Tracing management system for machine circulation among multi-stage construction units

Abstract

The invention relates to the technical field of machine tool circulation management, in particular to a traceability management system for machine tool circulation among multistage construction units. According to the method, cross validation is carried out on the vibration influence, the appearance abnormity and the transportation aging deviation, intelligent analysis is carried out by utilizing the trained responsibility attribution model, the responsibility party of machine tool damage is effectively traced and defined, the management cost and disputes are greatly reduced, and the cooperation efficiency and the asset management level between construction units are improved.

Description

Technical Field

[0001] This invention relates to the field of machinery and equipment circulation management technology, specifically to a traceability management system for machinery and equipment circulation among multi-level construction units. Background Technology

[0002] In large-scale engineering projects such as railway, highway, and water conservancy construction, multiple levels of construction units, including general contractors, subcontractors, and labor providers, typically share or lease large construction machinery to optimize resource allocation, reduce overall costs, and avoid waste. Therefore, machinery frequently circulates among different levels of construction units. A simple traceability management system, primarily based on manual records and supplemented by electronic spreadsheets, is in place to provide data support for subsequent verification of machinery compliance with construction safety requirements and for completing project cost accounting.

[0003] However, the assessment of the physical condition of machinery upon entry and exit from storage relies heavily on visual judgment and photographic documentation, lacking standardized quantitative criteria. If new damage is discovered upon receipt, disputes over liability can easily arise between the sender, transporter, and receiver. Secondly, the transportation process is opaque, and damage caused during transport often only becomes apparent during subsequent use, making it impossible to trace responsibility at that point.

[0004] Furthermore, the fragmented information records throughout the entire equipment circulation management process make it difficult to form a data-driven evidence chain that runs through the entire circulation process, resulting in the inability to fully and quickly trace the circulation history and health status of the equipment. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art, objectively quantify and record the appearance of machinery at the transferor and receiver, transparently monitor the transportation process, and on this basis, intelligently analyze and define the responsibility for machinery damage through cross-verification of multi-dimensional data, thereby reducing disputes and improving management efficiency and asset security.

[0006] The technical solution adopted by the present invention to solve its technical problem is: a traceability management system for the transfer of machinery and equipment among multi-level construction units, including: a scheduling trigger module, used to generate scheduling instructions containing the sender code, receiver code, unique number of machinery and equipment and planned transportation duration, and to mark the status of the machinery and equipment.

[0007] The outbound record module is used to scan the electronic tag of the equipment when it leaves the dispatching party, record the actual dispatch time, collect images of key parts, and generate an outbound appearance index.

[0008] The transportation monitoring module is used to acquire vibration data during transportation through sensors installed on the equipment, and to obtain the impact of vibration.

[0009] The inbound verification module is used to scan the electronic tag of the equipment when it arrives at the recipient, record the actual arrival time, and obtain the transportation timeliness deviation by combining the actual dispatch time and the planned transportation time. At the same time, it collects images of key parts, generates an inbound appearance index, and obtains the appearance anomaly by combining it with the outbound appearance index.

[0010] The record generation module is used to perform cross-validation analysis based on transportation timeliness deviation, vibration impact, and appearance anomalies, and generate traceability records based on the cross-validation conclusions to update the equipment status.

[0011] The data storage module is used to associate traceability records, updated equipment status, and unique equipment numbers with the equipment and store them in the database.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention achieves refined and automated monitoring of the entire process of equipment transfer between multi-level construction units through closed-loop management of scheduling triggering, outbound recording, transportation monitoring, inbound verification, record generation and data storage, thereby improving the efficiency of equipment management.

[0013] 2. This invention cross-validates vibration impact, appearance anomalies, and transportation timeliness deviations, and uses a trained responsibility attribution model for intelligent analysis to effectively trace and define the responsible party for equipment damage, greatly reducing management costs and disputes, and improving the collaborative efficiency and asset management level among construction units. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0016] Figure 2 This is a schematic diagram of the cross-validation analysis process in this invention.

[0017] Figure 3 This is a schematic diagram of the process for generating traceability records and updating equipment status in this invention.

[0018] Figure 4 This is a schematic diagram of the process for obtaining transportation timeliness deviation in this invention. Detailed Implementation

[0019] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] The following description, in conjunction with the accompanying drawings, details the specific scheme of the traceability management system for the transfer of machinery and equipment among multi-level construction units provided by this invention.

