A resource dynamic allocation and risk early warning method and system of a material leasing platform

Abstract

The present application relates to material leasing management technical field, especially material leasing platform's resource dynamic allocation and risk early warning method and system. The method comprises the following steps: extracting material use parameters, abnormal work order in lease period and corresponding sequence of return damage result from historical turnover records of material leasing platform, forming frequency parameter table; when each material check is completed, calculating quality correction coefficient according to the completion time length and attachment integrity of this check work order, resetting the credibility value of the material to the reset initial value with the quality correction coefficient as the upper limit; after the material enters the lease use, the reset initial value is continuously attenuated according to time step based on the frequency parameter table to obtain real-time credibility value. The present application dynamically evaluates the confidence degree of material state information through the dynamic evaluation system, simultaneously drives allocation decision and segmented damage attribution, realizes the decoupling of book inventory and actual allocable resources and the fine attribution of damage responsibility.

Description

Technical Field

[0001] This invention relates to the field of material leasing management technology, and in particular to a method and system for dynamic resource allocation and risk warning of a material leasing platform. Background Technology

[0002] In industries such as construction, exhibitions, and municipal engineering, material rental platforms are widely used as information management systems connecting lessors and lessees. Existing material rental platforms typically include modules for material ledger management, rental contract management, inventory status inquiry, and tenant information management. By recording the inbound and outbound status of materials, contract duration, and return status, they provide platform operators with basic resource management capabilities. Regarding resource allocation, existing platforms usually generate allocation plans based on the book inventory quantity. When a tenant submits a rental application, the system queries the currently available inventory and responds accordingly. In terms of damage risk management, existing platforms typically require manual verification of material status upon return, recording the verification results in the system work order as the basis for damage assessment and liability attribution.

[0003] However, after materials have undergone varying degrees of rental use, the discrepancy between their actual condition and the condition described in the verification records continues to accumulate. Existing technologies lack a quantitative mechanism for this, resulting in distorted material condition information on which allocation decisions are based, and a systematic deviation between the actual allocable quantity and the book inventory. Summary of the Invention

[0004] Based on this, it is necessary for the present invention to provide a method and system for dynamic resource allocation and risk warning of a material rental platform, so as to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a method for dynamic resource allocation and risk warning in a material leasing platform includes the following steps: Step S1: Extract the material usage parameters, the corresponding sequence of abnormal work orders during the rental period and the return damage results from the historical circulation records of the material rental platform to form a frequency parameter table; Step S2: When each material verification is completed, calculate the quality correction coefficient based on the completion time of the verification work order and the completeness of the attachments, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. Step S3: After the materials are put into use for rental, the initial value is continuously decayed according to the time step based on the frequency parameter table to obtain the real-time reliability value; the responsibility segments are divided with the arrival time of abnormal work orders during the rental period as nodes, and the real-time reliability value at the beginning of each segment is recorded as the segment benchmark value. Step S4: Compare the real-time reliability value with the preset threshold, determine the material's eligibility for allocation based on the comparison result, and trigger material verification when the real-time reliability value is lower than the preset threshold; Step S5: When damage occurs during material return, the damage is assigned to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type, and a handling notice or a questionable work order is generated respectively.

[0006] This invention also provides a resource dynamic allocation and risk warning system for a material leasing platform, used to execute the resource dynamic allocation and risk warning method for a material leasing platform as described above. The resource dynamic allocation and risk warning system for the material leasing platform includes: The frequency parameter construction module is used to extract material usage parameters, the corresponding sequence of abnormal work orders and return damage results from the historical circulation records of the material rental platform, and form a frequency parameter table. The credibility reset module is used to calculate the quality correction coefficient based on the completion time of the current verification work order and the completeness of the attachments when each material verification is completed, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. The credibility decay module is used to continuously decay the reset initial value according to the time step based on the frequency parameter table after the material enters the rental period to obtain the real-time credibility value; the responsibility segment is divided with the arrival time of abnormal work orders during the rental period as the node, and the real-time credibility value at the beginning of each segment is recorded as the segment benchmark value. The allocation eligibility determination module is used to compare the real-time reliability value with a preset threshold, determine the allocation eligibility of materials based on the comparison result, and trigger material verification when the real-time reliability value is lower than the preset threshold. The damage attribution module is used to assign damage to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type when damage occurs during material return, and generate a disposal notice or a questionable work order respectively.

[0007] This invention introduces a dynamic state quantity called a reliability value, transforming the determination of material allocation eligibility from a static book inventory quantity to a dynamic assessment based on information reliability. This fundamentally solves the long-standing systematic discrepancy between book inventory and actual allocable resources in existing technologies. The reliability value starts with a verification quality correction coefficient and continuously decreases with material usage intensity and time, accurately reflecting the gradual weakening of the verification records' ability to describe the current state of the material. This ensures that the information upon which allocation decisions are based remains within a quantifiable confidence range, rather than blindly trusting potentially outdated historical verification conclusions.

[0008] The automatic construction mechanism of the frequency parameter table enables the platform to continuously improve its ability to recognize the probability of damage under different material types and usage intensities as historical data accumulates. The cross-lease term correction mechanism for material usage parameters incorporates the systematic overstay tendency implied in the tenant's historical behavior into the intensity assessment of the current lease term, making the prediction of damage risk more realistic and avoiding the risk underestimation caused by ignoring the patterns of tenant's historical behavior in existing technologies. The introduction of damage duration offset further reveals a pattern that has never been utilized in existing technologies: the directional difference in the probability of damage for different material types under overstay and early return scenarios. This ensures that the adjustment of the attenuation rate truly reflects the physical characteristics of each material type, rather than applying a uniform attenuation rule to all materials.

[0009] The responsibility segmentation mechanism refines the previously coarse attribution of damage over the entire lease period to the specific usage stage. This allows damage attribution judgment to consider three dimensions of information simultaneously: the probability of historical damage types in each segment, the segment's credibility level, and the deviation from actual usage intensity, significantly improving the accuracy of damage attribution. The introduction of an intensity underestimation marker enables the system to identify situations where the frequency parameter table underestimates the usage intensity of a specific segment and corrects this in the damage attribution weight calculation. Simultaneously, the deviation information is fed back to the frequency parameter table to trigger a local update, forming a self-optimizing loop where the parameter table continuously converges from the lease period level to the segment level.

