Equipment leasing process automatic management system and method

Abstract

The invention discloses an equipment lease process automatic management system and method, and belongs to the technical field of equipment lease, and the method comprises the steps: dividing the expected use time into minimum time units, and forming a refined time credibility scoring matrix through combining an equipment return behavior fluctuation index and a regional time efficiency compression index; meanwhile, a buffer time period is automatically inserted after a task is finished, a sliding time window is used for detecting a low-credibility time block and marking a conflict risk, recommendation equipment is further screened through a logic judgment mechanism, a replacement scheduling scheme is provided when necessary, and finally scheduling suggestions are output in a graphical mode. High-precision prediction, high-reliability recommendation and high-visual interaction of equipment scheduling in a multi-lease, cross-time-zone and high-density scene are realized, and the scheduling safety and the system intelligence level are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of equipment leasing technology, and specifically to an automated management system and method for equipment leasing processes. Background Technology

[0002] Equipment leasing process automation management refers to the systematic, standardized and automated management of all aspects of the equipment leasing process, including equipment application, approval, contract signing, equipment delivery, usage monitoring, return and settlement, through information technology means (such as software systems or platforms). This reduces manual operation, improves work efficiency, reduces error rate, and achieves transparent and controllable management of the entire leasing process.

[0003] The existing technology has the following shortcomings: In equipment leasing scheduling systems, scheduling logic may suffer from time overlap misjudgment, which is particularly serious in multi-lease, cross-day, or cross-time zone application scenarios. Existing systems often use a simplified start-end time interval judgment method, ignoring minute-level time granularity, time zone uniformity, and the buffer time requirements between leases, resulting in equipment being repeatedly allocated and used at the scheduling boundary. Summary of the Invention

[0004] The purpose of this invention is to provide an automated management system and method for equipment leasing processes to address the shortcomings in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an automated management method for equipment leasing processes, comprising: Receive rental request information and divide the expected usage period into several minimum time units to generate the corresponding time block sequence; Obtain the equipment return behavior fluctuation index and the regional timeliness compression index, calculate the time reliability score for each time block, and form a time reliability matrix; After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is considered an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Construct a sliding time window, and within the target time period and a preset range before and after it, detect the credibility score of candidate devices in each time block, and mark conflicting time blocks that are below a set threshold; Based on the time reliability score and conflict flag of each candidate device, output a recommended list of devices that meet the scheduling conditions. If there are no available devices, generate alternative scheduling suggestions. The final recommendation results and credibility score data are visualized.

[0006] Preferably, dividing the expected usage period into several minimum time units includes: The system receives rental request information submitted by users, including equipment type, start time, end time, location, and work requirements. Based on equipment category, rental period, historical scheduling density, and industry work rhythm, it dynamically determines the granularity of time units, including but not limited to 15 minutes, 30 minutes, or 1 hour. Using an adaptive time granularity determination algorithm, it outputs a recommended granularity based on historical rental data and equipment usage frequency. It determines whether the time period crosses natural boundaries or work shifts, prioritizing the division into main blocks by shift and then subdividing the time units within each main block. Finally, it generates a time block sequence containing start and end times, the corresponding stage, granularity, tags, and an initial credibility score.

[0007] Preferably, the obtained device return behavior fluctuation index includes: The planned and actual return times of the target equipment from the last N rental records are collected to calculate the return deviation sequence. The return deviation is assumed to follow a Gaussian distribution, and a normal-inverse gamma conjugate prior distribution is introduced. The posterior deviation mean and variance are calculated based on Bayesian inference. The posterior standard deviation is divided by the average rental period to obtain the equipment return behavior volatility index. Where: RVI represents the equipment return behavior volatility index. The standard deviation of the posterior bias. This indicates the average duration of the equipment leasing period.

[0008] Preferably, obtaining the regional timeliness compression index includes: modeling a region as an M / M / 1 single-channel queuing system; statistically analyzing the rental request arrival rate λ and equipment scheduling processing rate μ per unit time; and calculating the regional utilization rate. And using the computational domain time compression index under the condition λ < μ. The expression is: .

[0009] Preferably, the equipment return behavior fluctuation index and the regional timeliness compression index are converted into a comprehensive feature vector. The comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses the prediction of the time credibility score label of each time block of image recognition as the prediction objective, and minimizes the sum of prediction errors of the time credibility score label of each time block of all image recognition as the training objective. The machine learning model is trained until the sum of prediction errors converges and the model training stops. The time credibility score of each time block of image recognition is determined according to the model output. The machine learning model is a multinomial regression model.