[0024] Please see Figure 1 The diagram illustrates a module connection diagram of a traceability management system for the transfer of machinery and equipment among multi-level construction units, provided by an embodiment of the present invention. Specifically, it includes: a scheduling trigger module, an outbound record module, a transportation monitoring module, an inbound verification module, a record generation module, and a data storage module.

[0025] The outbound record module is connected to the scheduling trigger module and the transportation monitoring module, respectively. The transportation monitoring module and the inbound verification module are connected to the record generation module, respectively. The record generation module is connected to the data storage module.

[0026] The module connection diagram reflects the flow sequence of data in the system. Starting from the scheduling trigger module, the data passes through the outbound record module, transportation monitoring module, inbound verification module, record generation module, and finally is stored in the data storage module, ensuring that the traceability data of the entire equipment circulation process is continuous and traceable.

[0027] The scheduling triggering module is used to generate a scheduling instruction that includes the sender code, receiver code, equipment unique number and planned transportation duration, and to mark the equipment status.

[0028] In one embodiment of the present invention, before equipment is transferred between construction units, the system must first automatically initiate a scheduling task. To ensure that the equipment transfer process is traceable and that responsibility is delineated, the system needs to pre-generate standardized scheduling instructions and dynamically update the status of the relevant equipment.

[0029] Considering the differences in geographical distribution, construction progress, and resource allocation among different construction units, the generation of scheduling instructions needs to be based on matching actual needs with available resources. Therefore, in a preferred embodiment of the present invention, the implementation process of the scheduling trigger module is as follows: The system first obtains the equipment requirements submitted by the recipient, which includes the recipient code, the dispatcher code, and the required equipment type.

[0030] Next, based on the required equipment type, the system filters out equipment whose current system status is marked as transferable from the equipment management information database of the designated dispatcher, identifies them as equipment to be dispatched, and obtains their unique number. This number is usually bound to an electronic identifier fixed on the equipment, such as an RFID tag or QR code.

[0031] To avoid introducing substandard equipment into the construction process, transferable equipment must meet at least the following conditions: no fault repair records within the past three months; all core functions for safe operation of the equipment are in normal working order, such as braking and overload protection; and the equipment's physical parameters, such as size, weight, and power, comply with the spatial layout, foundation bearing capacity, and power supply capacity requirements of the receiving site.

[0032] The system first obtains the latitude and longitude coordinates of the caller and receiver based on their registered geographical location information. Then, using these coordinates as input parameters, it calls the map service provider's route planning interface and specifies the mode of transportation, such as truck transport. The map service provider then automatically plans a reasonable transportation route based on its comprehensive actual road network information, and adds up the length of each road segment along the route to obtain the theoretical transportation distance for the entire journey.

[0033] Furthermore, the system presets a reference unit distance-time, such as 0.5 hours / km, which can be set based on historical transportation data. Multiplying the theoretical transportation distance by this reference unit distance-time yields the planned transportation duration.

[0034] Finally, the scheduling trigger module combines the dispatcher's code, receiver's code, equipment's unique number, and planned transportation duration into a standardized scheduling instruction data packet, and sends it to the dispatcher via the network. At the same time, it updates the equipment's status in the system from "available for transfer" to "awaiting dispatch".

[0035] It should be noted that a floating range can also be set for the reference unit distance time according to different means of transportation, such as heavy trucks and light trucks, for example, 0.4-0.6 hours / km, to enhance the accuracy of equipment circulation traceability management.

[0036] The outbound record module is used to scan the electronic tag of the equipment when it leaves the dispatching party, record the actual dispatch time, collect images of key parts, and generate an outbound appearance index.

[0037] In one embodiment of the present invention, when a machine is ready to be dispatched, its physical condition must be objectively recorded as an important basis for subsequent liability determination. To this end, operators use handheld terminals or fixed readers to scan the electronic tags on the machines.

[0038] Upon successful scanning, the outbound record module automatically records this moment as the actual retrieval time. Simultaneously, the high-definition camera deployed in the outbound channel can be triggered to capture images of key parts of the equipment. At least three images of each key part are captured from three angles: front, side, and top. The acquisition environment must meet the requirement that the light intensity is greater than or equal to 500 Lux, avoiding shadows and ensuring coverage of vulnerable areas.