[0010] The introduction of batch-level material part number distribution dispersion enables the system to, for the first time, distinguish between usage scenario damage and damage during platform self-management at the information level. The causes of damage corresponding to the two forms of low-confidence value material part numbers, concentrated distribution and uniform dispersion, are fundamentally different in physics. The former indicates that adjacent stored materials are affected by the same external conditions, while the latter indicates that randomly selected materials are independently subjected to different usage intensities. The system transforms this morphological difference into an automatic judgment of attribution direction, avoiding the simplistic treatment in existing technologies that attribute all batch damage to the user tenant. Attached Figure Description

[0011] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the steps of a method for dynamic resource allocation and risk warning in a material rental platform according to the present invention. Figure 2 This is a schematic diagram of the credibility value decay timeline in one embodiment; Figure 3 This is a schematic diagram illustrating the responsibility segmentation and damage attribution output of one embodiment; Figure 4 This is a schematic diagram of the module of a material rental platform's dynamic resource allocation and risk warning system according to the present invention. Detailed Implementation

[0012] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0013] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0014] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0015] To achieve the above objectives, please refer to Figures 1 to 4 This invention provides a method for dynamic resource allocation and risk warning in a material leasing platform, the method comprising the following steps: Step S1: Extract the material usage parameters, the corresponding sequence of abnormal work orders during the rental period and the return damage results from the historical circulation records of the material rental platform to form a frequency parameter table; Step S2: When each material verification is completed, calculate the quality correction coefficient based on the completion time of the verification work order and the completeness of the attachments, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. Step S3: After the materials are put into use for rental, the initial value is continuously decayed according to the time step based on the frequency parameter table to obtain the real-time reliability value; the responsibility segments are divided with the arrival time of abnormal work orders during the rental period as nodes, and the real-time reliability value at the beginning of each segment is recorded as the segment benchmark value. Step S4: Compare the real-time reliability value with the preset threshold, determine the material's eligibility for allocation based on the comparison result, and trigger material verification when the real-time reliability value is lower than the preset threshold; Step S5: When damage occurs during material return, the damage is assigned to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type, and a handling notice or a questionable work order is generated respectively.

[0016] Furthermore, step S1 includes the following steps: Step S11: Extract the material number, tenant industry type, actual usage duration, agreed usage duration, abnormal work orders during the rental period, and return damage results for each rental record from the historical circulation records of the material rental platform, and align them one by one according to the rental record number to form a rental parameter sequence; In one embodiment, the system reads completed lease records one by one from the historical flow records, extracts six fields from each record, and stores them in alignment with the lease record number as an index. Among them, the return damage result field records the damage assessment conclusion when the material is returned, and the abnormal work order field records all abnormal events actively reported by the tenant during the lease period and their submission time.

[0017] For example, the rental record numbered R1025 corresponds to material number M003, the tenant's industry type is construction, the actual usage duration is 28 days, the agreed usage duration is 21 days, there are 2 abnormal work orders during the rental period, and the return damage result is that it exceeds the preset damage expectation. The above six fields are stored in the rental parameter sequence with R1025 as the index.

[0018] Step S12: Take the tenant industry type mapping value, the ratio of actual usage time to agreed usage time, and the number of abnormal work orders during the lease period from the lease parameter sequence, and sum them according to the preset weights to obtain the initial value of the material usage parameters; In one embodiment, three fields are extracted from the rental parameter sequence and weighted summed: the tenant industry type is converted into a numerical value according to a preset mapping table to obtain the industry intensity score; the ratio of actual usage time to agreed usage time is calculated to obtain the over-use ratio; the number of abnormal work orders during the rental period is directly obtained; and the initial value of the material usage parameters for the rental record is obtained by weighting and summing the three values ​​according to preset weights.

[0019] For example, in a certain record, the industry intensity score is 0.8, the time overtime ratio is 1.33, and the number of abnormal work orders is 2. By weighting and summing the values ​​with weights of 0.3, 0.4, and 0.3, the initial value of the material usage parameter is 0.8×0.3+1.33×0.4+2×0.3=1.172.

[0020] Step S13: Extract the subsequences in the rental parameter sequence where the return damage results exceed the preset damage expectation, and calculate the upper limit of correction based on the difference between the mean of the subsequence and the mean of the whole sequence.

[0021] In one embodiment, a subset of records in which the return damage result exceeds the preset damage expectation is selected from the rental parameter sequence, the initial values ​​of the material usage parameters of each record in the subset are extracted, the difference between the mean of the subset and the mean of the whole sequence is calculated, and the difference is recorded as the correction upper limit.

[0022] For example, the average initial value of the material usage parameters for all historical rental records of a certain tenant is 0.9, while the average initial value of the material usage parameters for records of returned damage exceeding expectations is 1.4. The difference between the two is 0.5, and the upper limit of the correction is 0.5.

[0023] Furthermore, step S1 also includes the following steps: Step S14: Extract the duration over-utilization ratio sequence of tenants' historical lease terms from the lease parameter sequence; In one embodiment, the system groups the rental parameter sequence by tenant number, extracts the over-use ratio of all historical rental periods for the same tenant, and arranges them in chronological order to form the over-use ratio sequence of that tenant.

[0024] For example, a tenant has completed 8 rental periods in history, with corresponding overtime ratios of 1.0, 1.2, 1.1, 1.3, 1.2, 1.4, 1.3, and 1.5 respectively. Arranging these 8 values ​​in order of rental period forms the tenant's overtime ratio sequence.

[0025] It should be noted that the overtime ratio series reflects the degree to which tenants adhere to the agreed time limits across multiple lease periods. An occasional high overtime ratio in a single lease period does not necessarily indicate that the tenant has a systematic tendency to exceed the lease period. It is necessary to make a comprehensive judgment by combining the mean and standard deviation of the entire series.

[0026] Step S15: When the mean of the long-term overuse ratio sequence is continuously greater than 1 and the standard deviation is less than the preset stability threshold, the difference between the mean and 1 is used as the correction amount and added to the initial value of the material usage parameters. When the correction amount exceeds the correction limit, the correction amount is replaced by the correction limit and added again to obtain the material usage parameters. In one embodiment, the system calculates the mean and standard deviation of the overtime ratio sequence. When the mean is consistently greater than 1 and the standard deviation is less than a preset stability threshold, it is determined that the tenant has a stable tendency to overuse the material. The difference between the mean and 1 is used as a correction amount and added to the initial value of the material usage parameters. If the correction amount exceeds the correction limit, the correction amount is replaced by the correction limit and then re-added to finally obtain the corrected material usage parameters.