[0010] Preferably, the step of inserting a buffer period after the end time of an existing rental task includes: Identify the task completion time of the equipment and set the buffer time length according to the equipment type; set a 2-hour buffer for heavy equipment, a 30-minute buffer for light equipment, and a cross-day buffer segment for high-frequency equipment; generate time blocks corresponding to the buffer time period and mark their status as unrentable; force the time reliability score of the buffer time block to be set to the system's preset minimum value.

[0011] Preferably, the construction of the sliding time window and detection of conflicting time blocks includes: A sliding time window is formed by expanding a fixed time length forward and backward from the user's application time period as the center; the time credibility scores of all candidate devices within the time window are scanned; and time blocks with scores lower than a set threshold are marked as conflict time blocks.

[0012] Preferably, the device for outputting recommendations based on scores and conflict markers includes: Filter all candidate devices and determine whether their credibility scores for all time blocks within the application period are higher than a set threshold; exclude devices with conflicting time blocks or scores below the tolerable range; for devices with scores in the critical range, trigger a manual review or risk warning mechanism; prioritize devices that meet the scheduling conditions according to their average credibility, distance from the usage location, historical success rate, and user preferences, and output a recommendation list.

[0013] Preferably, if none of the candidate devices meet the scheduling conditions, the generation of alternative scheduling suggestions includes: The original time period is shifted backward to set a step size to reassess future available time blocks; the time period that first meets the scheduling threshold is found and a postponement suggestion is generated; if the time cannot be postponed, device models with compatible parameters or similar functions are found and device replacement suggestions are proposed; if none of these are feasible, a scheduling failure message is output, and the user is advised to change the time period or split the task.

[0014] The present invention also provides an automated management system for equipment leasing processes, including a time parsing module, a credibility scoring module, a scheduling buffer management module, a conflict detection module, an intelligent scheduling recommendation module, and a scheduling visualization module; Time parsing module: Receives rental request information, divides the expected usage period into several minimum time units, and generates corresponding time block sequences; Credibility scoring module: Obtains the device return behavior fluctuation index and regional timeliness compression index, calculates the time credibility score for each time block, and forms a time credibility matrix; Scheduling buffer management module: After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is regarded as an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Conflict detection module: Constructs a sliding time window to detect the credibility score of candidate devices in each time block within the target time period and a preset range before and after it, and marks conflict time blocks that are below a set threshold; Intelligent scheduling recommendation module: Based on the time reliability score and conflict marker of each candidate device, outputs a recommended list of devices that meet the scheduling conditions. If there are no available devices, it generates alternative scheduling suggestions. Scheduling visualization module: Visualizes the final recommendation results and credibility score data.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention constructs fine-grained time block sequences and combines equipment return behavior fluctuation indices and regional timeliness compression indices for reliability modeling. This enables accurate prediction of the deployability probability of equipment within each time unit, overcoming the coarse-grained judgment of start-end time intervals in traditional scheduling methods. It is particularly suitable for complex application scenarios such as multi-lease, cross-day, and cross-time zone scenarios. Simultaneously, the automatically inserted buffer time period mechanism effectively blocks the risk of time overlap between consecutive scheduling, significantly improving the fault tolerance of scheduling boundaries and ensuring the executability and stability of scheduling results.

[0016] 2. This invention introduces a sliding time window detection mechanism and an intelligent scheduling recommendation strategy. Based on the prediction of conflict time blocks, it enables adjustable equipment screening, sorting, and generation of alternative suggestions, further enhancing the system's scheduling flexibility and risk control capabilities. Combining the credibility score prediction of a multinomial regression model with a graphical user interface, the system not only achieves intelligent output of equipment scheduling results but also improves users' understanding and trust in the system's recommendation logic. Overall, it realizes the intelligent transformation of the equipment leasing process from rule-based scheduling to data-driven scheduling. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a mind map of the method of the present invention.

