for class resources, For the The demand volatility of the resource class, and the weight coefficient for balancing cost and resource utilization. For the The resource requirements of the activity, For the Class resources for The contribution of a type of activity, n represents the type of resource, and m represents the type of activity.
[0114] The execution process is as follows:
[0115] First, the cost of each resource is calculated, taking into account demand fluctuations and resource risks, and by introducing a risk adjustment factor and demand volatility To adjust the unit cost of resources to more accurately reflect the cost situation, The determination can be set by analyzing the risk characteristics of resources through historical data, The fluctuation of resource demand can be calculated by analyzing historical demand data. Secondly, by adding a penalty term To consider resource utilization, It is a weight coefficient used to balance the relationship between minimizing costs and maximizing resource utilization. Its value can be set according to the specific situation through trial and error or expert experience. Finally, the improved objective function is solved by the linear programming algorithm. , get each resource.
[0116] See also Figure 2 and Fig. 9 , the fault response optimization module includes:
[0117] The fault detection submodule performs real-time abnormal pattern recognition based on equipment monitoring data and historical fault records, identifies abnormal behaviors in equipment operation data, locates potential fault starting points, and generates preliminary fault identification records;
[0118] First, integrate the monitoring data of the equipment, such as temperature, pressure, speed and other sensor data, and compare them with historical fault records. Apply threshold analysis technology to set the normal operating range of each sensor data. When the real-time data exceeds these thresholds, the system marks it as a potential abnormal event. Further, use the autoregressive model to analyze the short-term fluctuations and long-term trends in the data, so as to more accurately identify abnormal patterns, accurately locate abnormal behaviors that may occur in equipment operation, and generate preliminary fault identification records to provide a basis for subsequent fault diagnosis.
[0119] The fault diagnosis submodule uses a convolutional neural network model to diagnose abnormal events based on the preliminary fault identification records, analyze the fault type and cause, and generate detailed fault diagnosis results;
[0120] First, a convolutional neural network (CNN) is used to process the time series of monitoring data. The model learns multiple levels of features to identify complex patterns in the data. The CNN model is trained with historical fault data as the training set, so that the model can identify the characteristics of various faults. During the diagnosis process, the model analyzes abnormal data segments and identifies features that match known fault types. The diagnosis results are post-processed, including the classification of fault types and the inference of fault causes, to make the fault diagnosis results more detailed and accurate. Finally, a detailed fault diagnosis report is generated, including the fault type, cause, and recommended maintenance measures.
[0121] The response strategy optimization submodule defines the fault type and equipment status as the environmental status and the fault response measures as potential actions based on the detailed fault diagnosis results, performs iterative learning, and generates a fault response optimization strategy.
[0122] First, a reinforcement learning framework is used, with fault type and equipment status as the state space of the environment and fault response measures as the action space of the agent. By simulating the equipment response under different fault conditions through historical data, the agent learns how to select the optimal fault response measures to maximize the stable operation of the equipment and minimize the maintenance cost. Using Q learning or other suitable reinforcement learning algorithms, the decision-making strategy is continuously optimized, and the fault response efficiency is improved through the reward mechanism. In the iterative process, a set of response strategies for different fault conditions is finally generated, aiming to improve the reliability of the equipment and reduce the maintenance cost.
[0123] See also Figure 2 and Fig.10 , the full life cycle management module includes:
[0124] The performance evaluation submodule analyzes the relationship between IT equipment performance and usage patterns based on the optimized pricing scheme, fault response optimization strategy and user feedback, reveals the relationship between key performance indicators and equipment usage patterns, and generates comprehensive performance analysis results;
[0125] First, collect and organize the optimized pricing plan, the implementation effect of the fault response strategy and user feedback data, including user satisfaction and equipment failure rate. Apply correlation analysis methods, such as the Pearson correlation coefficient, to evaluate the relationship between equipment performance indicators (such as operating efficiency, reliability) and usage patterns (such as frequency of use, operating environment). At the same time, further analyze the specific influencing factors and strength of the relationship to reveal which performance indicators are particularly sensitive to equipment usage patterns, thereby providing data support for subsequent equipment design and usage strategies. The final comprehensive performance analysis results will provide management with a decision-making basis for optimizing equipment configuration and operation management.