[0027] For example, if a tenant's overtime usage ratio has a mean of 1.3 and a standard deviation of 0.08, which is lower than the preset stability threshold of 0.15, the correction amount is 0.3. The initial value of the material usage parameter is 1.172, and the corrected material usage parameter is 1.472. If the upper limit of correction is 0.5, the correction amount of 0.3 does not exceed the upper limit, and the correction result is valid.

[0028] Step S16: Using material usage parameters, actual usage time, and material type as three-dimensional coordinates, calculate the ratio of the number of unexpected damage events to the total number of records under each coordinate combination, record it as the damage frequency value, fill it into the corresponding coordinates, and form a frequency parameter table.

[0029] In one embodiment, all rental records are grouped according to three dimensions: material usage parameters, actual usage duration, and material type. For each coordinate combination, the ratio of the number of records in which the damage result exceeds the preset damage expectation to the total number of records in that coordinate combination is calculated. This ratio is recorded as the damage frequency value and filled into the corresponding coordinate. After all coordinates are filled, a frequency parameter table is formed.

[0030] For example, there are 40 records under the coordinate (material usage parameters in the range of 1.2 to 1.4, actual usage time of 15 to 20 days, material type of steel pipe), of which 12 records show damage exceeding expectations. The damage frequency value under this coordinate is 0.30.

[0031] Furthermore, step S16 also includes the following steps: Group the rental records by material type from the frequency parameter table, extract the rental records in each group whose damage return results exceed the preset damage expectation and the rental records whose damage return results are normal, calculate the difference between the average of the ratio of actual usage time to agreed usage time in the two groups, and obtain the damage time offset for each material type. In one embodiment, after the frequency parameter table is formed, the system groups all rental parameter sequences according to material type, and extracts two subsets for each material type: a subset of records where the damage result exceeds the preset damage expectation and a subset of records where the damage result is normal. The average of the ratio of actual usage time to agreed usage time in the two subsets is calculated, and the two averages are subtracted to obtain the damage time offset of the material type.

[0032] For example, the average overuse ratio of records of damage exceeding expectations for steel pipe materials is 1.35, while the average overuse ratio of records of normal return is 0.95, with a difference of 0.40. The damage duration offset of steel pipe materials is a positive value of 0.40, indicating that steel pipe materials are more prone to damage exceeding expectations when used beyond their service life.

[0033] Material types with positive damage duration offsets are recorded as time-damage positive correlation markers, and material types with negative damage duration offsets are recorded as time-damage negative correlation markers. The two types of markers are written into the corresponding material type coordinates in the frequency parameter table.

[0034] In one embodiment, based on the positive or negative direction of the damage duration offset for each material type, a positive or negative damage duration correlation marker is assigned, and the corresponding marker is written into the coordinates of the material type in the frequency parameter table. For material types carrying a positive damage duration correlation marker, the attenuation accelerates when the actual usage time exceeds the agreed usage time in the attenuation calculation; for material types carrying a negative damage duration correlation marker, the attenuation accelerates when the actual usage time is less than the agreed usage time.

[0035] For example, if the damage duration offset of steel pipe materials is positive, a positive correlation marker for damage duration is written; if the damage duration offset of a certain type of exhibition truss is negative, a negative correlation marker for damage duration is written, indicating that this type of material has a higher probability of damage under short-term rental and high-intensity use scenarios, and needs to be accelerated to decay even if the use duration does not exceed the agreed duration.

[0036] Furthermore, step S2 includes the following steps: Step S21: Read the completion time and number of attachments from the current verification work order. When the completion time is less than 60% of the lower limit of the standard verification time for this material type, record a time abnormality mark. When the number of attachments is zero, record an attachment missing mark. In one embodiment, when a verification work order is submitted, two objective fields are automatically read: the completion time from the start of the verification operation to submission, and the number of files uploaded in the work order attachment column. The completion time is compared with the preset standard verification time lower limit for this material type. When the completion time is less than 60% of the lower limit, a time anomaly mark is recorded. The number of attachments is directly checked to see if it is zero. If it is zero, an attachment missing mark is recorded.

[0037] For example, the standard verification time for template-type materials is limited to 10 minutes. If a verification work order shows a completion time of 5 minutes, which is less than 60% of 10 minutes (6 minutes), the system will record an abnormal time mark. At the same time, the number of attachments for this work order is 0, and the system will record an attachment missing mark.

[0038] Step S22: Determine the quality correction factor based on the duration anomaly marker and the attachment missing marker; In one embodiment, the quality correction coefficient for this verification is determined based on the combination of the duration anomaly marker and the attachment missing marker. If neither marker exists, the higher quality correction coefficient is applied, indicating that no quality issues were found at the objective recording level. If only one marker exists, the intermediate quality correction coefficient is applied. If both markers exist simultaneously, the lower quality correction coefficient is applied, indicating that there are anomalies in both the completion time and attachment completeness dimensions of this verification.

[0039] For example, if a check records both the duration anomaly flag and the attachment missing flag, the quality correction factor is 0.4; if another check only records the attachment missing flag, the quality correction factor is 0.7; if neither flag exists, the quality correction factor is 1.0.

[0040] Step S23: Reassign the confidence value of the material with the quality correction factor as the upper limit, and record it as the reset initial value.

[0041] In one embodiment, the reliability value of the material is reassigned with the quality correction factor as the upper limit, and the reassignment result is recorded as the reset initial value, which serves as the starting point for the reliability decay of the material after entering the next lease period.

[0042] For example, if the quality correction factor for a certain verification is 0.7, the credibility value of the material is reset to an initial reset value not exceeding 0.7. If the platform is set to restore the credibility value to the full value corresponding to the quality correction factor after the verification is passed, then the initial reset value is 0.7. If the material has historical damage records that cause the basic credibility to be lower than 0.7, then the initial reset value is the smaller value between the actual assigned value and the quality correction factor.

[0043] Of particular importance, step S2 also includes the following: When the material outbound verification is completed, the material is scanned in three dimensions using the contour acquisition device configured at the verification station. The obtained contour data is stored with the material number as the index and recorded as the outbound contour baseline. When materials are returned, a three-dimensional contour scan is performed on the returned materials again using a contour acquisition device. The results of this scan are compared with the outline baseline of the outbound materials, and the difference between the two scan results is calculated and recorded as the contour difference. When the manual inspection results show that the material condition is normal but the profile difference exceeds the preset upper limit of normal wear difference for this material type, a deviation mark is recorded in the inspection results. The deviation marker of the verification conclusion is used as an additional input for calculating the quality correction coefficient. When the deviation marker of the verification conclusion exists, the quality correction coefficient shall not be higher than the preset deviation discount limit. The contour difference is written as an independent field in the current rental record and participates in the statistical calculation of the damage frequency value under each coordinate combination.