[0019] Figure 2 This is a mind map of the system modules of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, please refer to Figure 1 As shown in this embodiment, an automated management method for equipment leasing processes includes: Receive rental request information and divide the expected usage period into several minimum time units to generate the corresponding time block sequence; Obtain the equipment return behavior fluctuation index and the regional timeliness compression index, calculate the time reliability score for each time block, and form a time reliability matrix; After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is considered an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Construct a sliding time window, and within the target time period and a preset range before and after it, detect the credibility score of candidate devices in each time block, and mark conflicting time blocks that are below a set threshold; Based on the time reliability score and conflict flag of each candidate device, output a recommended list of devices that meet the scheduling conditions. If there are no available devices, generate alternative scheduling suggestions. The final recommendation results and credibility score data are visualized.

[0022] Receive rental request information submitted by users, including: equipment type and quantity; expected start and end time of use; location and type of operation; and any special requirements (such as day and night operation, power requirements, etc.).

[0023] Standardize time information: convert start and end times to UTC format or use a unified reference time zone; if scheduling is across time zones, mark the original time zone code of each time segment.

[0024] The time unit granularity (i.e., the length of a single time block) is dynamically determined based on the following factors: Equipment category (e.g., heavy hoisting equipment is recommended to be granular at 1 hour, while short-term rental electrical equipment can be fined down to 15 minutes); user-specified task cycle length; historical scheduling density (granularity is recommended for high-density scenarios); industry characteristics of the project (e.g., when the construction industry has a significant day-night rhythm, the granularity needs to match the work shifts).

[0025] An adaptive time granularity determination algorithm is used to output a recommended granularity based on the input information and the device's historical data. The granularity can be a fixed length (e.g., 30 minutes, 1 hour) or a non-uniform division (e.g., a 1-hour work period and a 2-hour rest period).

[0026] Determine whether the requested time period crosses natural boundaries (such as the day boundary or shift handover); if it crosses shifts (such as 08:00–16:00, 16:00–00:00, 00:00–08:00), the system prioritizes dividing the time into main blocks according to shifts; each main block is further subdivided into the smallest time units.

[0027] A hierarchical time division model is adopted: the first layer is divided into "time phases" (such as day / night, shifts); the second layer is further subdivided into "micro-time blocks" within each phase; it can dynamically adapt to industry standards or customer-defined shift configurations.

[0028] If the target time period includes public holidays or nighttime restricted working hours, the system will automatically mark that time range as a "non-recommended time block"; it can connect to the national statutory holiday API or the enterprise's custom operation calendar. Based on the impact of weather / construction environment (if IoT or third-party environmental data is connected): if extreme weather is expected during the time period, the corresponding time block will be marked as a "risk time block".

[0029] Based on the above rules, the entire expected usage period is divided into an ordered sequence of time blocks, with the following structure: Each time block contains the following attribute fields: start time; end time; stage (e.g., shift number, date); minimum time unit granularity; special tags (e.g., holidays, risk periods, nighttime, environmental restrictions); initial credibility score (default value or inherited from the previous stage); time zone identifier (used for subsequent scheduling matching and equipment location coordination).

[0030] The system acquires the equipment return behavior fluctuation index and the regional timeliness compression index, calculates the time reliability score for each time block, and forms a time reliability matrix, which specifically includes: The method for obtaining the equipment return behavior fluctuation index is as follows: For target device d, obtain its most recent N rental records: Indicates the planned return time for the i-th lease; Indicates the actual return time of the i-th rental; This represents the return time deviation (in hours or minutes, can be positive or negative); forming a deviation sample sequence: ; Assume the return bias follows a Gaussian distribution: ; Set the prior to a conjugate prior distribution (commonly the normal-inverse gamma prior): , ;in: Indicates the empirical average deviation (e.g., the whole system average); This indicates the uncertainty of the prior mean; This represents a priori estimate of the degree of volatility.

[0031] Given sample data Δ, update the posterior parameters according to the Bayesian inference formula (derivation omitted from the standard formula), and calculate the posterior mean: Posterior variance estimation (simplified form): ;in: The sample mean representing historical bias; This represents the posterior fluctuation estimate; This indicates the prior mean deviation, which can be set as the average deviation of the system or similar equipment. This represents the prior uncertainty; the larger the value, the lower the confidence in the prior mean. The return behavior volatility index is defined as the standardized posterior bias per unit time, expressed as: Where: RVI represents the equipment return behavior volatility index. The standard deviation of the posterior bias. This indicates the average duration of the equipment leasing period (unit as above, to prevent the amplification of the impact of short-term leasing deviations); if When the value is small, a minimum threshold can be set to avoid division by zero.