[0039] It should be noted that critical parts are those parts of the machinery that are easily damaged and that affect safe use. Taking tower cranes as an example, critical parts include, but are not limited to, tower body connecting bolts, boom pins, and hook anti-detachment devices.

[0040] Furthermore, the specific method for generating the outbound appearance index is as follows: First, image recognition algorithms are used to identify defective areas in the acquired images, such as rust, scratches, and deformation. Next, the ratio of the total area of ​​all defective areas to the total area of ​​the image is calculated to obtain the area proportion S.

[0041] It should be noted that the image recognition algorithm can be edge detection or threshold segmentation technology based on the OpenCV library; to further improve the accuracy of defect region identification and classification, preferably, a deep learning-based target detection or semantic segmentation model, such as YOLO or U-Net, can be used.

[0042] Taking the image of the crane boom pin as an example: The system identified one scratch defect with an area of ​​10128 pixels. That is, the area accounts for 0.5%.

[0043] Then, pixel-level height differences between each defective region and the normal surface are obtained through image feature analysis.

[0044] The specific process is as follows: Convert the acquired color images of key areas into grayscale images, according to the calculation formula: This yields the gradient magnitude of each pixel. .in , These are the gradient values ​​of the pixel in the X and Y directions, respectively. The larger the gradient magnitude, the more drastic the change in image brightness at that pixel, which usually corresponds to the boundaries of edges, cracks, or depressions.

[0045] Accordingly, for each identified defective region, all pixels within it are traversed to obtain the gradient magnitude of each defective pixel. Then, the average gradient magnitude of all pixels on the normal surface is calculated, and the absolute value of the difference between the gradient magnitude of each defective pixel and the average value is taken as the pixel-level drop.

[0046] Next, the arithmetic mean of all pixel-level drops is calculated as the average defect depth. Subsequently, the area proportions were determined. With average defect depth Multiply them to obtain the overall damage degree.

[0047] Finally, considering that machinery is a long-term asset, its appearance damage is cumulative. Therefore, the appearance index of the machinery upon its last transfer is retrieved from the data storage module. ,Will Subtracting the overall damage level calculated in this study from the final value yields the final outbound appearance index. ,Right now .

[0048] It should be noted that a normal surface refers to an intact reference area adjacent to the defective area but not identified as a defect in the same key area image. The system automatically selects areas with uniform texture and color in this image as normal surface reference areas to ensure the objectivity of the comparison benchmark.

[0049] The transportation monitoring module is used to acquire vibration data during transportation through sensors installed on the equipment, and to analyze the impact of vibration accordingly.

[0050] Physical impacts during transportation are one of the main factors causing equipment damage. To objectively assess transportation quality, this invention includes a transportation monitoring module that collects dynamic vibration data in real time during transportation.

[0051] In one embodiment of the invention, an IoT sensor integrating GPS and a triaxial accelerometer is pre-installed on the machine. During transportation, the IoT sensor continuously collects triaxial acceleration data of the machine at a fixed sampling frequency, for example, 10Hz. , , .

[0052] The raw data is first filtered, and then the triaxial acceleration components at each sampling time point are synthesized according to the calculation formula: The resultant acceleration value is obtained. Arranging the combined acceleration values ​​of all sampling points in chronological order creates a vibration intensity sequence.

[0053] Next, the statistical standard deviation of the vibration intensity sequence is calculated to characterize the degree of dispersion of the vibration.

[0054] Then, all values ​​in the vibration intensity sequence are sorted in descending order, and the position estimation algorithm is used to determine the corresponding preset high percentile value in the sequence as an estimate to reflect the extreme vibration intensity encountered during transportation.

[0055] In this invention, the position estimation algorithm can be a linear interpolation method, and the calculation formula for the linear interpolation method is as follows: .

[0056] in, This represents the total number of samples in the vibration intensity sequence. , The theoretical position of the target percentile in the sequence, with a value range of 1- , This is the preset high percentile.

[0057] Assuming, , ,but Then, in the vibration intensity sequence arranged in descending order, the value corresponding to the 50th position is the estimated value. When If it is an integer, directly take the first integer in the sequence. One value is used as an estimate; when If it is a non-integer, then... The average of the values ​​in the corresponding sequence after rounding down and rounding up is used as the estimate.