[0044] The calculation of the contour difference also includes the following steps: Extract the contour difference sequence corresponding to the rental records with normal return damage results from the historical rental records according to material type, calculate the mean and standard deviation of the sequence, and obtain the normal wear difference distribution of the material type. When the current profile difference exceeds the mean of the normal wear difference distribution plus twice the standard deviation, it is determined that the current profile difference exceeds the upper limit of the normal wear difference. The difference between the current profile difference and the mean of the normal wear difference distribution is recorded as the profile deviation, and the profile deviation is written into the current rental record.

[0045] Furthermore, in step S3, the real-time confidence value is obtained by continuously decaying the reset initial value according to the time step based on the frequency parameter table, including: Extract the arrival times of abnormal work orders that have arrived within the rental period, calculate the ratio of the time from each arrival time to the start time of the rental period to the total rental period, and form a location ratio sequence. In one embodiment, at each time step, it is checked whether a new abnormal work order has arrived. If so, the system record time of the work order is extracted, the time from the start time of the lease period is calculated, and then divided by the total lease period to obtain the position ratio of the work order. The position ratios of all arrived abnormal work orders are arranged in the order of arrival to form a position ratio sequence.

[0046] For example, the total rental period for a certain material is 30 days. The first abnormal work order arrives on the 6th day with a location ratio of 6 / 30 = 0.20. The second abnormal work order arrives on the 18th day with a location ratio of 18 / 30 = 0.60. The location ratio sequence is [0.20, 0.60], and the sequence mean is 0.40.

[0047] Extract the damage frequency value from the frequency parameter table and use it as the decay increment for the current time step, corresponding to the current material usage parameters, actual usage time and material type. In one embodiment, at the beginning of each time step, the corresponding coordinates are located in the frequency parameter table using the current material usage parameters, the cumulative actual usage time from the start of the lease period to the current time, and the material type as coordinates, and the damage frequency value under that coordinate is extracted as the attenuation increment of the current time step.

[0048] For example, the current material usage parameters of a certain steel pipe are 1.2, the cumulative actual usage time is 12 days, and the material type is steel pipe. The system locates the coordinates (1.2 interval, 10 to 15 days interval, steel pipe) in the frequency parameter table, extracts the damage frequency value of 0.025 under the coordinates, and uses it as the attenuation increment for this time step.

[0049] When the mean of the position ratio sequence is less than 0.5, the front segment is marked, and the attenuation increment is multiplied by the preset acceleration coefficient before attenuation is performed; otherwise, attenuation is performed directly with the attenuation increment. In one embodiment, the mean of the current position ratio sequence is calculated within each time step. When the mean is less than 0.5, it is determined that the abnormal work orders are concentrated in the first half of the rental period. The concentration of the previous period is recorded, and the attenuation increment of this time step is multiplied by a preset acceleration coefficient before attenuation is performed. When the mean is greater than or equal to 0.5, the concentration of the previous period is canceled, and attenuation is performed directly with the attenuation increment.

[0050] For example, the average of the above position ratio sequence is 0.40, which is less than 0.5. The first segment is marked. If the preset acceleration coefficient is 1.5, the decay increment of this time step changes from 0.025 to 0.025×1.5=0.0375 before decay is performed.

[0051] The decay increment executed at each time step is successively subtracted from the reset initial value to obtain the real-time confidence value at the current moment.

[0052] In one embodiment, the decay increment (including the acceleration increment triggered by the previous concentrated marker) executed at each time step is sequentially subtracted from the reset initial value. At the end of each time step, the current cumulative result is output as the real-time confidence value at that moment, which is used for threshold comparison in step S4.

[0053] For example, a material is reset to an initial value of 0.85. The decay increment on day 1 is 0.025, and the real-time confidence value at the end of day 1 is 0.825. On day 2, the front-end centralized marking is triggered, and the decay increment after acceleration is 0.0375. The real-time confidence value at the end of day 2 is 0.7875, and so on, gradually decreasing.

[0054] Of particular importance, step S3, which involves continuously decaying the reset initial value according to the time step based on the frequency parameter table, also includes: Deploy transit sensing devices in the passage between the material use area and the storage area. Each time the transit sensing devices detect the passage of materials, they record the time and direction of passage. Passage from the storage area to the use area is recorded as an entry event, and passage from the use area to the storage area is recorded as an exit event. Within each time step, the difference between the number of inbound events and outbound events that have occurred during the current lease period is calculated to obtain the cumulative net inbound count of materials in the usage area at the current moment. The real-time usage intensity coefficient at the current moment is obtained by the ratio of the cumulative net inbound count to the total number of inbound events during the lease period. Multiply the real-time usage intensity coefficient by the damage frequency value under the corresponding coordinate in the frequency parameter table to obtain the attenuation increment after real-time usage intensity correction. Replace the attenuation increment obtained by looking up the table based on the fixed material usage parameters and perform attenuation of the confidence value at the current time step. When returning each lease term, the average value of the real-time usage intensity coefficient sequence at each time step within the lease term is used as the measured usage intensity for this lease term. This value is compared with the initial value of the material usage parameters calculated based on the information field. When the difference between the two exceeds the preset deviation threshold, the industry intensity mapping value corresponding to the tenant's industry type is updated with the measured usage intensity, thus completing the automatic calibration of the industry intensity mapping table.

[0055] The calculation of the strength coefficient in real time also includes: Extract the occurrence time of each migration event within the current lease period from the records of the transit sensing device, calculate the ratio of the duration of each migration time from the start time of the lease period to the total lease period, and form a migration location ratio sequence. The moving-in position ratio sequence is compared with the position ratio sequence. When the difference between the mean of the moving-in position ratio sequence and the mean of the position ratio sequence is less than the preset synchronization threshold, it is determined that the transit perception event and the abnormal work order arrival event are highly consistent in time distribution, and the hardware and software synchronization mark is recorded. When hardware and software synchronization markers exist, the trigger judgment of the previous centralized markers is replaced by the real-time intensity coefficient, and the adjustment of the attenuation increment is directly driven by the real-time intensity coefficient, eliminating the dependence on the condition that the mean of the position ratio sequence is less than 0.5.