[0032] The method for obtaining the regional timeliness compression index is as follows: Assume that equipment rental requests in a certain region Z can be abstracted as a single-channel queuing system, denoted as an M / M / 1 model, including the following assumptions: Request arrival follows a Poisson distribution (an exponential distribution of interval time); service time (equipment schedulable processing time) follows an exponential distribution; a single service channel or equivalent service capacity is aggregated into one channel.

[0033] Recent data for region Z is obtained from the scheduling system, with a commonly used sliding time window (e.g., the last 7 days): λ represents the rental request arrival rate per unit time (e.g., the number of rental tasks submitted per hour); μ represents the equipment scheduling processing capacity per unit time (e.g., the number of scheduling tasks completed or the number of successfully released devices per hour); ρ=λ / μ: is called the region utilization rate.

[0034] The expression for calculating the Regional Time Compression Index (RTCR) is as follows: (Prerequisite: λ < μ, system stable); If λ ≥ μ, the system enters saturation, and RTCR is set to the maximum threshold or marked as a high-voltage area. When λ approaches μ, RTCR approaches infinity, indicating that regional scheduling is strained and equipment availability is extremely unreliable; an upper limit pruning threshold RTCRmax can be set to avoid model instability.

[0035] The equipment return behavior fluctuation index and the regional timeliness compression index are converted into a comprehensive feature vector. The comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses the prediction of the time credibility score label of each time block of image recognition as the prediction objective, and minimizes the sum of prediction errors of the time credibility score label of each time block of all image recognition as the training objective. The machine learning model is trained until the sum of prediction errors converges and the model training stops. The time credibility score of each time block of image recognition is determined according to the model output. The machine learning model is a multinomial regression model.

[0036] The trained machine learning model is used to predict the credibility of all devices and time blocks, resulting in a complete two-dimensional credibility score matrix (device × time block); the value range is usually [0,1], and it is visualized as a heatmap for scheduling reference; it can also be used in automatic scheduling systems to select the optimal available time period.

[0037] In this invention, in order to solve the problems that may occur in the continuous scheduling of equipment, such as return delays, unclear status, and incomplete allocation, and thus effectively prevent scheduling conflicts, equipment unavailability or work interruption caused by overly tight scheduling, the system automatically inserts a buffer period after the end time of each equipment rental task. This period is regarded by the system as an exclusive time segment that cannot be used for subsequent scheduling.

[0038] Specifically, after a device completes a rental task scheduling in the scheduling system, the system automatically matches a standardized minimum buffer duration based on the device's category or model, for example: For heavy lifting equipment, the buffer time can be set to 2 hours for unloading, transportation and maintenance. For mobile power generation equipment, the buffer time can be as short as 30 minutes, which is used for operations such as cable power outage and parameter reset; For high-frequency equipment that requires regular maintenance, a day-to-day buffer can be inserted.

[0039] The buffer period can be statically configured (such as system preset) or dynamically generated (such as automatically adjusted according to equipment usage frequency and fault records), and recorded as a special type of system task, marked as unleasable.

[0040] To ensure the accuracy of credibility modeling during the scheduling process, the system sets a preset minimum value (e.g., 0 or 0.01) for the time credibility score of all time blocks within the buffer segment when generating the time block sequence. Regardless of whether other conditions meet the scheduling requirements, the system will not recommend or match any equipment resources within that time block. This score, as part of the scheduling boundary anti-collision mechanism, serves as a clear scheduling prohibition signal in the equipment availability matrix, scheduling heatmap, and AI scoring model.

[0041] In addition, this buffering mechanism has the following additional functions: Visual presentation: In the scheduling front-end interface, buffer time blocks are marked with gray, red or diagonal lines, so that the scheduler can clearly identify unavailable time periods; Configurability support: Enterprise users can customize buffering strategies based on factors such as device category, customer type, and project level; Data-driven optimization: The system can dynamically adjust the buffer time based on historical scheduling failure rates or return delay probabilities.

[0042] This automatic insertion mechanism for buffer periods effectively improves the scheduling system's fault tolerance to scheduling boundary uncertainties, enhances the security of equipment allocation and the executability of scheduling plans, and reduces multi-tenant conflicts and lease failure rates. It has significant technical value for achieving intelligent and highly reliable automated equipment leasing.

