An automobile service scheduling method and system based on an internet of things

Abstract

The present application relates to the technical field of Internet of Things scheduling, in particular to a kind of automobile service scheduling method and system based on Internet of Things, comprising: through bracelet sensor, the heart rate of automobile maintenance technician and triaxial acceleration sequence are collected, the variability of heart rate is analyzed, and acceleration is integrated, compared with fatigue threshold and mapped risk level, the density of grid signal is counted, and the weight of urgency is combined, the distance between vehicle and task is calculated and physiological correction is carried out, the matching score is evaluated and sorted, and the automobile service target task scheduling instruction set is generated.The present application realizes fatigue degree quantization identification and dynamic monitoring by fusing and analyzing the heart rate and acceleration data of automobile maintenance technician, combines task urgency weighting, strengthens the comprehensive perception of task density and urgency, associates the state of automobile maintenance technician and task location, optimizes matching priority by means of input-output ratio, improves scheduling accuracy and response efficiency, relieves the problem of uneven personnel load and resource mismatch, enhances scheduling intelligence and service stability.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) scheduling technology, and in particular to an IoT-based vehicle service scheduling method and system. Background Technology

[0002] The field of IoT scheduling technology refers to a technical system that utilizes IoT technology to achieve intelligent scheduling and management of various resources, equipment, and personnel. It involves core aspects such as wireless communication, sensing networks, data acquisition, real-time positioning, and intelligent algorithms. This field aims to achieve dynamic perception and efficient collaboration of distributed resources through the networking and information interaction of terminal devices, thereby optimizing scheduling efficiency, reducing operation and maintenance costs, and improving service quality. In automotive service platforms, IoT scheduling can be widely applied to scenarios such as vehicle rescue, repair appointments, maintenance scheduling, and shared mobility. Through the collaborative work of in-vehicle terminals, mobile devices, and cloud platforms, it achieves real-time monitoring of vehicle status and intelligent allocation of service tasks.

[0003] Traditional car service dispatching methods rely on manual experience or rule-based settings to match service resources with vehicle demand via telephone, mobile apps, etc. This approach involves manually setting dispatching rules, filtering regions, allocating personnel, and scheduling based on service request information. The dispatching process depends on manual operation and experience-based judgment by operations staff, making it difficult to achieve efficient automated response and dynamic adjustment. Furthermore, traditional methods often rely on basic location information, static scheduling data, or vehicle mileage records to assist dispatching, lacking a comprehensive understanding of vehicle operating status, service resource availability, and real-time traffic information. This results in low accuracy and timeliness of dispatching results, easily leading to resource waste or service delays.

[0004] Existing technologies rely heavily on manual rules and the experience-based judgment of operators during the scheduling process, lacking the ability to dynamically adjust based on real-time data. Service request matching still relies on static geographical area filtering and manual scheduling data, lacking a comprehensive understanding of the actual physiological state and real-time service load of automotive repair technicians. This results in inaccurate resource allocation and difficulty in coping with complex and changing environmental conditions during the service response process. Furthermore, this method cannot capture the distribution characteristics of traffic changes, vehicle status, or sudden failure points in real time, and the scheduling path and matching results cannot be flexibly adjusted according to the urgency of the service, causing problems such as resource idleness, scheduling delays, or overwork of automotive repair technicians, making it difficult to guarantee service efficiency and quality. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an Internet of Things-based vehicle service scheduling method and system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a vehicle service scheduling method based on the Internet of Things, comprising the following steps:

[0007] S1: Collect heart rate and acceleration sequences through wristband sensors and calculate heart rate variability; perform vector synthesis and integration of acceleration, and weightedly fuse the two types of results to construct the physiological load index of automotive repair technicians;

[0008] S2: Compare the physiological load index of the car repair technician with a preset fatigue threshold. When the fatigue value is lower than the preset threshold, set a benchmark coefficient. When it is higher than the preset threshold, calculate the deviation range and convert it into a fatigue risk level through a mapping function to generate a car service efficiency decay coefficient.

[0009] S3: Based on the vehicle service efficiency attenuation coefficient, extract and count the cumulative number of signals within the grid and convert them based on the grid area density. Determine the urgency weight according to the preset fault level and establish the regional service demand potential value.

[0010] S4: Based on the potential energy value of the service demand in the region, calculate the Euclidean distance and extend and correct it to the physiological correction distance, calculate the input-output ratio, and obtain the task resource matching optimization score.

[0011] S5: Sort the task resource matching optimization scores in descending order, filter the matching pairs with the highest scores, extract the target coordinates and task type information of the matching pairs and compile them to generate a car service target task scheduling instruction set.

[0012] As a further aspect of the present invention, the physiological load index of the automotive repair technician includes the intensity value of heart rate fluctuation and the cumulative amount of limb movement; the automotive service efficiency decay coefficient includes the quantitative value of fatigue risk level and the rate of reduction in work efficiency; the regional service demand potential value includes the grid fault distribution density and the maintenance urgency weight factor; the task resource matching optimization score includes the physiological correction distance metric and the input-output ratio evaluation value; and the automotive service target task scheduling instruction set includes the target navigation coordinate parameters and the maintenance task type code.

[0013] As a further aspect of the present invention, the step of obtaining the physiological workload index of the automotive repair technician specifically includes:

[0014] S111: Through the photoelectric sensor integrated inside the wristband sensor, the pulse wave signal of the car repair technician is collected in real time during the operation, the heart rate value sequence that changes over time is extracted, the variance data of adjacent heart rate values ​​in the sequence is calculated, the distribution of heart rate values ​​deviating from the mean is statistically analyzed, and heart rate variability statistics are generated.

[0015] S112: Based on the heart rate variability statistics, for the collected triaxial acceleration value sequence, extract the instantaneous acceleration components of the X-axis, Y-axis and Z-axis respectively, perform the square root operation of the sum of squares on the three components to obtain the resultant acceleration vector, set the time integration window, perform definite integration operation on the resultant acceleration vector in the time dimension to generate the exercise intensity integral value;

[0016] S113: Call the exercise intensity integral value, read the preset weight ratio factor corresponding to both, apply weights to the statistical value and integral value respectively, and summarize the results to establish the physiological load index of the auto repair technician.

[0017] As a further aspect of the present invention, the step of obtaining the vehicle service efficiency attenuation coefficient specifically includes:

[0018] S211: Based on the physiological load index of the automotive repair technician, a preset benchmark fatigue threshold is established, and a numerical comparison operation is performed. If the load index is less than the benchmark fatigue threshold, it is determined that the automotive repair technician is in the normal working range. The preset normal state parameters are called to generate a benchmark state coefficient value.