[0126] The maintenance decision submodule analyzes the impact of equipment parameters on maintenance needs based on the comprehensive performance analysis results. The parameters include equipment age, usage frequency, and past failure history. It analyzes the factors that have a significant impact on maintenance needs, proposes targeted maintenance plans, and generates customized maintenance strategies.
[0127] First, through statistical analysis methods such as factor analysis and variance analysis, identify the main factors affecting equipment maintenance needs. For variables such as equipment age, frequency of use, and failure history, evaluate the correlation and impact strength between variables and maintenance needs. Based on the analysis results, design specific maintenance activities, including preventive maintenance plans and failure response strategies, as well as more frequent inspections of old equipment. At the same time, customize maintenance strategies to adapt to the specific needs and operating conditions of different types of equipment, aiming to improve equipment reliability and performance while reducing downtime and maintenance costs.
[0128] The recycling strategy submodule analyzes the remaining useful life and recycling potential of IT equipment based on customized maintenance strategies, integrates data, redistributes or updates old equipment, and generates a full life cycle management plan;
[0129] First, evaluate the remaining useful life of the equipment, use predictive models such as remaining life analysis (RLA), combine equipment maintenance records and usage data to estimate and analyze the recycling potential of the equipment, including the identification of renewable components and recycling cost-effectiveness analysis. Integrate data to develop targeted recycling strategies, such as reconfiguring some old equipment to an operating environment with lower demand, or making necessary upgrades to extend the life of the equipment. Through this approach, optimize the overall efficiency and cost-effectiveness of the equipment while supporting environmental sustainability goals. Ultimately, the generated life cycle management plan will provide a comprehensive management framework for the procurement, use, maintenance and retirement of equipment.
[0130] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. A time-sharing leasing solution system for IT equipment, It is characterized in that The system comprises: The user cognitive analysis module analyzes user preferences and sensitivities based on user interaction data, identifies differentiated user groups, and generates user cognitive bias records; The market demand analysis module analyzes the current market capacity and potential growth space based on the user cognitive bias record, evaluates the market demand elasticity and the feasibility of the pricing strategy, and generates a market demand analysis record; The pricing strategy optimization module adjusts the pricing strategy based on the market demand analysis record, simulates the impact of the differentiated pricing scheme on the market response, and generates an optimized pricing scheme; The event monitoring and processing module captures events in real time based on online user behavior and market dynamics, analyzes the impact of rental demand and user behavior, and generates event response records; The behavior trend prediction module classifies and analyzes the user behavior data based on the event response records to generate a behavior trend prediction result; The resource allocation adjustment module optimizes the allocation of available IT equipment rental resources based on the behavior trend prediction results and generates a resource adjustment plan; The fault response optimization module optimizes the fault handling process and response strategy based on equipment monitoring data and historical fault records, and generates a fault response optimization strategy; The full life cycle management module comprehensively evaluates the equipment performance based on the optimized pricing scheme, resource adjustment scheme, fault response optimization strategy and user feedback, and generates a full life cycle management plan; The behavior trend prediction module includes: The data classification submodule groups the user behavior data based on the event response records, distinguishes the user behavior patterns, and generates user behavior grouping results; The trend analysis submodule analyzes the change trends of different user behavior groups based on the user behavior group division results and generates trend analysis results; The trend prediction submodule uses the ARIMA model based on the trend analysis results to train the model according to the historical user behavior trend data, predict future user behavior and rental demand trends, and generate behavior trend prediction results; The ARIMA model, according to the formula: ; Calculate future user behavior and rental demand trends , generate behavioral trend prediction results, where is a constant term, , is the coefficient of the autoregressive term, which indicates the impact of the previous one or two periods of the time series on the current value. , Respectively represent the values