[0043] In this invention, in order to improve the system's ability to predict equipment scheduling conflicts in advance and prevent potential scheduling failures caused by imprecise time granularity or uneven resource allocation, the system designs a sliding time window conflict detection mechanism to identify time blocks with insufficient credibility during the scheduling calculation stage, and dynamically adjust the equipment recommendation results or provide risk warnings accordingly.

[0044] Around the rental period requested by the user, a sliding time window is constructed, centered on the target time period and extending forward and backward. This time window may include: the start and end time periods reserved by the user; buffer time periods extending forward and backward (e.g., each extended by 1 hour, 2 hours, or user-defined duration); the time window as a whole is divided into the smallest time block units consistent with the system (e.g., 30 minutes or 15 minutes). For each time block within this time window, the system performs the following operations: Obtain all devices that may be used for scheduling this task; Based on the aforementioned time reliability scoring mechanism, the score value of each device in each time block is obtained; The score of each time block is compared with a preset threshold (such as 0.4 or 0.5). If it is lower than the preset threshold, it is marked as a conflicting time block. The detected conflict time blocks are used as conflict risk data for use by the following modules: Schedule recommendation engine: Automatically avoids conflicting time blocks; Visual interface: Displays conflicting time blocks using methods such as color enhancement, diagonal lines, and layer masking to help dispatchers quickly identify them; Log archiving system: Records conflict detection results, supporting subsequent scheduling optimization and behavior learning.

[0045] In addition, to enhance adaptability, this sliding time window can be configured with the following capabilities: Dynamic window range adjustment: The sliding range length is adaptively determined based on device type or task urgency; Personalized scoring threshold settings: Different types of equipment, different customers, and different project levels can use different credibility scoring lower limits; Parallel batch detection mechanism: For a large number of devices or dense time periods, concurrent detection can be enabled to improve efficiency.

[0046] Through the aforementioned sliding window mechanism, the system can not only determine whether equipment is available for allocation within the user's requested time period, but also perceive the scheduling risks in adjacent time periods, achieving "proactive" scheduling risk management. Especially in multi-tenancy, cross-scheduling, and high-frequency call scenarios, this mechanism can significantly reduce equipment conflict scheduling problems caused by boundary overlap, misjudgment of credibility, etc., and is a key supporting link for the intelligence and robustness of equipment leasing scheduling.

[0047] In this invention, after completing the device time reliability score and conflict time block marking, the system generates the final scheduling recommendation result for the user.

[0048] The system retrieves a complete set of candidate devices that match the rental application. Matching criteria include, but are not limited to: device model, geographical location, current status as "schedulable", maintenance qualification, and availability of inventory.

[0049] For each candidate device, the system finds its time credibility score on the time block requested by the user and performs a joint evaluation based on whether the time block is marked as conflicting.

[0050] The system determines device schedulability using the following logic: Condition A: The credibility score of all application time blocks is not lower than the set threshold (e.g., 0.6). Condition B: No time blocks are marked as conflicting time blocks; Condition C: If some time blocks are below the threshold but within a tolerable range (e.g., 0.5~0.6), a manual confirmation mechanism or risk warning mechanism can be triggered; The system will only determine a device as a schedulable device if conditions A and B are met simultaneously, or if conditions A and C are met.

[0051] All devices that meet the above judgment logic are prioritized according to the following dimensions, and a device recommendation list is output: average time reliability (from high to low); distance of the device's current geographical location from the place of use (from near to far); historical scheduling success rate; customer preference records for the device or model.

[0052] If the system determines that no candidate device meets the scheduling conditions, that is, the time block confidence score of all candidate devices is lower than the threshold; or the time block is in a conflict state; or all schedulable devices are excluded; The system will then automatically enter the alternative scheduling suggestion generation module and perform the following steps: The system shifts the original application time period forward by a sliding window in increments of 5 or 15 minutes, recalculates the credibility score within the future time range, identifies the time period that first meets the scheduling threshold, and provides the following suggestions: If the time cannot be postponed, the system may attempt to find compatible devices, i.e., other models of devices with similar parameters that support the same type of task, as alternatives, and note the results: If none of these conditions can be met, the system will output the following message: The current request cannot meet the scheduling conditions within the specified time period. It is recommended to change the time or split the task and have it executed by multiple devices in a time-sharing manner.