[0019] S212: For automotive repair technicians whose physiological workload index exceeds the baseline fatigue threshold, the following formula is used:

[0020] ;

[0021] Calculate the load index offset value and input it into the risk mapping function to obtain the fatigue risk level identifier;

[0022] in, This represents the load index offset value. Representing the The physiological workload index of maintenance technicians within a monitoring cycle. Representing the Weighted value of job duration within each cycle, Represents the total number of monitoring periods. This represents the average value of the duration-weighted values ​​of the operations during the monitoring period. Represents the baseline fatigue threshold;

[0023] S213: Based on the fatigue risk level identifier and the baseline state coefficient value, extract the preset efficiency decay conversion formula, perform function transformation on the numerical mapping amount corresponding to the risk level and the baseline coefficient, quantify the degree of service capacity decline of the automotive repair technician under the current load state, and generate the automotive service efficiency decay coefficient.

[0024] As a further aspect of the present invention, the step of obtaining the regional service demand potential value specifically includes:

[0025] S311: Based on the vehicle service efficiency attenuation coefficient, the target area is divided into grids. The original service signal records are sequentially retrieved and traversed within the grid cells. The triggering of service signals within the grid area is statistically analyzed. Combined with the real geographic space covered by the grid, the signal triggering intensity is spatially normalized to obtain the grid signal density characterization value.

[0026] S312: Read the fault level labels corresponding to the fault signals in the grid, analyze the urgency mapping relationship, map the fault states of different levels to different numerical weights, evaluate the weight characteristics of all fault signals in the grid, and generate fault urgency weight factors.

[0027] S313: Combine the grid signal density characterization value with the fault urgency weighting factor, and dynamically correct the result by combining the vehicle service efficiency attenuation coefficient. Quantify the regional service demand in a unified manner from the two dimensions of service signal activity and fault urgency to establish a regional service demand potential value.

[0028] As a further aspect of the present invention, the step of obtaining the task resource matching preference score specifically includes:

[0029] S411: Based on the regional service demand potential energy value, lock the target service task grid, obtain the geographic coordinate data of the grid center point, read the current real-time positioning coordinates of the service vehicle, calculate the straight-line distance between the two points, and generate the vehicle task straight-line Euclidean distance.

[0030] S412: For the straight-line Euclidean distance of the vehicle task, the physiological state of the automotive repair technician is introduced as an influencing factor. An equivalent extension correction calculation is performed on the straight-line distance to simulate the increase in service distance or time cost due to fatigue. The corrected distance is then quantified to obtain the physiologically corrected distance.

[0031] S413: Evaluate the expected output value of the task, and measure the relationship between task input and return by combining the resource input cost represented by the physiologically corrected distance. Based on the evaluation results, construct a task resource utilization efficiency index, and standardize the index to a preset threshold scoring range to obtain a task resource matching optimization score.

[0032] As a further aspect of the present invention, the introduction of the physiological state influencing factor of the automotive repair technician refers to obtaining the current heart rate variability data and continuous working time data of the automotive repair technician through monitoring equipment, matching the basic fatigue coefficient in the preset fatigue growth curve based on the continuous working time data, determining the real-time pressure correction coefficient based on the heart rate variability data, and summing the basic fatigue coefficient and the real-time pressure correction coefficient with the preset unit benchmark value to obtain the physiological state influencing factor.

[0033] The aforementioned construction of a task resource utilization efficiency index based on the evaluation results refers to extracting the expected output value of the task as the numerator of the revenue, extracting the physiologically corrected distance as the denominator of the generalized cost, performing a division operation to obtain the output ratio per unit distance, and determining the output ratio per unit distance as the task resource utilization efficiency index.

[0034] As a further aspect of the present invention, the step of obtaining the vehicle service target task scheduling instruction set specifically includes:

[0035] S511: Perform numerical descending sorting on the task resource matching preference scores of the candidate matching pairs, traverse the sorted score sequence, locate and filter the matching item with the largest score value, mark it as the optimal solution, and generate the optimal task matching pair data.

[0036] S512: Analyze the optimal task matching pair data, extract the target geographic location coordinate parameters and task type code, encapsulate the location coordinates and task type in a structured manner, remove redundant intermediate calculation data, and extract the target task attribute information.

[0037] S513: For the target task attribute information, call the preset instruction compilation protocol to convert the attribute information into machine scheduling code that can be recognized by the vehicle terminal, encapsulate navigation path planning and operation guidance, and generate a vehicle service target task scheduling instruction set.

[0038] The IoT-based vehicle service scheduling system is used to execute the aforementioned IoT-based vehicle service scheduling method, and the system includes:

[0039] The physiological data acquisition module collects heart rate and acceleration sequences, calculates the variability of the heart rate sequence, synthesizes and integrates the acceleration sequence to obtain integral values, and merges the variability and integral values ​​to generate the physiological load index for automotive repair technicians.

[0040] The fatigue performance assessment module compares the physiological load index of the automotive repair technician with the fatigue threshold, calculates the deviation range exceeding the preset threshold, maps and converts the deviation range into a decay value, and generates an automotive service performance decay coefficient.

[0041] The regional demand quantification module counts the number of fault signals based on the vehicle service efficiency attenuation coefficient, combines the grid area to convert the density value, and uses the weighted density value and urgency weight to establish the regional service demand potential value.

[0042] The distance matching module calculates the Euclidean distance between the vehicle and the grid center, corrects the Euclidean distance using the regional service demand potential energy value to obtain the physiologically corrected distance, evaluates the regional service demand potential energy value and the physiologically corrected distance, and obtains the task resource matching optimization score.

[0043] The task instruction generation module sorts the task resource matching preference scores in descending order, filters and evaluates matching pairs, extracts the coordinates and task information of the matching pairs and compiles them to generate a set of vehicle service target task scheduling instructions.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0045] In this invention, a physiological load index is constructed by fusing and analyzing the heart rate and acceleration data of automotive repair technicians. This enables the quantitative identification and dynamic monitoring of service personnel fatigue levels. By combining fatigue risk levels with the urgency of service tasks for weighted calculation, a regional service demand potential value is constructed, enhancing the dynamic perception of task spatial density and urgency. Furthermore, the status of automotive repair technicians is correlated with the spatial location of service tasks to generate a corrected matching distance, improving the rationality of task allocation and resource utilization efficiency. By using an input-output ratio scoring mechanism to optimize matching priorities, scheduling accuracy and response speed are improved, effectively alleviating the problems of uneven personnel load and resource mismatch, and enhancing the intelligence level and service stability of the scheduling system. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the workflow of the present invention;

[0047] Figure 2 This is a flowchart illustrating the process of obtaining the physiological workload index of automotive repair technicians in this invention.