of the time series after difference in the previous period and the previous two periods, , is the coefficient of the moving average term, , Represent the random error terms of the time series in the previous period and the previous two periods, respectively. is the random error term at the current time point, is the seasonal influence coefficient, is the weight coefficient of seasonal impact, is the industry dynamics index, is the weight coefficient of industry dynamics, For specific events, is the weight coefficient of the impact of a specific event, is the technological progress adjustment coefficient, is the weight coefficient of the impact of technological progress; The resource allocation adjustment module includes: The resource assessment submodule analyzes the usage, maintenance history and performance indicators of the equipment based on the behavior trend prediction results, identifies resources with high maintenance requirements and frequent use, and generates comprehensive resource assessment results; The inventory management submodule dynamically calculates inventory levels and replenishment points based on the comprehensive resource evaluation results, analyzes the difference between forecasted demand and existing inventory, adjusts inventory strategies, and generates inventory optimization plans; The dynamic adjustment submodule of the scheme uses a linear programming method based on the inventory optimization scheme to optimize and adjust resource allocation, formulate a resource allocation plan, balance resource utilization and cost-effectiveness, and generate a resource adjustment scheme; The linear programming method is according to the formula: ; Calculate the optimal allocation of each resource , generate a resource adjustment plan, where For the The unit cost of the class resource, For the The amount of class resources allocated, For the Risk adjustment factor for class resources, For the The demand volatility of the resource class, is the weight coefficient for balancing cost and resource utilization, For the The resource requirements of the activity, For the Class resources for The contribution of a type of activity, n represents the type of resource, and m represents the type of activity.
2. The IT equipment time-sharing leasing solution system according to claim 1, Features: The user cognitive bias record includes the user's anchoring effect analysis, risk preference analysis and time preference inconsistency analysis; the market demand analysis record includes the supply and demand balance status, price elasticity analysis and market saturation index; the optimized pricing plan includes the pricing strategy for differentiated time periods and its expected impact on user acceptance and system utilization; the event response record includes the event type, impact scope and potential demand changes; the behavior trend prediction results include future user rental behavior trend prediction and potential market demand change analysis; the resource adjustment plan includes the resource optimized allocation plan, expected inventory level and adjusted response strategy; the fault response optimization strategy includes the accuracy of fault detection, the efficiency of the diagnostic process and the improvement of response measures; the full life cycle management plan includes equipment maintenance plan, upgrade recommendations and reuse strategy.
3. The IT equipment time-sharing leasing solution system according to claim 1, Features: The pricing strategy optimization module includes: The pricing model building submodule builds a pricing model for the rental period based on the market demand analysis record, solves the price setting for maximizing the revenue, and generates a preliminary pricing strategy; The user acceptance evaluation submodule performs regression analysis on user purchase data at different price levels based on the preliminary pricing strategy, calculates price elasticity, evaluates user acceptance, and generates user price sensitivity analysis results; The pricing optimization submodule performs price optimization based on the user price sensitivity analysis results, iteratively optimizes the preliminary pricing strategy, and generates an optimized pricing plan.
4. The IT equipment time-sharing leasing solution system according to claim 1, Features: The event monitoring processing module includes: The event capture submodule filters and converts data based on online user behavior and market dynamics, retains key event data, including login anomalies, transaction operations or major market news, and obtains key event data streams; The event analysis submodule configures a sliding event window based on the key event data stream to aggregate and analyze time series data, identify abnormal behaviors and important market changes, evaluate potential impacts, and generate event impact analysis results; The event record submodule analyzes the event details, including event type, time, affected users and market reactions, based on the event impact analysis results, optimizes data access and query efficiency, and establishes event response records.
Citation Information
Cited By
Asset management and distribution method based on leasing business
CN119444373A