[0053] In this invention, to improve scheduling transparency and system interpretability, the system outputs a list of recommended device scheduling, and graphically presents the time reliability score of each candidate device in the time block it covers. This information is then combined with information such as recommendation priority, conflicting time blocks, and alternative suggestions to generate a visual results interface.

[0054] Construct a device time reliability heatmap, where rows represent candidate device numbers; columns represent continuous time blocks (e.g., 30-minute granularity); each cell's color represents a reliability score (color transitions from red to yellow to green); dark red (0~0.3): high risk, unavailable; yellow (0.3~0.6): low reliability, requires careful evaluation; green (0.6~1.0): high reliability, recommended time period; conflicting time blocks will be highlighted with diagonal lines, borders, or masks.

[0055] Construct a bar chart to prioritize device recommendations, displaying the overall credibility score (e.g., average over a time period) of the top N recommended devices; each bar represents a device, and the color is synchronized with the heatmap.

[0056] The scheduling results preview panel displays key metrics for each device's recommended solution: start and end time periods; overall reliability score; presence of buffer zone interference; alternative suggestion indicator (if device downgrade recommendation exists); hovering the mouse over the device displays detailed information, such as: "Recommendation reason is the lowest conflict rate in the region".

[0057] The visual behavioral interaction description includes: Click on the device row: This will expand the scoring trend chart (such as a line chart) for the device's time block. Double-click the conflict time block: a pop-up will appear with details of the scoring reason (e.g., low RTCR, high RVI, holiday period). Select the "Show only high credibility segments" option: filter devices with time block scores higher than the set threshold; supports exporting visualized data as PDF or images for project review and filing.

[0058] Example 2, please refer to Figure 2 As shown in this embodiment, an automated equipment leasing process management system includes a time parsing module, a credibility scoring module, a scheduling buffer management module, a conflict detection module, an intelligent scheduling recommendation module, and a scheduling visualization module. Time parsing module: Receives rental request information, divides the expected usage period into several minimum time units, and generates corresponding time block sequences; Credibility scoring module: Obtains the device return behavior fluctuation index and regional timeliness compression index, calculates the time credibility score for each time block, and forms a time credibility matrix; Scheduling buffer management module: After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is regarded as an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Conflict detection module: Constructs a sliding time window to detect the credibility score of candidate devices in each time block within the target time period and a preset range before and after it, and marks conflict time blocks that are below a set threshold; Intelligent scheduling recommendation module: Based on the time reliability score and conflict marker of each candidate device, outputs a recommended list of devices that meet the scheduling conditions. If there are no available devices, it generates alternative scheduling suggestions. Scheduling visualization module: Visualizes the final recommendation results and credibility score data.

[0059] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0060] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0061] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0062] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for automating the management of equipment leasing processes, characterized in that: include: Receive rental request information and divide the expected usage period into several minimum time units to generate the corresponding time block sequence; Obtain the equipment return behavior fluctuation index and the regional timeliness compression index, calculate the time reliability score for each time block, and form a time reliability matrix; After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is considered an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Construct a sliding time window, within the target time period and a preset range before and after it, detect the credibility score of candidate devices in each time block, and mark conflicting time blocks that are below a set threshold; Based on the time reliability score and conflict flag of each candidate device, output a recommended list of devices that meet the scheduling conditions. If there are no available devices, generate alternative scheduling suggestions. The final recommendation results and credibility score data are visualized.

2. The automated management method for equipment leasing process according to claim 1, characterized in that: The division of the expected usage period into several minimum time units includes: The system receives rental request information submitted by users, including equipment type, start time, end time, location, and work requirements. Based on equipment category, rental period, historical scheduling density, and industry work rhythm, it dynamically determines the granularity of time units, including but not limited to 15 minutes, 30 minutes, or 1 hour. Using an adaptive time granularity determination algorithm, it outputs a recommended granularity based on historical rental data and equipment usage frequency. It determines whether the time period crosses natural boundaries or work shifts, prioritizing the division into main blocks by shift and then subdividing the time units within each main block. Finally, it generates a time block sequence containing start and end times, the corresponding stage, granularity, tags, and an initial credibility score.