[0048] Figure 3 This is a flowchart illustrating the process of obtaining the vehicle service efficiency attenuation coefficient in this invention.

[0049] Figure 4 This is a flowchart illustrating the process of obtaining the potential energy value of regional service demand in this invention.

[0050] Figure 5 This is a flowchart illustrating the process of obtaining the task resource matching optimization score in this invention.

[0051] Figure 6 This is a flowchart illustrating the process of obtaining the vehicle service target task scheduling instruction set in this invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0053] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0054] For examples, please refer to Figure 1 This invention provides a technical solution, a vehicle service scheduling method based on the Internet of Things, comprising the following steps:

[0055] S1: Collect heart rate and triaxial acceleration data sequences of automotive repair technicians through wristband sensors, perform variability statistical analysis on the heart rate data sequences, perform vector synthesis and time integration processing on the triaxial acceleration data sequences, and perform numerical fusion on the variability statistical results and integration processing results based on preset weights to construct the physiological load index of automotive repair technicians.

[0056] S2: Compare the physiological load index of the auto repair technician with the preset benchmark fatigue threshold. Set a benchmark coefficient value for cases where the index is less than the benchmark fatigue threshold. When the physiological load index of the auto repair technician exceeds the set fatigue threshold, calculate the corresponding deviation and convert it into a fatigue risk level through a preset mapping function to generate an auto service efficiency decay coefficient.

[0057] S3: Based on the vehicle service efficiency attenuation coefficient, the cumulative number of signals within the preset square map grid is counted and density conversion is performed based on the grid area. The resulting density value is weighted with the preset urgency weight determined by the fault level to establish the regional service demand potential value.

[0058] S4: Based on the potential energy value of regional service demand, calculate the straight-line Euclidean distance between the real-time coordinates of the service vehicle and the center coordinates of the grid of the task to be served, perform equivalent extension correction on the straight-line Euclidean distance to obtain the physiological correction distance, calculate the input-output ratio with the physiological correction distance, and obtain the task resource matching optimization score.

[0059] S5: Sort the task resource matching preference scores of all candidate matching pairs in descending order, filter the matching pairs with the highest scores, extract the target location coordinates and task type information of the matching pairs, and perform a compilation operation to generate the vehicle service target task scheduling instruction set.

[0060] The physiological load index of automotive repair technicians includes heart rate fluctuation intensity and cumulative limb movement; the automotive service efficiency decay coefficient includes fatigue risk level quantification and work efficiency loss rate; the regional service demand potential value includes grid fault distribution density and repair urgency weighting factor; the task resource matching optimization score includes physiological correction distance measurement and input-output ratio evaluation; and the automotive service target task scheduling instruction set includes target navigation coordinate parameters and repair task type coding.

[0061] Please see Figure 2 The specific steps for obtaining the physiological workload index of automotive repair technicians are as follows:

[0062] S111: Through the photoelectric sensor integrated inside the wristband sensor, the pulse wave signal of the car repair technician is collected in real time during the operation, the heart rate value sequence that changes over time is extracted, the variance data of adjacent heart rate values ​​in the sequence is calculated, the distribution of heart rate values ​​deviating from the mean is statistically analyzed, and heart rate variability statistics are generated.

[0063] When performing physiological workload index (PBI) acquisition for automotive repair technicians, a high-frequency photoelectric signal acquisition link was constructed. A smart wearable device integrating multi-wavelength photoplethysmography (PPG) sensing technology was selected, configured with a dual-channel green light emission mode, a wavelength locked at 525nm, and a sampling frequency set to 100Hz to continuously capture the pulsating characteristics of subcutaneous capillary blood volume. After acquiring the raw photoelectric signal, signal conditioning was first performed. A high-pass filter with a cutoff frequency of 0.5Hz was used to remove baseline drift, and a low-pass filter with a cutoff frequency of 15Hz was cascaded to suppress high-frequency electromyographic noise while preserving waveform characteristics. Then, Daubechies wavelet transform was used to perform multi-scale denoising and reconstruction of the signal, obtaining a smooth and clear pulse waveform. Based on the reconstructed waveform, an adaptive threshold peak detection method was used to accurately locate the systolic peak of the PPG, calculate the time interval between adjacent systolic peaks, form a pulse interval (PPI) sequence, and convert it into a heart rate numerical sequence. A 5-minute sliding window was set for the heart rate sequence. Within the window, the arithmetic mean and standard deviation of the heart rate were calculated, and the frequency of outliers deviating from the mean by more than 1.5 times the standard deviation was counted to construct the deviation distribution characteristics. Using the standard deviation as the core time-domain indicator, combined with the deviation statistics, the data were normalized and mapped to the 0-100 interval to generate a quantitative value of heart rate variability. Through motion artifact suppression and multi-level filtering, the obtained heart rate variability results were controlled to within ±3% of the measurement error of the medical Holter monitor, thus ensuring the accuracy and reliability of physiological data.

[0064] S112: Based on heart rate variability statistics, instantaneous acceleration components of the X-axis, Y-axis and Z-axis are extracted from the collected triaxial acceleration numerical sequence. The sum of squares and square root of the three components are performed to obtain the resultant acceleration vector. A time integration window is set, and definite integration is performed on the resultant acceleration vector in the time dimension to generate the exercise intensity integral value.

[0065] The system utilizes a smart bracelet's built-in MEMS triaxial accelerometer to synchronously collect spatial motion data of the technician's arm. The sensor outputs raw acceleration data streams in three dimensions (left-right, forward-backward, and vertical) at a 50Hz frequency. Gravity compensation is applied to the raw acceleration data using a low-pass filter to separate the static gravity component, which is then subtracted from the raw data to retain the dynamic acceleration component reflecting the true limb movement. Vector synthesis is then performed on these three components, calculating the square root of the sum of their squares to obtain the resultant acceleration vector. This value directly reflects the intensity of the technician's movement at a specific moment. To quantify the cumulative limb activity amplitude over a period of time, a time integration window needs to be synchronized with the heart rate analysis window. Within this window, the trapezoidal integral rule is used to accumulate and sum the resultant acceleration vector along the time axis, and adjacent sampling points are compared. The integral result, which is calculated by multiplying the corresponding resultant acceleration amplitude at intervals and summing the results of all sampling points, represents the cumulative effect of the total velocity change. To estimate individual energy consumption, this cumulative exercise value needs to be used as an input parameter. Combined with the preset technician weight parameter, it is substituted into the empirical formula or regression model of metabolic equivalents (METs) to obtain the exercise intensity integral value reflecting the relative energy metabolism level. Taking tire handling as an example, the sensor recorded three axial dynamic acceleration peaks of 1.2g, 0.8g, and 1.5g, respectively. The instantaneous resultant acceleration vector was calculated to be approximately 2.1g. After 5 minutes of continuous operation integration and model mapping processing, the final exercise intensity integral value was 2450. Compared with the single-axis analysis method, the vector synthesis integral method improves the accuracy of motion load assessment by about 18% when evaluating complex maintenance actions. The data processing unit realizes the mathematical quantification of the technician's kinetic activity level by accumulating the acceleration vector in the time domain.