[0058] Finally, the statistical standard deviation is multiplied by the estimated value. This product is the vibration impact. The larger the value, the higher the potential risk of damage to the machinery from vibration during transportation.

[0059] To avoid misjudgments due to a single indicator, the statistical standard deviation and preset high percentile are used to comprehensively consider both routine fluctuations and extreme shock events, thus fully assessing the potential impact of vibrations on machinery during transportation. For example, even with occasional extremely high turbulence during a period of continuous and stable transportation, the overall impact may be less than that of a period of continuous and severe turbulence.

[0060] It should be noted that the settings can be adjusted according to the precision and vibration resistance requirements of different machines. For precision equipment, the preset high percentile can be set to 99 to more sensitively capture extreme vibrations; for heavy and durable equipment, the preset high percentile can be set to 90 or 85.

[0061] Step S4: The inbound verification module is used to scan the electronic tag of the equipment when it arrives at the recipient, record the actual arrival time, and obtain the transportation timeliness deviation by combining the actual dispatch time and the planned transportation time; at the same time, it collects images of key parts, generates the inbound appearance index, and obtains the appearance anomaly by combining the outbound appearance index.

[0062] Once the equipment arrives at the receiving end, the system enters the warehousing verification phase, which aims to complete the closed loop of the transfer and evaluate the transportation results.

[0063] Specifically, when the equipment arrives at the receiving site, the receiving personnel scan its electronic tag, and the module records that moment as the actual arrival time.

[0064] First, please refer to Figure 4 For calculating the timeliness deviation in transportation: the inbound verification module obtains the actual dispatch time from the outbound record, and calculates the actual transportation time according to the formula: Actual transportation time = Actual arrival time - Actual dispatch time. .

[0065] Next, obtain the planned transportation duration from the scheduling instructions. Then, based on the preset transportation method, the system retrieves the corresponding standard allowable deviation from the preset table. Transportation timeliness deviation The calculation formula is: .

[0066] Taking the use of heavy trucks for road transportation of machinery as an example, the planned transportation time of the machinery can be found from the preset table as 10 hours. Under normal road conditions without extreme weather or sudden accidents, the actual transportation time fluctuates between 9.2 and 10.8 hours, and 9.5 hours can be taken. Considering the common uncontrollable factors such as traffic lights, temporary stops at service areas, and temporary speed limits on some road sections, a 2% buffer space is reserved for the system. Therefore, the allowable deviation range of the comprehensive setting standard is 10% of the planned transportation time, that is, 1 hour.

[0067] As a preferred embodiment of the present invention, if the system identifies that the planned transportation route contains known high-frequency congestion sections, a correction mechanism will be activated.

[0068] It should be noted that transportation routes can be queried and frequently congested road sections can be identified through the integrated traffic big data platform.

[0069] The correction mechanism specifically involves obtaining the actual and theoretical transport times for the same route segment from historical data for the same time period, calculating the difference between the two, and using this difference as the average delay time. Then, according to the calculation formula The net transport timeliness deviation was obtained. .

[0070] Finally, use the corrected formula: This allows for a fairer assessment of the timeliness performance of transport providers.

[0071] Meanwhile, regarding the judgment of appearance anomalies: upon entry into the warehouse, the system adopts the same image acquisition and processing procedure as upon exit, generating the current entry appearance index of the equipment. Next, the difference between the current exit appearance index and the entry appearance index is calculated, and the maximum value between this difference and zero is taken to avoid negative value interference caused by image recognition errors.

[0072] Then, to eliminate the interference of slow and uniform appearance degradation caused by normal use, wind and sun exposure on the judgment of abnormal damage, the system retrieves the outbound appearance index and inbound appearance index of the same type and the same path of equipment in the historical flow tasks of a preset number of times from the database. Then, it first calculates the difference between the current outbound appearance index and the inbound appearance index, and then calculates the average of all differences as the natural aging benchmark value.

[0073] As a preferred example, the preset number of times can be set according to the turnover frequency of the machine type. For example, for general-purpose machines with high-frequency turnover, the preset number of times can be 15 times, and for special machines with low-frequency turnover, the preset number of times can be 6 times.

[0074] Finally, if the current change in appearance is greater than the natural aging baseline value, it is determined that the appearance of the machine has been abnormally damaged during this transfer and transportation, and the value of the current change in appearance is taken as the appearance abnormality; otherwise, it is determined that the appearance of the machine is not abnormal, and the appearance abnormality is marked as zero.