[0056] Furthermore, in step S3, the responsibility is divided into segments based on the arrival time of abnormal work orders within the lease period, and the real-time reliability value at the start time of each segment is recorded as the segmentation benchmark value, including: At the moment each abnormal work order arrives, the current real-time credibility value is recorded as the segmentation benchmark value for responsibility segmentation, where the arrival time of the abnormal work order is the system record time when the tenant corresponding to the abnormal work order submits the work order; In one embodiment, at the start of the lease term, the initial value is reset and recorded as the segmentation benchmark value of the first responsibility segment. Subsequently, whenever a new abnormal work order arrives during the lease term, the real-time reliability value at that moment is extracted as the segmentation benchmark value of the new responsibility segment and stored in conjunction with the arrival time of the corresponding abnormal work order.

[0057] For example, the initial value of a template rental period reset is 0.90. On the 8th day, the first abnormal work order arrives, at which time the real-time reliability value is 0.74, and the segmentation benchmark value of the first responsibility segment is recorded as 0.74. On the 20th day, the second abnormal work order arrives, at which time the real-time reliability value is 0.61, and the segmentation benchmark value of the second responsibility segment is recorded as 0.61. When the rental period ends, the real-time reliability value is 0.52, and the segmentation benchmark value of the third responsibility segment is recorded as 0.52.

[0058] It should be noted that the starting point of the first responsibility segment is the start time of the lease period, not the arrival time of the first abnormal work order. Therefore, the initial reset value at the start of the lease period should also be recorded as the segment baseline value to ensure that the baseline value sequence of each segment covers the entire lease period.

[0059] The difference between the baseline values ​​of adjacent segments is calculated to obtain the actual decrease in the reliability of each segment. The number, duration and material type of abnormal work orders in each segment are used as coordinates to locate the segment in the frequency parameter table. The product of the damage frequency value under the corresponding coordinate and the duration of the segment is extracted to obtain the predicted decrease of the segment. In one embodiment, at the end of each responsibility segment, the actual decrease in reliability of the segment is calculated by the difference between the baseline values ​​of two adjacent segments; at the same time, the number of abnormal work orders, the segment duration, and the material type in the segment are used as coordinates to locate the segment in the frequency parameter table, the damage frequency value under the corresponding coordinates is extracted, and the predicted decrease is obtained by multiplying it by the segment duration.

[0060] For example, the first responsibility segment lasts for 8 days, and the segment baseline value drops from 0.90 to 0.74, with an actual decrease in reliability of 0.16. Using the number of abnormal work orders in this segment (1), the segment duration (8 days), and the material type template as coordinates, the damage frequency value is found to be 0.015, and the predicted decrease is 0.015 × 8 = 0.12.

[0061] When the actual decline in confidence exceeds the predicted decline, the strength of the marker should be underestimated and the underestimated marker should be appended to the segmentation baseline value.

[0062] In one embodiment, the actual decrease in credibility of each responsibility segment is compared with the predicted decrease. When the actual decrease exceeds the predicted decrease, it is determined that the frequency parameter table underestimates the intensity of use for that segment. An underestimation mark is recorded for that segment and attached to the corresponding segment baseline value.

[0063] For example, the actual decrease in the first responsibility segment was 0.16, while the predicted decrease was 0.12. The actual decrease exceeded the prediction, so an underestimation marker was added to the first responsibility segment. The actual decrease in the second responsibility segment was 0.13, while the predicted decrease was 0.15. The actual decrease did not exceed the prediction, so no underestimation marker was added.

[0064] Of particular importance, step S3 also includes: When returning the leased property, extract the segments with underestimated strength markers from the baseline values ​​of each segment, calculate the difference between the actual decrease in confidence and the predicted decrease, and obtain the confidence deviation value. Based on the reliability deviation value of each segment of the material type archive, when the reliability deviation value of multiple consecutive lease periods under the same material type is continuously positive, the damage frequency value under the corresponding coordinate in the frequency parameter table is updated with the average reliability deviation value, and the damage frequency value before and after the update and the lease record number are written into the correction log.

[0065] Furthermore, step S3, which divides responsibility segments based on the arrival time of abnormal work orders within the lease period, also includes: Extract the location ratio sequence of arrival times of abnormal work orders for all historical lease periods of the tenant from the historical transfer records by tenant number, calculate the mean of the location ratio sequence for each historical lease period, and form the historical reported location mean sequence for the tenant. Extract the start time of the actual damage attribution segment within the tenant's historical lease term from the questionable work order processing conclusion, calculate the ratio of the time between the start time of the actual damage attribution segment and the start time of the corresponding lease term to the total lease term, and form the average sequence of the tenant's historical damage locations. Calculate the difference sequence between the historical average reported location sequence and the historical average damaged location sequence corresponding to the lease period to obtain the reporting delay deviation sequence for the tenant. When the mean of the reported delay deviation sequence is consistently positive and the standard deviation is less than the preset stability threshold, it is determined that the tenant has a systematic tendency to report delays, and the delay reporting flag and the mean of the reported delay deviation sequence are recorded. During the current lease term, when the tenant carries a delayed reporting flag, the arrival time of the abnormal work orders that have arrived during the current lease term is subtracted from the duration corresponding to the average of the reporting delay deviation sequence to obtain the corrected arrival time of each abnormal work order. The corrected arrival time is used instead of the actual arrival time as the segmentation node for responsibility segmentation, and the real-time reliability value at each corrected segmentation node is recorded as the segmentation benchmark value.

[0066] The calculation of the reported delay deviation sequence further includes: Group the reported delay deviation sequence by material type, calculate the mean and standard deviation of the subsequence corresponding to each material type, and obtain the classification reporting delay deviation value of the tenant under each material type; When the difference between the classification reporting delay deviation value under a certain material type and the average value of all material types exceeds the preset significance threshold, a specific delay marker for that tenant under that material type is recorded.

[0067] It also includes the following: when the current lease term involves a material type with a specific delay mark, the calculation of the correction arrival time is performed by replacing the average value of all types with the classification reporting delay deviation value corresponding to that material type, so that the correction accuracy of the split node converges from the tenant level to the tenant-material type intersection level.

[0068] Furthermore, step S4 includes: Set an upper threshold and a lower threshold. When the real-time confidence value is higher than the upper threshold, the material is determined to be eligible for allocation. In one embodiment, two preset thresholds, an upper threshold and a lower threshold, are set for each type of material. At the end of each time step, the current real-time confidence value is compared with the two thresholds, and the allocation qualification of the material is processed accordingly based on the comparison result.

[0069] For example, a platform sets the upper threshold for steel pipe materials to 0.6 and the lower threshold to 0.3. The current real-time reliability value of a certain batch of steel pipes is 0.75, which is higher than the upper threshold of 0.6. The system determines that the batch of steel pipes is qualified for normal allocation and will respond first when a tenant applies for it.