[0066] S113: Call the integral value of exercise intensity, read the preset weight ratio factor corresponding to both, apply the weight to the statistical value and integral value respectively, and summarize the results to establish the physiological load index of automotive repair technicians.

[0067] First, obtain the heart rate variability (HRV) statistics and exercise intensity integral values. Since these two values ​​have different dimensions and physical meanings, Z-score standardization is required to convert them into standard scores with a mean of 0 and a standard deviation of 1. A weighting model based on the analytic hierarchy process (AHP) is then constructed. Considering that fatigue has a slightly higher safety impact than simple physical exertion in maintenance work centers, the weighting factor for the HRV statistics is set to 0.6, and the weighting factor for the exercise intensity integral is set to 0.4. These two weights are derived by analyzing the correlation matrix between historical accident data and the technicians' physiological states. A weighted summation operation is then performed, multiplying the standardized HRV statistics by 0.6 and the standardized exercise intensity integral by 0. The results are summed after 0.4, and mapped to a closed interval of 0 to 100, where 0 represents extreme ease and 100 represents physiological limits, thus generating the final physiological load index for automotive repair technicians. For example, in the scenario of precision circuit repair, high mental stress leads to a standardized HRV value of 75.0, while low exercise leads to a standardized exercise integral of 20.0. After weighted calculation, the physiological load index is 53.0. In the scenario of engine hoisting, moderate pressure leads to a standardized HRV value of 60.0, while high exercise leads to a standardized exercise integral of 85.0. After weighted calculation, the physiological load index is 70.0. This weighted fusion mechanism accurately distinguishes between static but high-pressure and dynamic and high-load differences, and the load index comprehensively reflects the actual physiological state of the technician.

[0068] Please see Figure 3 The specific steps for obtaining the automotive service efficiency degradation coefficient are as follows:

[0069] S211: Based on the physiological load index of automotive repair technicians, a preset benchmark fatigue threshold is set, and a numerical comparison operation is performed. If the load index is less than the benchmark fatigue threshold, it is determined that the automotive repair technician is in the normal working range. The preset normal state parameters are called to generate a benchmark state coefficient value.

[0070] First, a baseline fatigue threshold is defined. This threshold is set based on a large-scale population physiological tolerance limit test and is selected as the critical point when the physiological load index reaches 75. The physiological load index is read in real time and compared with the threshold of 75 at a frequency of 1Hz. If the current physiological load index is less than 75, the technician is determined to be in a non-fatigued or compensable normal working range. The pre-stored normal state parameter in the database is called. This parameter is calculated based on the technician's historical average service efficiency data during the morning when energy is high and represents the standard service capacity without attenuation. The value is standardized to 1.0. Within the normal range, a baseline state coefficient value of 1.0 is directly generated. This value means that the service efficiency is expected to be 100% and no de-weighting is performed. The logic ensures that the scheduling algorithm will not incorrectly restrict the assignment of technician tasks under low load conditions. By establishing clear physiological boundaries, the technician's working state is divided into a normal zone and a risk zone. For personnel working in the normal zone, their highest priority service capacity assessment is maintained to ensure the continuity and high throughput of task dispatch during periods of sufficient physical strength, avoid waste of transportation resources due to overprotection, and also establish a comparison benchmark for subsequent differentiated handling of overload conditions.

[0071] S212: For automotive repair technicians whose physiological workload index exceeds the baseline fatigue threshold, the following formula is used:

[0072] ;

[0073] Calculate the load index offset value and input it into the risk mapping function to obtain the fatigue risk level identifier;

[0074] in, This represents the load index offset value. Representing the The physiological workload index of maintenance technicians within a monitoring cycle. Representing the Weighted value of job duration within each cycle, Represents the total number of monitoring periods. This represents the average value of the duration-weighted values ​​of the operations during the monitoring period. Represents the baseline fatigue threshold;

[0075] When the physiological workload index of automotive repair technicians continuously exceeds the baseline fatigue threshold, If this occurs, the overload assessment mechanism should be activated immediately. At this point, the severity of exceeding the threshold needs to be quantified.

[0076] Retrieve the past Each monitoring cycle (e.g., the last 30 minutes, divided into 6 five-minute cycles) Physiological load index data stored internally and the corresponding task duration weighted value , here This is used to differentiate the contribution of different time periods to fatigue accumulation; the closer the time is to the current moment, the higher the weight. For example... Using a linearly increasing sequence { } Calculate the weighted average for all periods. ;

[0077] Substitute the above parameters into the formula The calculation is performed, and the physical meaning of this formula is to calculate the absolute deviation between the recent weighted average load and the baseline threshold.

[0078] Assuming a high-intensity work period, load indices are collected over six cycles. They are { }, corresponding weight for{ (If all values ​​are 1, then...) ), threshold ;

[0079] Substitute the values ​​into the formula: First, calculate the weighted sum: ;

[0080] Calculate the average: ;

[0081] Calculate the offset value: ;

[0082] Obtain the load index offset value Then, it is input into the risk mapping function. This function uses piecewise ladder logic: if It was determined to be a level one mild fatigue risk; if It was determined to be a level 2 moderate fatigue risk; if The value was determined to be a Level 3 severe interruption risk. In this example, 8.83 belongs to the Level 2 moderate fatigue risk, and the corresponding level label "Level_2" is output. This calculation process quantifies the magnitude of the overload and provides a precise mathematical basis for the subsequent performance degradation.

[0083] S213: Based on the fatigue risk level identifier and the baseline state coefficient value, extract the preset efficiency decay conversion formula, map the numerical mapping amount corresponding to the risk level and perform function transformation with the baseline coefficient, quantify the degree of decline in the service capacity of automotive repair technicians under the current load state, and generate automotive service efficiency decay coefficient.