[0075] The record generation module is used to perform cross-validation analysis based on transportation timeliness deviation, vibration impact, and appearance anomalies, and generate traceability records based on the cross-validation conclusions to update the equipment status.

[0076] To achieve accurate attribution of responsibility, this invention designs a record generation module that performs fusion analysis on multi-dimensional data. The specific implementation of the record generation module includes two main parts: cross-validation analysis and equipment status updates.

[0077] Please see Figure 2 During the cross-validation analysis phase, the record generation module immediately obtains the vibration impact of this transfer and transportation after the receiver scans the electronic tag of the equipment, and constructs a feature vector by combining the vibration impact, appearance anomaly and transportation timeliness deviation.

[0078] Subsequently, the feature vector is input into a preset responsibility attribution model, which outputs mutually exclusive responsibility attribution probabilities; among which, the responsibility attribution probabilities include the responsibility probability of the originating party and the responsibility probability of the transporting party.

[0079] Next, the record generation module compares the numerical values ​​of the caller's liability probability and the transporter's liability probability, and takes the party with the highest probability value as the initial liability determination result.

[0080] Finally, in this invention, the system sets a confidence threshold of 0.75. If the highest probability value is greater than or equal to the confidence threshold, the corresponding responsibility determination signal is output, for example, the responsible party: the transporter. Otherwise, an uncertainty signal is output.

[0081] It should be noted that the confidence threshold is not limited to a fixed value and can be set differently depending on the value of different equipment. For example, for high-value, precision equipment, a confidence threshold of 0.85 can be used to minimize risk; for conventional equipment, a confidence threshold of 0.75 can be maintained. As the performance of the attribution model continues to improve and stabilize, the confidence threshold can be gradually lowered, for example, to 0.7, to improve the level of system automation.

[0082] Please see Figure 3 During the equipment status update phase, the record generation module combines transportation timeliness deviation, vibration impact, appearance abnormality, and liability determination signals or uncertainty signals to generate a complete traceability record.

[0083] Meanwhile, the status of the equipment is updated based on the appearance abnormality and the liability determination result: if the appearance abnormality is greater than zero, regardless of the liability determination result, that is, regardless of whether a liability determination signal or an uncertainty signal is received, for safety reasons, the status of the equipment is updated to "in storage and awaiting maintenance"; this avoids equipment with potential safety hazards from being directly put into use next time, thus ensuring construction safety.

[0084] Under the above conditions, if a responsibility determination signal is received, the corresponding responsible party will be associated with the traceability record simultaneously; otherwise, the responsible party will not be associated, and the traceability record will remain unchanged.

[0085] If the appearance anomaly is zero, the equipment status is updated to "in stock and awaiting evaluation." Similarly, if a responsibility determination signal is received, the corresponding responsible party is also linked in the traceability record. Otherwise, no responsible party is linked, and the traceability record remains unchanged.

[0086] It should be noted that the "to be evaluated" status refers to a situation where the equipment has no visible abnormalities but other indicators may be questionable, such as excessive vibration. In this case, the receiving party's management personnel are allowed to conduct further manual verification.

[0087] Furthermore, the process of constructing the responsibility attribution model is explained as follows: First, collect data on vibration impact, appearance anomalies, transportation timeliness deviations, and identified responsible parties from historical transfer task data to form a dataset; among which, the responsible party can be confirmed manually or through arbitration.

[0088] Then, the dataset was randomly divided into training, validation, and test sets in a 7:2:1 ratio. Vibration impact, appearance anomalies, and transportation timeliness deviations from the training set were used as input features, and the responsible party was used as the target variable.

[0089] Next, supervised training is performed using a gradient boosting tree model. During training, the classification mapping relationship between features and responsible parties is established by minimizing the loss function, thereby constructing the initial responsibility attribution model.

[0090] Subsequently, the initial model was optimized using the validation set to prevent overfitting.

[0091] Finally, the responsibility attribution labels of the test set are predicted by the initial responsibility attribution model. The initial responsibility attribution model is adjusted and optimized by the classification accuracy. When the performance index (accuracy) reaches the preset standard, the final responsibility attribution model is output.