[0070] When the real-time credibility value is between the upper and lower thresholds, the material is arranged in the allocation order after other materials of the same type that are eligible for allocation. In one embodiment, when the real-time confidence value of a material is between the upper threshold and the lower threshold, its allocation eligibility is not cancelled, but when generating the allocation scheme, it is arranged after the materials of the same type whose real-time confidence value is higher than the upper threshold, and it is only included in the allocation when the inventory of high-confidence materials is insufficient to meet the current application.

[0071] For example, a tenant requests 50 steel pipes. Currently, there are 30 steel pipes with a real-time credibility value higher than the upper threshold, and 40 steel pipes with a credibility value between the two thresholds. The 30 high-credibility steel pipes are allocated first, and the remaining 20 are supplemented from the steel pipes with a credibility value between the two thresholds.

[0072] When the real-time credibility value falls below the lower threshold, the material's allocation eligibility is revoked and a verification is triggered.

[0073] In one embodiment, when the real-time reliability value of a material falls below a lower threshold, the system immediately cancels the material's allocation eligibility, removes it from the allocatable inventory, and automatically generates a verification work order to trigger the material verification process. After verification is completed, the system returns to step S2, recalculates the quality correction coefficient based on the completion time and attachment completeness of the current verification work order, and reassigns the reliability value of the material. Only after the reassignment is completed can the material re-participate in the allocation eligibility determination.

[0074] For example, if the real-time credibility value of a batch of templates drops to 0.28, which is below the lower threshold of 0.3, the templates will be disqualified from being allocated and a verification work order will be generated. After the verifier completes the verification and submits the work order, the credibility value will be reset to 0.85 based on the quality of this verification, and the templates will regain the eligibility to be allocated.

[0075] Of particular importance, step S4, which determines the allocation eligibility of batch-level materials, also includes the following steps: The distribution dispersion is obtained by calculating the standard deviation of the part number interval sequence and identifying the material part numbers whose real-time reliability values ​​are below the lower threshold within the batch. When the distribution dispersion is lower than the preset concentration threshold, the qualification for allocating the entire batch of materials is suspended and the damage tendency during the platform's self-management period is marked. When the distribution dispersion is higher than the preset concentration threshold, only the allocation qualification of a single material with a low real-time reliability value is cancelled.

[0076] Furthermore, step S5 includes the following steps: Step S51: Extract the material usage parameters, actual usage time, and historical type distribution of returned damage results under the material type coordinates corresponding to each responsibility segment from the frequency parameter table to obtain the damage type probability vector of each responsibility segment; In one embodiment, a frequency parameter table is queried for each responsibility segment. The material usage parameters, actual usage time and material type corresponding to the segment are used as coordinates for positioning. The occurrence ratio of each damage type in the historical return damage results under the coordinates is extracted to form the damage type probability vector of the responsibility segment.

[0077] For example, the lease term of a template is divided into three responsibility segments. The third segment has 80 historical damage records under the corresponding coordinates. Among them, surface damage accounts for 55%, structural bending accounts for 30%, and no damage accounts for 15%. The probability vector of damage type for this segment is [surface damage: 0.55, structural bending: 0.30, no damage: 0.15].

[0078] Step S52: Match the damage type recorded at the time of return with the damage type probability vector of each responsibility segment to obtain the initial damage type matching degree of each responsibility segment; In one embodiment, the actual damage type recorded when the material is returned is compared with the damage type probability vector of each responsible segment, and the probability value corresponding to the actual damage type in the probability vector of each segment is extracted as the initial damage type matching degree of that segment.

[0079] For example, the damage type recorded during this return is surface layer damage. The probability of surface layer damage in the first segment damage type probability vector is 0.20, the second segment is 0.35, and the third segment is 0.55. The initial damage type matching degree of the three segments is 0.20, 0.35, and 0.55, respectively.

[0080] Step S53: Weight the initial damage type matching degree with the benchmark value of each segment, and positively correct the weight of the benchmark value of the segment carrying the underestimation of intensity to obtain the corrected damage type matching degree; In one embodiment, the initial damage type matching degree of each segment is weighted by the segment baseline value of each responsible segment. The higher the segment baseline value, the more reliable the material state record at the beginning of the segment and the more sufficient the damage attribution basis of the segment, so the greater the weight. For segments carrying the strength underestimation mark, a positive correction is applied on the basis of the segment baseline value weight to further improve the damage attribution weight of the segment, and the corrected damage type matching degree is obtained.

[0081] For example, the segmentation baseline values ​​for the three segments are 0.90, 0.74, and 0.61, respectively, and the initial damage type matching degrees are 0.20, 0.35, and 0.55, respectively. After weighting, the values ​​are 0.20×0.90=0.18, 0.35×0.74=0.26, and 0.55×0.61=0.34. The third segment carries an underestimation of intensity, and after applying a positive correction coefficient of 1.2, it is corrected to 0.34×1.2=0.41. After correction, the damage type matching degrees for the three segments are 0.18, 0.26, and 0.41, respectively.

[0082] Step S54: Normalize the damage type matching degree of each responsibility segment to obtain the damage allocation ratio of each responsibility segment; In one embodiment, the damage type matching degree of each responsible segment after correction is added together to obtain a total, and then the matching degree of each segment is divided by the total to obtain the damage allocation ratio of each responsible segment. The sum of the damage allocation ratios of each segment is 1.

[0083] For example, the damage type matching degrees after the three segment corrections are 0.18, 0.26, and 0.41, respectively, with a total of 0.85. After normalization, the damage distribution ratios of each segment are 0.18 / 0.85≈0.21, 0.26 / 0.85≈0.31, and 0.41 / 0.85≈0.48, respectively.

[0084] Step S55: Generate a handling notice for responsibility segments whose damage allocation ratio is higher than the preset attribution threshold, and generate a questionable work order for responsibility segments whose damage allocation ratio is lower than the preset attribution threshold.

[0085] In one embodiment, the damage allocation ratio of each responsibility segment is compared with a preset attribution threshold one by one. For responsibility segments with a damage allocation ratio higher than the preset attribution threshold, a disposal notice is generated to clarify that the tenant corresponding to the segment bears the corresponding damage responsibility. For responsibility segments with a damage allocation ratio lower than the preset attribution threshold, a questionable work order is generated for the platform operation personnel to intervene and handle.

[0086] For example, the preset attribution threshold is 0.40, and the damage allocation ratio of the third segment is 0.48, which is higher than the threshold. The system generates a disposal notice to the tenant corresponding to the third segment. The damage allocation ratio of the first segment is 0.21, and that of the second segment is 0.31, both of which are lower than the threshold. The system generates a corresponding questionable work order.