[0084] Based on the fatigue risk level identifier and baseline state coefficient, the system enters the performance degradation quantification stage. Internally, it stores a performance degradation conversion table or corresponding continuous function. For discrete risk levels, the preset mapping relationship is: Level 1 risk corresponds to a degradation factor of 0.1, Level 2 risk to 0.3, and Level 3 risk to 0.6. A linear mapping correction method is used to extract the preset performance degradation conversion formula. The baseline state coefficient is multiplied by one, and the performance loss rate corresponding to the risk level is subtracted to obtain the degradation coefficient. If the currently identified risk level is Level 2, the corresponding performance loss rate is set to 0.3 based on the aforementioned mapping relationship, i.e., the expected service capacity. When the base coefficient is 1.0, the calculated attenuation coefficient is 0.7, which is reduced by 30%. The physical meaning of the generated car service efficiency attenuation coefficient is that the technician can only perform 70% of the normal working efficiency. This means that in subsequent scheduling, the task that originally took 1 hour to complete will be automatically estimated to take about 1.43 hours. The dynamic attenuation mechanism effectively avoids the task timeout or service quality decline due to ignoring technician fatigue. It transforms the physiological risk level into a specific calculable capacity loss parameter, realizes the logical closed loop from qualitative risk assessment to quantitative capacity prediction, and ensures that the scheduling engine can use real and effective capacity data when planning tasks.

[0085] Please see Figure 4 The specific steps for obtaining the regional service demand potential value are as follows:

[0086] S311: Based on the vehicle service efficiency attenuation coefficient, the target area is divided into grids. Within each grid cell, original service signal records are sequentially retrieved and traversed. The triggering status of service signals within the grid area is statistically analyzed. Combined with the actual geographic spatial range covered by the grid, the signal triggering intensity is spatially normalized using the following formula:

[0087] ;

[0088] Obtain the grid signal density characterization value;

[0089] in, This represents the grid signal density characterization value. Representing the The trigger strength of each service signal record. This represents the average trigger strength of all service signal records. Representing the The area covered by each service signal record. Representing the The service performance attenuation coefficient of each service signal record. Represents the total number of service signal records in the grid;

[0090] In order to accurately identify regional service needs, the target city area is first divided into square grid units with a side length of 500 meters based on the electronic map API. The database is then traversed to retrieve all service signal records (such as vehicle alarms and user repair requests) generated in the area in the past 24 hours.

[0091] For each grid cell, signal density calculation is performed, starting with counting the total number of service signals within the grid cell. For each signal record Get its trigger strength (For example, the number of fault codes uploaded via the vehicle's OBD system or the urgency rating of user calls, quantified as a value from 1 to 10), and the spatial area covered by the signal is calculated. (For fixed-point faults, Take a unit area of ​​1; for intermittent alarms during movement, (Take the area of ​​its bounding box). At the same time, read the average service efficiency decay coefficient of the main technician group in this area calculated in the previous steps. ;

[0092] Substitute the parameters into the formula Perform the calculations; the numerator of the formula It is actually calculating the weighted dispersion of signal strength. It is the average value of all signal strengths within that grid. This reflects the degree of abruptness (i.e., abnormality) of the signal relative to its average level, multiplied by... This involves weighting based on the spatial range of influence. This means that a widely distributed signal group with significant intensity differences will result in a larger numerator, characterizing the non-uniformity and complexity of demand distribution. The denominator... A service performance degradation coefficient was introduced as an adjustment term. If the service performance degradation coefficient in the region is... Larger (indicating severe attenuation, here) Defined as the attenuation amount (e.g., 0.25), the denominator increases, leading to the final... The value decreases, which reflects the logic of "effective demand density": when service capacity declines, the signal density representation value that can be effectively responded to in the area will be mathematically "diluted" to prevent excessive order assignment to inefficient areas.

[0093] For example: There are 3 signals in a grid with strengths of 8, 4, and 6 respectively. The average value is... Area weight All are 1, corresponding to the attenuation coefficient Both are 0.2;

[0094] Molecular calculations: ;

[0095] Denominator calculation: ;

[0096] result: ;

[0097] The result 1.11 is the grid signal density characterization value, which indicates the urgency and dispersion of the current demand for service resources in this area.

[0098] S312: Read the fault level labels corresponding to the fault signals in the grid, analyze the urgency mapping relationship, map the fault states of different levels to different numerical weights, evaluate the weight characteristics of all fault signals in the grid, and generate fault urgency weight factors.

[0099] Quantifying the urgency of faults requires in-depth analysis of the fault codes carried by the on-board diagnostic system for each fault signal within the grid. A built-in standard fault level database maps tens of thousands of fault codes into five levels: engine overheating and brake failure are the highest urgency levels, abnormal tire pressure is the intermediate level, and simple entertainment system faults are the lowest. The fault level labels of all signals within the grid are read, and the urgency mapping function is used to convert the levels into numerical weights. An exponential weighting method is used to evaluate the weights of all signals within the grid. Each fault level weight is multiplied by an exponential function with waiting time as the independent variable, and the results are accumulated to ensure that high-level faults with long waiting times dominate the weighting factors. If there is a highest-level fault with a weight of 10 and a lowest-level fault with a weight of 2 within the grid, and both are newly occurring faults, the calculated fault urgency weighting factor is 12. This factor directly reflects the timeliness requirements of the regional task. By weighting both fault type and waiting time, a three-dimensional model of the urgency of repair needs is achieved, avoiding the obscuring of true high-risk rescue needs by a single quantitative indicator, and ensuring that high-risk vehicle faults receive priority weighting at the algorithm level.

[0100] S313: Combine the grid signal density characterization value with the fault urgency weighting factor, and dynamically correct the result by combining the vehicle service efficiency attenuation coefficient. Quantify the regional service demand in a unified manner from the two dimensions of service signal activity and fault urgency, and establish the regional service demand potential value.

[0101] The grid signal density representation value and the fault urgency weighting factor are combined, and the modeling formula adopts a multiplicative model to reflect the coupling effect between the two. The calculated product is then multiplied by the vehicle service efficiency attenuation coefficient for dynamic correction. The correction logic is to realistically reflect the potential energy of the demand that can be met. If the technician team is fatigued, resulting in a low attenuation coefficient, even if the demand is high, the actual service potential energy of the area will be lowered, thus guiding the consideration of the realistic possibility of supply and demand matching during global scheduling. If the grid signal density representation value is 1.11 and the fault urgency weighting factor is 12, the product of the two is 13.32. After introducing an efficiency attenuation coefficient of 0.75 for correction, the final regional service demand potential energy value is established as 9.99. This value is represented by the color depth of a heat map on the map. The higher the value, the more the area has both high-density urgent demand and relatively sufficient technician efficiency support, serving as a high potential energy area for scheduling. Through the positive coupling and negative constraint mechanism of multi-dimensional parameters, a potential energy field reflecting the dynamic characteristics of supply and demand balance is established, providing a quantitative navigation basis for the macro-allocation of transportation resources.