3. The automated management method for equipment leasing process according to claim 1, characterized in that: The fluctuation index of the acquired device return behavior includes: The planned and actual return times of the target equipment in the last N rental records are collected to calculate the return deviation sequence. The return deviation is assumed to follow a Gaussian distribution, and a normal-inverse gamma conjugate prior distribution is introduced. The posterior deviation mean and variance are calculated based on Bayesian inference. The posterior standard deviation is divided by the average rental period to obtain the equipment return behavior volatility index. Where: RVI represents the equipment return behavior volatility index. The standard deviation of the posterior bias. This indicates the average duration of the equipment leasing period.

4. The automated management method for equipment leasing process according to claim 3, characterized in that: The acquisition of the regional timeliness compression index includes: modeling a certain region as an M / M / 1 single-channel queuing system; statistically analyzing the rental request arrival rate λ and equipment scheduling processing rate μ per unit time; and calculating the regional utilization rate. And using the computational domain time compression index under the condition λ < μ. The expression is: .

5. The automated management method for equipment leasing process according to claim 4, characterized in that: The equipment return behavior fluctuation index and the regional timeliness compression index are converted into a comprehensive feature vector. The comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses the prediction of the time credibility score label of each time block of image recognition as the prediction objective, and minimizes the sum of prediction errors of the time credibility score label of each time block of all image recognition as the training objective. The machine learning model is trained until the sum of prediction errors converges and the model training stops. The time credibility score of each time block of image recognition is determined according to the model output. The machine learning model is a multinomial regression model.

6. The automated management method for equipment leasing process according to claim 5, characterized in that: The insertion of a buffer period after the end time of an existing rental task includes: Identify the task completion time of the equipment and set the buffer time length according to the equipment type; set a 2-hour buffer for heavy equipment, a 30-minute buffer for light equipment, and a cross-day buffer segment for high-frequency equipment; generate time blocks corresponding to the buffer time period and mark their status as unrentable; force the time reliability score of the buffer time block to be set to the system's preset minimum value.

7. The automated management method for equipment leasing process according to claim 6, characterized in that: The process of constructing a sliding time window and detecting conflicting time blocks includes: A sliding time window is formed by expanding a fixed time length forward and backward from the user's application time period as the center; the time credibility scores of all candidate devices within the time window are scanned; and time blocks with scores lower than a set threshold are marked as conflict time blocks.

8. The automated management method for equipment leasing process according to claim 7, characterized in that: The device that outputs recommendations based on scores and conflict markers includes: Filter all candidate devices and determine whether their credibility scores for all time blocks within the application period are higher than a set threshold; exclude devices with conflicting time blocks or scores below the tolerable range; for devices with scores in the critical range, trigger a manual review or risk warning mechanism; prioritize devices that meet the scheduling conditions according to their average credibility, distance from the usage location, historical success rate, and user preferences, and output a recommendation list.

9. The automated management method for equipment leasing process according to claim 8, characterized in that: If none of the candidate devices meet the scheduling conditions, the proposed alternative scheduling includes: The original time period is shifted backward to set a step size to reassess future available time blocks; the time period that first meets the scheduling threshold is found and a postponement suggestion is generated; if the time cannot be postponed, device models with compatible parameters or similar functions are found and device replacement suggestions are proposed; if none of these are feasible, a scheduling failure message is output, and the user is advised to change the time period or split the task.

10. An automated management system for equipment leasing processes, used to implement the automated management method for equipment leasing processes as described in any one of claims 1-9, characterized in that: It includes a time parsing module, a credibility scoring module, a scheduling buffer management module, a conflict detection module, an intelligent scheduling recommendation module, and a scheduling visualization module; Time parsing module: Receives rental request information, divides the expected usage period into several minimum time units, and generates corresponding time block sequences; Credibility scoring module: Obtains the device return behavior fluctuation index and regional timeliness compression index, calculates the time credibility score for each time block, and forms a time credibility matrix; Scheduling buffer management module: After the end time of an existing rental task, a buffer period is automatically inserted based on the equipment type. The buffer period is regarded as an unrentable period, and its corresponding time reliability score is forcibly set to a preset minimum value. Conflict detection module: Constructs a sliding time window to detect the credibility score of candidate devices in each time block within the target time period and a preset range before and after it, and marks conflict time blocks that are below a set threshold; Intelligent scheduling recommendation module: Based on the time reliability score and conflict marker of each candidate device, outputs a recommended list of devices that meet the scheduling conditions. If there are no available devices, it generates alternative scheduling suggestions. Scheduling visualization module: Visualizes the final recommendation results and credibility score data.

Images