[0087] Step S5 further includes the following steps: When the current lease record carries a profile deviation, the actual confidence decrease of each responsible segment is extracted from the benchmark value sequence of each segment, and the ratio of the actual confidence decrease of each responsible segment to the predicted decrease is calculated to obtain the intensity deviation ratio of each responsible segment. The baseline weight of each responsibility segment is adjusted twice based on the strength deviation ratio of each responsibility segment. The larger the strength deviation ratio, the greater the adjustment range of the weight of the responsibility segment. When the contour deviation exceeds the preset severe damage threshold, the work order corresponding to the responsible segment with the highest damage allocation ratio will be automatically upgraded to a handling notice, without waiting for manual intervention.

[0088] Figure 2 This is a time-series diagram illustrating the decay of credibility values. In this diagram, 300 represents the lease period time axis, 302 the credibility value coordinate axis, 304 the credibility value decay line, 306 the segment node corresponding to the arrival time of the first abnormal work order, 308 the segment node corresponding to the arrival time of the second abnormal work order, 310 the first responsibility segment, 312 the second responsibility segment, 314 the third responsibility segment, 316 the upper threshold, 318 the lower threshold, and 320 the node that triggers verification when the real-time credibility value reaches the lower threshold.

[0089] like Figure 2 As shown, after the material verification is completed, the confidence value is reset to the initial value and then decreases at a normal decay rate in the first responsibility segment. At point 306, the first abnormal work order arrives, the mean of the position ratio sequence is less than 0.5, the front-end centralized marker is triggered, and the confidence value decays at a significantly higher slope than the first segment in the second responsibility segment. After entering the third responsibility segment, the confidence value continues to decrease and touches the lower threshold of 318 in the middle of the segment. The system triggers material verification at point 320. At this time, the material is still within the lease period, which reflects the system's early warning capability before damage occurs.

[0090] Figure 3 This diagram illustrates the division of responsibility segments and damage attribution output. 600 represents the rental period timeline, 602 the rental start date, 604 the return date, 606 the arrival date of the first abnormal work order, 608 the arrival date of the second abnormal work order, 610 the first responsibility segment, 612 the second responsibility segment, 614 the third responsibility segment, 624 the duration of the first responsibility segment (8 days), 626 the duration of the second responsibility segment (12 days), 628 the duration of the third responsibility segment (10 days), 630 the additional strength underestimation marker within the third responsibility segment, 632 the damage type recorded upon material return, 634 the damage allocation ratio attributing damage to the first responsibility segment, 636 the damage allocation ratio attributing damage to the second responsibility segment, and 638 the damage allocation ratio attributing damage to the third responsibility segment.

[0091] like Figure 3As shown, the rental period time axis 600 uses the arrival times of two abnormal work orders 606 and 608 as nodes, dividing the complete rental period into three responsibility segments 610, 612, and 614, with segment durations of 8 days, 12 days, and 10 days, respectively. When materials are returned, damage 632 (surface layer damage) is discovered. The system extracts the damage type probability vector under the corresponding coordinates of each responsibility segment from the frequency parameter table, matches the returned damage type with the probability vector of each segment, and normalizes it after weighting by the baseline value of each segment, obtaining the damage allocation ratios of each responsibility segment: 634 (0.21), 636 (0.31), and 638 (0.48). The third responsibility segment carries an underestimation of strength marker 630, and its damage allocation ratio is the highest after positive correction of the damage attribution weight. Based on this, the system generates a disposal notification to the tenant corresponding to the third responsibility segment.

[0092] See Figure 4 The present invention also provides a resource dynamic allocation and risk warning system 100 for a material leasing platform, used to execute the resource dynamic allocation and risk warning method for a material leasing platform as described above. The resource dynamic allocation and risk warning system 100 for the material leasing platform includes: The frequency parameter construction module 101 is used to extract the material usage parameters, the corresponding sequence of abnormal work orders and return damage results from the historical circulation records of the material rental platform, and form a frequency parameter table. The credibility reset module 102 is used to calculate the quality correction coefficient based on the completion time of the current verification work order and the completeness of the attachments when each material verification is completed, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. The credibility decay module 103 is used to continuously decay the reset initial value according to the time step based on the frequency parameter table after the material enters the rental period to obtain the real-time credibility value; the responsibility segment is divided with the arrival time of abnormal work orders during the rental period as the node, and the real-time credibility value at the beginning of each segment is recorded as the segment reference value. The allocation qualification determination module 104 is used to compare the real-time confidence value with a preset threshold, determine the allocation qualification of the material based on the comparison result, and trigger material verification when the real-time confidence value is lower than the preset threshold. The damage attribution module 105 is used to assign damage to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type when damage occurs during material return, and generate a disposal notice or a questionable work order respectively.

[0093] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0094] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A resource dynamic allocation and risk early warning method for a material leasing platform, characterized in that, Includes the following steps: Step S1: Extract the material usage parameters, the corresponding sequence of abnormal work orders during the rental period and the return damage results from the historical circulation records of the material rental platform to form a frequency parameter table; Step S2: When each material verification is completed, calculate the quality correction coefficient based on the completion time of the verification work order and the completeness of the attachments, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. Step S3: After the materials are put into use for rental, the initial value is continuously decayed according to the time step based on the frequency parameter table to obtain the real-time reliability value; the responsibility segments are divided with the arrival time of abnormal work orders during the rental period as nodes, and the real-time reliability value at the beginning of each segment is recorded as the segment benchmark value. Step S4: Compare the real-time reliability value with the preset threshold, determine the material's eligibility for allocation based on the comparison result, and trigger material verification when the real-time reliability value is lower than the preset threshold; Step S5: When damage occurs during material return, the damage is assigned to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type, and a handling notice or a questionable work order is generated respectively.

2. The method of claim 1, wherein, Step S1 includes the following steps: Step S11: Extract the material number, tenant industry type, actual usage duration, agreed usage duration, abnormal work orders during the rental period, and return damage results for each rental record from the historical circulation records of the material rental platform, and align them one by one according to the rental record number to form a rental parameter sequence; Step S12: Take the tenant industry type mapping value, the ratio of actual usage time to agreed usage time, and the number of abnormal work orders during the lease period from the lease parameter sequence, and sum them according to the preset weights to obtain the initial value of the material usage parameters; Step S13: Extract the subsequences in the rental parameter sequence where the return damage results exceed the preset damage expectation, and calculate the upper limit of correction based on the difference between the mean of the subsequence and the mean of the whole sequence.