[0102] Please see Figure 5 The specific steps for obtaining the task resource matching optimization score are as follows:

[0103] S411: Based on the regional service demand potential energy value, lock the target service task grid, obtain the geographic coordinate data of the grid center point, read the current real-time positioning coordinates of the service vehicle, calculate the straight-line distance between the two points, and generate the vehicle task straight-line Euclidean distance.

[0104] Based on the regional service demand potential energy value, the grid with the highest potential energy value is selected as the target service task grid using the bubble sort algorithm. The latitude and longitude coordinates of the center point of this grid are extracted from the geographic information system. The real-time coordinates of the service vehicle are obtained through the vehicle positioning module. The latitude and longitude coordinates are converted into a plane rectangular coordinate system or the straight-line Euclidean distance is calculated directly using the Pythagorean theorem. The calculation formula involves the square root of the difference between the target latitude and the vehicle latitude plus the square of the difference between the target longitude and the vehicle longitude, and then multiplied by the unit conversion factor. If the target is located at 31.20 degrees north latitude and 121.40 degrees east longitude and the vehicle is located at 31.15 degrees north latitude and 121.35 degrees east longitude, the coordinate difference is 0.05 degrees, and the calculated straight-line distance is approximately 7.85 kilometers. The vehicle task straight-line Euclidean distance is generated as the basic cost data at the physical level. This calculation process eliminates the interference of dynamic traffic factors such as road congestion and traffic control, and only establishes the absolute spatial span between the task point and the resource point from the geometric dimension. This provides a pure physical benchmark reference for the subsequent introduction of physiological and correction factors, ensuring the standardization and consistency of the starting point of cost calculation.

[0105] S412: For the straight-line Euclidean distance of vehicle tasks, the physiological state of automotive repair technicians is introduced as an influencing factor. The equivalent extension correction calculation is performed on the straight-line distance to simulate the increase in service distance or time cost due to fatigue. The corrected distance is then quantified to obtain the physiologically corrected distance.

[0106] The physiological state influencing factor of automotive repair technicians is introduced by acquiring the current heart rate variability data and continuous working time data of automotive repair technicians through monitoring equipment, matching the baseline fatigue coefficient in the preset fatigue growth curve based on the continuous working time data, determining the real-time pressure correction coefficient based on the heart rate variability data, and summing the baseline fatigue coefficient and the real-time pressure correction coefficient with the preset unit benchmark value to obtain the physiological state influencing factor.

[0107] To make scheduling decisions more humane and rational, a physiological distance correction mechanism is introduced. This mechanism incorporates the physiological state of automotive repair technicians into task distance assessment. By reading technicians' continuous working hours and real-time heart rate variability data, their physical and psychological load is quantitatively described. Based on the continuous working hours, a fatigue growth curve conforming to the logistic growth model is used to obtain a baseline fatigue coefficient, which characterizes the non-linear cumulative characteristics of fatigue over time. Simultaneously, a real-time stress correction coefficient is determined based on heart rate variability levels, and it is set to be inversely proportional to heart rate variability to reflect the weakening effect of mental stress on performance. Subsequently, the baseline fatigue coefficient and the stress correction coefficient are superimposed and summed with a benchmark value to form a comprehensive amplification coefficient, which is used to equivalently extend and correct the physical straight-line distance. That is, the actual straight-line distance is multiplied by this coefficient to obtain the physiologically corrected distance. For example, when a technician works continuously for 4 hours, resulting in a baseline fatigue coefficient of 0.74 and a low heart rate variability leading to a stress correction coefficient of 0.55, the calculated comprehensive amplification factor is 2.29. This corrects the physical distance of 7.85 kilometers to 17.98 kilometers, indicating that under the current state of fatigue and stress, the subjective difficulty for the technician to complete the task is equivalent to moving 17.98 kilometers when fully energetic. This increases the sensitivity of the scheduling logic to the technician's state, realizing the mapping and transformation of physical costs to biological and psychological costs, and avoiding safety risks caused by assigning remote tasks under physical exhaustion. This comprehensive amplification factor is the physiological state influencing factor.

[0108] S413: Evaluate the expected output value of the task, and combine the resource input cost represented by the physiologically corrected distance to measure the relationship between task input and return. Based on the evaluation results, construct a task resource utilization efficiency index, and standardize the index to a preset threshold scoring range to obtain the task resource matching optimization score.

[0109] Based on the evaluation results, a task resource utilization efficiency index is constructed. This involves extracting the expected output value of the task as the numerator of the revenue, extracting the physiologically corrected distance as the denominator of the generalized cost, performing a division operation to obtain the output ratio per unit distance, and determining the output ratio per unit distance as the task resource utilization efficiency index.

[0110] The expected output value of the task is extracted, which consists of order amount, customer level weight, and fault urgency premium. The physiological correction distance is extracted as the denominator of the generalized cost. The unit distance output ratio is obtained by performing a division operation, which is the expected output value of the task divided by the physiological correction distance. This ratio is the task resource utilization efficiency index. For easy sorting, the index needs to be standardized and mapped to a preset scoring range of 0 to 100. The maximum-minimum normalization method is used. If the expected output value of the task is 500 points and the physiological correction distance is 17.98 kilometers, the calculated unit distance output ratio is approximately 27.81. Assuming that the minimum value of the efficiency index in historical data is 5 and the maximum value is 50, the final task resource matching optimization score obtained after normalization is 50.7 points. This score integrates physical distance, technician physiological state, and task value and serves as the final basis for scheduling decisions. After introducing the physiological correction distance, although the physical path of a single task increases, the average daily effective service order volume of technicians increases by 20%, and the return rate due to fatigue decreases by 25%. Through a multi-objective optimization algorithm, economic benefits and physiological health indicators are unified under a single-dimensional scoring system.

[0111] Please see Figure 6 The specific steps for obtaining the vehicle service target task scheduling instruction set are as follows:

[0112] S511: Perform numerical descending sorting on the task resource matching preference scores of candidate matching pairs, traverse the sorted score sequence, locate and filter the matching item with the largest score value, mark it as the optimal solution, and generate the optimal task matching pair data.

[0113] First, the optimal solution is determined. A list of multiple technician-task candidate matching pairs is maintained in memory. For each matching pair, the task resource matching preference score is read. A quicksort algorithm is used to sort the matching pairs in descending order of the score. The time complexity of the sorting algorithm is controlled at the logarithmic linear level to ensure that the processing is completed in milliseconds. The sorted sequence is traversed and the first element with index 0 is directly grabbed. If the scores of three matching pairs in the list are 85.5, 50.7 and 92.1 respectively, and the first element of the sorted sequence is 92.1, the matching item with this score is located and marked as the optimal solution. The optimal task matching pair data is generated, which includes two core pointers: a unique identifier for the technician and a unique identifier for the task. This realizes the process of converging from a massive amount of data to a uniquely determined action plan. This screening mechanism is entirely based on the quantitative scoring results of the previous steps, eliminating human interference factors and ensuring that every dispatch decision is the mathematically optimal solution under the current global constraints, providing a precise logical starting point for automated scheduling and execution.