[0092] It should be noted that in the initial stage of system operation, in order to start the automation process as soon as possible, a slightly relaxed standard can be temporarily accepted, such as a preset standard of 85%. As the system gradually stabilizes, the preset standard can be increased to 90%. In addition, for the circulation of equipment with particularly high value or extremely high security requirements, a more stringent standard can be adopted, set at 93%.

[0093] The data storage module is used to associate traceability records, updated equipment status, and unique equipment numbers, and store them in the database.

[0094] All data from the entire process is ultimately archived by the data storage module. The system first binds the traceability records and updated equipment status with the equipment's unique number, and then uses blockchain technology to store the data in a distributed database or relational database, ensuring data security, traceability, and rapid retrieval.

[0095] In summary, this invention achieves refined and automated monitoring of the entire machinery circulation process through closed-loop management of scheduling triggering, outbound recording, transportation monitoring, inbound verification, record generation, and data storage. In particular, through multi-dimensional data cross-verification and intelligent analysis of vibration data, appearance images, and transportation timeliness, it can effectively trace and define the responsible party for machinery damage, greatly reducing management costs and disputes, improving the management efficiency and transparency of machinery circulation between construction units, and also significantly improving the collaborative efficiency and asset management level among multi-level construction units.

[0096] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0097] Those skilled in the art will recognize that the modules 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 implementations should not be considered beyond the scope of this invention.

[0098] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0099] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0100] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A traceability management system for the transfer of machinery and equipment among multi-level construction units, characterized in that, include: The scheduling trigger module is used to generate scheduling instructions that include the dispatcher code, receiver code, equipment unique number and planned transportation duration, and to mark the equipment status. The outbound record module is used to scan the electronic tag of the equipment when it leaves the sending location, record the actual sending time, collect images of key parts, and generate an outbound appearance index. The transportation monitoring module is used to acquire vibration data during transportation through sensors installed on the machinery, and to obtain the vibration impact. The inbound verification module is used to scan the electronic tag of the equipment when it arrives at the recipient, record the actual arrival time, and combine the actual dispatch time and the planned transportation time to obtain the transportation timeliness deviation; at the same time, it collects images of key parts, generates an inbound appearance index, and combines it with the outbound appearance index to obtain the appearance anomaly. The record generation module is used to perform cross-validation analysis based on transportation timeliness deviation, vibration impact, and appearance anomalies, and generate traceability records based on the cross-validation conclusions to update the equipment status. The data storage module is used to associate traceability records, updated equipment status, and unique equipment numbers with the equipment and store them in the database.

2. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 1, characterized in that, The process of generating a dispatch instruction containing the sender's code, receiver's code, equipment's unique number, and planned transportation duration, and marking the equipment's status, specifically involves: Obtain the receiver's code, the sender's code, and the required equipment type; Based on the required equipment type, select equipment with a status of "available for transfer" from the equipment management information of the dispatching party, identify them as equipment to be dispatched, and obtain their unique numbers. The theoretical transportation distance is obtained based on the difference in geographical coordinates between the sender and receiver. The planned transportation time is the product of the theoretical transportation distance and the time required for the reference unit distance. The dispatcher's code, receiver's code, equipment's unique number, and planned transportation time are used as standardized dispatch instructions, and the equipment's system status is updated to "pending dispatch".

3. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 1, characterized in that, The image generation of the outbound appearance index from key parts of the acquisition equipment includes: Images of key parts of the equipment are captured when the equipment leaves the receiving location, and defect areas in the images of key parts are identified. Calculate the ratio of the total area of ​​the defective region to the total area of ​​the image to obtain the area proportion; The pixel-level drop of each defect region relative to the normal surface is obtained by image feature analysis, and the average value of all pixel-level drops is taken as the average defect depth. Multiplying the area percentage by the average defect depth yields the overall damage level; The difference between the appearance index of the equipment during its last entry into the warehouse and the overall damage level is used as the appearance index upon exiting the warehouse.

4. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 1, characterized in that, The method of acquiring vibration data during transportation through sensors installed on the machinery to determine the impact of vibration specifically involves: The triaxial acceleration of the equipment during transportation is continuously collected by sensors installed on the equipment at a fixed sampling frequency; The triaxial acceleration components at each sampling time point are synthesized to obtain the resultant acceleration value. All the resultant acceleration values ​​are arranged in chronological order to form a vibration intensity sequence. Calculate the statistical standard deviation of the vibration intensity sequence, sort all values ​​in the vibration intensity sequence in descending order, and determine the value of the corresponding preset high percentile in the sequence through the position estimation algorithm as the estimated value; The product of the statistical standard deviation and the estimated value is used as the vibration impact.

5. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 2, characterized in that, The actual arrival time is recorded and combined with the actual dispatch time and planned transportation duration to obtain the transportation timeliness deviation, specifically: The difference between the actual arrival time and the actual dispatch time is taken as the actual transportation time. Obtain the standard allowable deviation corresponding to the preset transportation method; Calculate the difference between the actual transport time and the planned transport time, and use the ratio of this difference to the standard allowable deviation as the transport timeliness deviation.

6. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 5, characterized in that, The process for obtaining transportation timeliness deviation also includes: Identify whether the transportation route on which the planned transportation time is based contains known high-frequency congested road sections; if so, obtain the actual transportation time and theoretical transportation time of the road section in the historical transportation data of the same time period, and calculate the difference between the two as the average delay time. The net transport time deviation is obtained by subtracting the average delay time from the difference between the actual transport time and the planned transport time. The ratio of the net transport time deviation to the standard allowable deviation is used as the corrected transport time deviation.

7. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 3, characterized in that, The process of generating the inbound appearance index and combining it with the outbound appearance index to obtain appearance anomalies is as follows: Obtain the outbound and inbound appearance indices from the historical flow task data of the same type of machinery with the same transportation route, calculate the difference, and use the average of all differences as the natural aging benchmark value. Similarly, the current inbound appearance index is obtained by obtaining the outbound appearance index. The difference between the current outbound appearance index and the inbound appearance index is calculated, and the maximum value of the difference between the outbound appearance index and zero is taken as the current appearance change. If the current change in appearance is greater than the natural aging baseline value, it is determined that the appearance of the machine has been abnormally damaged during this transfer and transportation, and the current change in appearance is taken as the appearance abnormality; otherwise, it is determined that the appearance of the machine is not abnormal, and the appearance abnormality is marked as zero.

8. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 7, characterized in that, The cross-validation analysis based on transportation timeliness deviation, vibration impact, and appearance anomalies includes: After the recipient scans the electronic tag of the equipment, the vibration impact of this transfer and transportation is obtained, and a feature vector including vibration impact, appearance abnormality and transportation timeliness deviation is constructed. The feature vector is input into a preset responsibility attribution model, and mutually exclusive responsibility attribution probabilities are output; the responsibility attribution probabilities include the responsibility probability of the originating party and the responsibility probability of the transporting party; The party with the highest probability between the transporter's and the sender's liabilities is used as the initial liability determination result. If the highest probability value is greater than or equal to the confidence threshold, output the responsibility determination signal corresponding to the initial responsibility determination result; otherwise, output the uncertainty signal.

9. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 8, characterized in that, The process of generating traceability records and updating equipment status based on cross-validation conclusions specifically involves: Deviations in transportation timeliness, impact of vibration, abnormal appearance, and signals or uncertainties in liability determination are all recorded as traceability data. If the appearance abnormality is greater than zero, regardless of whether a responsibility determination signal or an uncertainty signal is received, the status of the machine will be updated to "in storage and awaiting maintenance"; when a responsibility determination signal is received, the corresponding responsible party will be associated with the traceability record simultaneously. If the appearance anomaly is zero, the status of the machine is updated to "in stock and awaiting evaluation"; when a responsibility determination signal is received, the corresponding responsible party is simultaneously associated in the traceability record.

10. The traceability management system for the transfer of machinery and equipment among multi-level construction units according to claim 8, characterized in that, The process of constructing the responsibility attribution model includes: The dataset is constructed by collecting data on vibration impact, appearance anomalies, transportation timeliness deviations, and identified responsible parties from historical transfer task data. The dataset is divided into training, validation and test sets according to a set ratio. Vibration impact, appearance anomaly and transportation timeliness deviation in the training set are used as input features and the responsible party is used as the target variable. The gradient boosting tree model is used for supervised training. The classification mapping relationship between the input features and the target variable is established by minimizing the loss function to build the initial responsibility attribution model. The initial responsibility attribution model is used to predict the responsibility attribution labels of the test set. The initial responsibility attribution model is then adjusted and optimized based on the classification accuracy, and the final responsibility attribution model is output.

Images