3. The method of claim 2, wherein, Step S1 also includes the following steps: Step S14: Extract the duration over-utilization ratio sequence of tenants' historical lease terms from the lease parameter sequence; Step S15: When the mean of the long-term overuse ratio sequence is continuously greater than 1 and the standard deviation is less than the preset stability threshold, the difference between the mean and 1 is used as the correction amount and added to the initial value of the material usage parameters. When the correction amount exceeds the correction limit, the correction amount is replaced by the correction limit and added again to obtain the material usage parameters. Step S16: Using material usage parameters, actual usage time, and material type as three-dimensional coordinates, calculate the ratio of the number of unexpected damage events to the total number of records under each coordinate combination, record it as the damage frequency value, fill it into the corresponding coordinates, and form a frequency parameter table.

4. The method of claim 3, wherein, Step S16 also includes the following steps: Group the rental records by material type from the frequency parameter table, extract the rental records in each group whose damage return results exceed the preset damage expectation and the rental records whose damage return results are normal, calculate the difference between the average of the ratio of actual usage time to agreed usage time in the two groups, and obtain the damage time offset for each material type. Material types with positive damage duration offsets are recorded as time-damage positive correlation markers, and material types with negative damage duration offsets are recorded as time-damage negative correlation markers. The two types of markers are written into the corresponding material type coordinates in the frequency parameter table.

5. The resource dynamic allocation and risk early warning method of the material leasing platform according to claim 4, characterized in that, Step S2 includes the following steps: Step S21: Read the completion time and number of attachments from the current verification work order. When the completion time is less than 60% of the lower limit of the standard verification time for this material type, record a time abnormality mark. When the number of attachments is zero, record an attachment missing mark. Step S22: Determine the quality correction factor based on the duration anomaly marker and the attachment missing marker; Step S23: Reassign the confidence value of the material with the quality correction factor as the upper limit, and record it as the reset initial value.

6. The resource dynamic allocation and risk early warning method of the material leasing platform according to claim 5, characterized in that, Step S3, which involves continuously decaying the reset initial value according to the time step based on the frequency parameter table to obtain the real-time confidence value, includes: Extract the arrival times of abnormal work orders that have arrived within the rental period, calculate the ratio of the time from each arrival time to the start time of the rental period to the total rental period, and form a location ratio sequence. Extract the damage frequency value from the frequency parameter table and use it as the decay increment for the current time step, corresponding to the current material usage parameters, actual usage time and material type. When the mean of the position ratio sequence is less than 0.5, the front segment is marked, and the attenuation increment is multiplied by the preset acceleration coefficient before attenuation is performed; otherwise, attenuation is performed directly with the attenuation increment. The decay increment executed at each time step is successively subtracted from the reset initial value to obtain the real-time confidence value at the current moment.

7. The resource dynamic allocation and risk early warning method of the material leasing platform according to claim 6, characterized in that, In step S3, the responsibility is divided into segments based on the arrival time of abnormal work orders within the lease period. The real-time reliability value at the start time of each segment is recorded as the segmentation benchmark value, including: At the moment each abnormal work order arrives, the current real-time credibility value is recorded as the segmentation benchmark value for responsibility segmentation, where the arrival time of the abnormal work order is the system record time when the tenant corresponding to the abnormal work order submits the work order; The difference between the baseline values ​​of adjacent segments is calculated to obtain the actual decrease in the reliability of each segment. The number, duration and material type of abnormal work orders in each segment are used as coordinates to locate the segment in the frequency parameter table. The product of the damage frequency value under the corresponding coordinate and the duration of the segment is extracted to obtain the predicted decrease of the segment. When the actual decline in confidence exceeds the predicted decline, the strength of the marker should be underestimated and the underestimated marker should be appended to the segmentation baseline value.

8. The method of claim 7, wherein, Step S4 includes: Set an upper threshold and a lower threshold. When the real-time confidence value is higher than the upper threshold, the material is determined to be eligible for allocation. When the real-time credibility value is between the upper and lower thresholds, the material is arranged in the allocation order after other materials of the same type that are eligible for allocation. When the real-time credibility value falls below the lower threshold, the material's allocation eligibility is revoked and a verification is triggered.

9. The method for dynamic resource allocation and risk warning of a material leasing platform according to claim 8, characterized in that, Step S5 includes the following steps: Step S51: Extract the material usage parameters, actual usage time, and historical type distribution of returned damage results under the material type coordinates corresponding to each responsibility segment from the frequency parameter table to obtain the damage type probability vector of each responsibility segment; Step S52: Match the damage type recorded at the time of return with the damage type probability vector of each responsibility segment to obtain the initial damage type matching degree of each responsibility segment; Step S53: Weight the initial damage type matching degree with the benchmark value of each segment, and positively correct the weight of the benchmark value of the segment carrying the underestimation of intensity to obtain the corrected damage type matching degree; Step S54: Normalize the damage type matching degree of each responsibility segment to obtain the damage allocation ratio of each responsibility segment; Step S55: Generate a handling notice for responsibility segments whose damage allocation ratio is higher than the preset attribution threshold, and generate a questionable work order for responsibility segments whose damage allocation ratio is lower than the preset attribution threshold.

10. A resource dynamic allocation and risk early warning system for a material leasing platform, characterized in that, For executing the resource dynamic allocation and risk warning method of the material leasing platform as described in claim 1, the resource dynamic allocation and risk warning system of the material leasing platform includes: The frequency parameter construction module is used to extract material usage parameters, the corresponding sequence of abnormal work orders and return damage results from the historical circulation records of the material rental platform, and form a frequency parameter table. The credibility reset module is used to calculate the quality correction coefficient based on the completion time of the current verification work order and the completeness of the attachments when each material verification is completed, and reset the credibility value of the material to the initial reset value with the quality correction coefficient as the upper limit. The credibility decay module is used to continuously decay the reset initial value according to the time step based on the frequency parameter table after the material enters the rental period to obtain the real-time credibility value; the responsibility segment is divided with the arrival time of abnormal work orders during the rental period as the node, and the real-time credibility value at the beginning of each segment is recorded as the segment benchmark value. The allocation eligibility determination module is used to compare the real-time reliability value with a preset threshold, determine the allocation eligibility of materials based on the comparison result, and trigger material verification when the real-time reliability value is lower than the preset threshold. The damage attribution module is used to assign damage to the corresponding responsible segment based on the matching degree between the baseline value of each segment and the pre-determined damage type when damage occurs during material return, and generate a disposal notice or a questionable work order respectively.

Images