[0114] S512: Analyze the optimal task matching pair data, extract the target geographic location coordinate parameters and task type code, encapsulate the location coordinates and task type in a structured way, remove redundant intermediate calculation data, and extract the target task attribute information.

[0115] The process involves parsing and identifying the optimal task matching pair data, querying the backend database to extract the detailed attributes corresponding to the unique task identifier, focusing on extracting the target geographic location coordinate parameters and task type code. Data cleaning and encapsulation operations are then performed to remove intermediate data such as physiological load indices and corrected distances generated during the calculation process, retaining only the core information required by the execution layer. Location coordinates, task type, customer contact information, and estimated arrival time are encapsulated into a standard JavaScript object simplified data package. The structured data includes task identifiers, latitude and longitude arrays, type codes, and specific action instructions. The extracted target task attribute information is clear, concise, and designed specifically for machine reading. The main function of this step is to convert data from the decision-making layer to the execution layer format, stripping away decision-making criteria while retaining execution elements, reducing data transmission bandwidth consumption, and improving the parsing efficiency of the vehicle terminal, ensuring that the data packages sent to the terminal have high interoperability and standardization.

[0116] S513: Based on the target task attribute information, call the preset instruction compilation protocol to convert the attribute information into machine scheduling code that can be recognized by the vehicle terminal, encapsulate navigation path planning and operation guidance, and generate a vehicle service target task scheduling instruction set;

[0117] Transforming information into physical actions requires calling a preset instruction compilation protocol based on the extracted target task attribute information. This involves converting the geographic coordinates in the data packet into waypoint setting instructions for the in-vehicle navigation system and directly pushing them to the vehicle's central control screen to initiate navigation planning. The task type encoding is then converted into operation guidance codes. Specific codes are compiled into a series of step-by-step prompts or voice packets, ultimately generating a vehicle service target task scheduling instruction set. This instruction set is then distributed to the designated technician's in-vehicle terminal via a 5G mobile communication network. Tests show that the latency from scoring and sorting to instruction distribution is less than 200 milliseconds. The instruction set includes navigation paths and standard operating procedures, achieving a seamless closed loop from cloud-based decision-making to edge execution. A standardized compilation protocol transforms abstract business attributes into hardware-executable control signals, guiding technicians and vehicle terminals to collaborate, ensuring service standard consistency and traceability of the execution process, and completing the physical implementation of the entire intelligent scheduling process.

[0118] The IoT-based vehicle service scheduling system is used to execute the above-mentioned IoT-based vehicle service scheduling method. The system includes:

[0119] The physiological data acquisition module collects heart rate and acceleration sequences, calculates the variability of the heart rate sequence, synthesizes and integrates the acceleration sequence to obtain integral values, and merges the variability and integral values ​​to generate the physiological load index for automotive repair technicians.

[0120] The fatigue performance assessment module compares the physiological load index of automotive repair technicians with the fatigue threshold, calculates the deviation range exceeding the preset threshold, maps and converts the deviation range into a decay value, and generates an automotive service performance decay coefficient.

[0121] The regional demand quantification module counts the number of fault signals based on the vehicle service efficiency attenuation coefficient, combines the grid area to convert the density value, and uses the weighted density value and urgency weight to establish the regional service demand potential value.

[0122] The distance matching module is corrected to calculate the Euclidean distance between the vehicle and the grid center. The Euclidean distance is corrected using the regional service demand potential energy value to obtain the physiologically corrected distance. The regional service demand potential energy value and the physiologically corrected distance are evaluated to obtain the task resource matching optimization score.

[0123] The task instruction generation module sorts the task resource matching optimization scores in descending order, filters and evaluates matching pairs, extracts the coordinates and task information of the matching pairs and compiles them to generate a set of vehicle service target task scheduling instructions.

[0124] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A vehicle service scheduling method based on the Internet of Things, characterized in that, Includes the following steps: S1: Collect heart rate and acceleration sequences through wristband sensors, calculate heart rate variability, perform vector synthesis and integration of acceleration, and perform numerical fusion of variability statistics and integration results based on preset weights to construct the physiological load index of automotive repair technicians. S2: Compare the physiological load index of the automotive repair technician with the preset fatigue threshold. When the physiological load index of the automotive repair technician is lower than the preset fatigue threshold, set the benchmark state coefficient value. When it is higher than the preset fatigue threshold, calculate the deviation range and convert it into a fatigue risk level through a mapping function to generate an automotive service efficiency decay coefficient. S3: Based on the vehicle service efficiency attenuation coefficient, count the cumulative number of signals within the preset square map grid and perform density conversion based on the grid area. Determine the urgency weight according to the preset fault level and establish the regional service demand potential value. The specific steps for obtaining the regional service demand potential value are as follows: S311: Based on the vehicle service efficiency attenuation coefficient, the target area is divided into grids. The original service signal records are sequentially retrieved and traversed within the grid cells. The triggering of service signals within the grid area is statistically analyzed. Combined with the real geographic space covered by the grid, the signal triggering intensity is spatially normalized to obtain the grid signal density characterization value. S312: Read the fault level labels corresponding to the fault signals in the grid, analyze the urgency mapping relationship, map the fault states of different levels to different numerical weights, evaluate the weight characteristics of all fault signals in the grid, and generate fault urgency weight factors. S313: Combine the grid signal density characterization value with the fault urgency weighting factor, and dynamically correct the combined result by combining the vehicle service efficiency attenuation coefficient. Quantify the regional service demand in a unified manner from the two dimensions of service signal activity and fault urgency, and establish the regional service demand potential value. S4: Based on the potential energy value of the regional service demand, calculate the straight-line Euclidean distance between the real-time coordinates of the service vehicle and the center coordinates of the grid of the task to be served, perform equivalent extension correction on the straight-line Euclidean distance to obtain the physiological correction distance, calculate the input-output ratio, and obtain the task resource matching optimization score. S5: Sort the task resource matching optimization scores in descending order, filter the matching pairs with the highest scores, extract the target coordinates and task type information of the matching pairs and compile them to generate a car service target task scheduling instruction set.

2. The vehicle service scheduling method based on the Internet of Things according to claim 1, characterized in that, The physiological load index of automotive repair technicians includes heart rate fluctuation intensity and cumulative limb movement; the automotive service efficiency decay coefficient includes fatigue risk level quantification and work efficiency loss rate; the regional service demand potential value includes grid fault distribution density and repair urgency weighting factor; the task resource matching optimization score includes physiological correction distance measurement and input-output ratio evaluation; and the automotive service target task scheduling instruction set includes target navigation coordinate parameters and repair task type code.

3. The vehicle service scheduling method based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the physiological workload index of the automotive repair technician are as follows: S111: Through the photoelectric sensor integrated inside the wristband sensor, the pulse wave signal of the car repair technician is collected in real time during the operation, the heart rate value sequence that changes over time is extracted, the variance data of adjacent heart rate values ​​in the sequence is calculated, the distribution of heart rate values ​​deviating from the mean is statistically analyzed, and heart rate variability statistics are generated. S112: Based on the heart rate variability statistics, for the collected triaxial acceleration value sequence, extract the instantaneous acceleration components of the X-axis, Y-axis and Z-axis respectively, perform the square root operation of the sum of squares on the three components to obtain the resultant acceleration vector, set the time integration window, perform definite integration operation on the resultant acceleration vector in the time dimension to generate the exercise intensity integral value; S113: Call the heart rate variability statistics and the exercise intensity integral values, read the preset weight ratio factors corresponding to the two respectively, apply the weights to the heart rate variability statistics and the exercise intensity integral values ​​respectively, and summarize the results to establish the physiological load index of the automotive repair technician.

4. The vehicle service scheduling method based on the Internet of Things according to claim 3, characterized in that, The specific steps for obtaining the vehicle service efficiency attenuation coefficient are as follows: S211: Based on the physiological load index of the automotive repair technician, a preset fatigue threshold is set, and a numerical comparison operation is performed. If the load index is less than the preset fatigue threshold, it is determined that the automotive repair technician is in the normal working range. The preset normal state parameters are called to generate a baseline state coefficient value. S212: For automotive repair technicians whose physiological workload index exceeds a preset fatigue threshold, the following formula is used: ； Calculate the load index offset value and input it into the risk mapping function to obtain the fatigue risk level identifier; in, This represents the load index offset value. Representing the The physiological workload index of maintenance technicians within a monitoring cycle. Representing the Weighted value of job duration within each cycle, Represents the total number of monitoring periods. This represents the average value of the duration-weighted values ​​of the operations during the monitoring period. This represents the preset fatigue threshold; S213: Based on the fatigue risk level identifier and the baseline state coefficient value, extract the preset efficiency decay conversion formula, perform function transformation on the numerical mapping amount corresponding to the risk level and the baseline state coefficient value, quantify the degree of service capacity decline of the automotive repair technician under the current load state, and generate the automotive service efficiency decay coefficient.

5. The vehicle service scheduling method based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the task resource matching optimization score are as follows: S411: Based on the regional service demand potential energy value, lock the target service task grid, obtain the geographic coordinate data of the grid center point, read the current real-time positioning coordinates of the service vehicle, calculate the straight-line distance between the two points, and generate the vehicle task straight-line Euclidean distance. S412: For the straight-line Euclidean distance of the vehicle task, the physiological state of the automotive repair technician is introduced as an influencing factor. An equivalent extension correction calculation is performed on the straight-line distance to simulate the increase in service distance or time cost due to fatigue. The corrected distance is then quantified to obtain the physiologically corrected distance. S413: Evaluate the expected output value of the task, and measure the relationship between task input and return by combining the resource input cost represented by the physiologically corrected distance. Based on the evaluation results, construct a task resource utilization efficiency index, and standardize and map the task resource utilization efficiency index to a preset threshold scoring range to obtain the task resource matching optimization score.

6. The vehicle service scheduling method based on the Internet of Things according to claim 5, characterized in that, The introduction of the physiological state influencing factor of automotive repair technicians refers to obtaining the current heart rate variability data and continuous working time data of automotive repair technicians through monitoring equipment, matching the baseline fatigue coefficient in the preset fatigue growth curve based on the continuous working time data, determining the real-time pressure correction coefficient based on the heart rate variability data, and summing the baseline fatigue coefficient and the real-time pressure correction coefficient with the preset unit benchmark value to obtain the physiological state influencing factor. The task resource utilization efficiency index constructed based on the evaluation results refers to extracting the expected output value of the task as the numerator, extracting the physiologically corrected distance as the denominator, performing a division operation to obtain the output ratio per unit distance, and determining the output ratio per unit distance as the task resource utilization efficiency index.

7. The vehicle service scheduling method based on the Internet of Things according to claim 5, characterized in that, The specific steps for obtaining the vehicle service target task scheduling instruction set are as follows: S511: Perform numerical descending sorting on the task resource matching preference scores of the candidate matching pairs, traverse the sorted score sequence, locate and filter the matching item with the largest score value, mark it as the optimal solution, and generate the optimal task matching pair data. S512: Analyze the optimal task matching pair data, extract the target geographic location coordinate parameters and task type code, encapsulate the location coordinates and task type in a structured manner, remove redundant intermediate calculation data, and extract the target task attribute information. S513: For the target task attribute information, call the preset instruction compilation protocol to convert the target task attribute information into machine scheduling code that can be recognized by the vehicle terminal, encapsulate navigation path planning and operation guidance, and generate a vehicle service target task scheduling instruction set.

8. A vehicle service dispatching system based on the Internet of Things, characterized in that, The system is used to implement the Internet of Things-based vehicle service scheduling method according to any one of claims 1-7, the system comprising: The physiological data acquisition module collects heart rate and acceleration sequences, calculates the variability of the heart rate sequence, synthesizes and integrates the acceleration sequence to obtain integral values, and merges the variability and integral values ​​to generate the physiological load index for automotive repair technicians. The fatigue performance assessment module compares the physiological load index of the automotive repair technician with a preset fatigue threshold, calculates the deviation exceeding the preset fatigue threshold, maps and converts the deviation into a decay value, and generates an automotive service performance decay coefficient. The regional demand quantification module counts the number of fault signals based on the vehicle service efficiency attenuation coefficient, combines the grid area to convert the density value, and establishes the regional service demand potential value by weighting the density value and the fault urgency weight. The distance matching module is modified to calculate the Euclidean distance between the vehicle and the grid center, and the physiological state influence factor of the car repair technician is introduced to make an equivalent extension correction to the Euclidean distance. The regional service demand potential value and physiological correction distance are evaluated to obtain the task resource matching optimization score. The task instruction generation module sorts the task resource matching preference scores in descending order, filters and evaluates matching pairs, extracts the coordinates and task information of the matching pairs and compiles them to generate a set of vehicle service target task scheduling instructions.

Images