Drug logistics transportation monitoring method and system

Abstract

The invention relates to the technical field of electric data processing, in particular to a medicine logistics transportation monitoring method and system, and the method comprises the steps: collecting temperature data in a box body on a transport vehicle; the validity of the temperature data at the current moment is calculated, the validity is the bumping strength of the box body at the moment and the optimal matching degree of a temperature change sequence at the current moment in the current state and a standard temperature sequence after the box body is normally opened, and the length of the temperature change sequence at the current moment is consistent with that of the standard temperature sequence; and acquiring a temperature prediction value at the next moment based on a sliding moving average method. According to the method, the size of the moving average window is dynamically adjusted by combining the matching degree of the jolting strength and the temperature change trend, and self-adaptive smoothing processing of the temperature data is achieved. Noise can be suppressed during jolting, sensitivity can be kept during real fluctuation, and therefore prediction accuracy is improved, and the false alarm rate of abnormal detection is reduced.

Description

Technical Field

[0001] This invention relates to the field of electrical data processing technology. More specifically, this invention relates to a method and system for monitoring the logistics and transportation of pharmaceuticals. Background Technology

[0002] In the field of pharmaceutical transportation, ensuring the temperature stability of medicines throughout the entire transportation process is crucial, as many medicines are highly sensitive to temperature; exceeding a certain temperature range can lead to reduced efficacy, spoilage, or even inactivation. Refrigerated trucks, used for the road transport of frozen or fresh goods such as frozen foods, dairy products, fruits and vegetables, vaccines, and pharmaceuticals, are enclosed box trucks with independent refrigeration units. They can be used for cold chain transportation of pharmaceuticals, maintaining the required storage temperatures to ensure the quality of the medicines during transport.

[0003] In related technologies, for example, Chinese patent application document with publication number CN119204907A discloses a method, product, equipment and medium for temperature monitoring during drug transportation. By acquiring the temperature of different drug categories and different locations inside the box in real time, multi-point temperature monitoring can provide more comprehensive temperature data, which helps to more accurately assess the temperature distribution inside the box, reduce the errors that may be caused by a single monitoring point, and compare the measured temperature with the temperature range required for drug transportation. This can ensure that the drug is always within the specified temperature range throughout the transportation process, thereby ensuring the quality and safety of the drug.

[0004] Currently, when using the moving average method to predict the temperature inside a transport container, it does not fully consider factors such as varying degrees of vehicle jolting due to complex road conditions during transportation, and the instantaneous temperature rise and rapid recovery caused by the short-term opening of the container door during unloading, inspection, or customs clearance. Although these short-term temperature fluctuations do not constitute a real cold chain breakdown, they appear as abrupt changes or abnormal fluctuations in the data collected by sensors. This causes traditional prediction methods to lag in their response to temperature changes or make misjudgments, resulting in inaccurate prediction results and making it impossible to accurately monitor abnormal temperatures inside the transport container. Summary of the Invention

[0005] This invention provides a method and system for monitoring pharmaceutical logistics transportation, aiming to solve the problem that traditional prediction methods in related technologies are slow to respond to or misjudge temperature changes, resulting in inaccurate prediction results and thus failing to accurately monitor abnormal temperatures inside the transport container.

[0006] In a first aspect, the present invention provides a method for monitoring pharmaceutical logistics transportation, wherein an alarm device is installed in the driver's cab of a transport vehicle. The monitoring method includes: collecting temperature data inside the container of the transport vehicle; determining a standard temperature sequence in response to the container being in an open mode; obtaining a predicted temperature value for the next moment based on a moving average method; determining whether to issue a warning through the alarm device based on the difference between the predicted temperature value and a first theoretical value, wherein the first theoretical value is determined based on the standard temperature sequence, and the window size of the moving average method is from the moment the container is opened to the current moment; the method for obtaining the standard temperature sequence includes: obtaining a temperature sequence composed of the duration of each time the container is opened in the current state and all temperature data within that duration from historical data; calculating the coverage rate of temperature data at each relative moment in all temperature sequences; taking the moment when the earliest coverage rate is less than a threshold as the cutoff node; calculating the average temperature at the same moment before the cutoff node; and arranging all the average temperature values ​​in chronological order to obtain the standard temperature sequence. By using the matching degree between the intensity of bumps during transportation and the trend of temperature changes as dual indicators of the validity of temperature data, and combining sliding window adaptive adjustment and standard temperature sequence matching mechanism, the accuracy and real-time performance of temperature anomaly detection in cold chain transportation environment are significantly improved.

[0007] Furthermore, in response to the transport vehicle being in a closed state, the validity of the temperature data at the current moment is calculated, where validity is the reciprocal of the turbulence intensity of the container at that moment. The predicted temperature value for the next moment is obtained based on the moving average method. Whether to issue an early warning is determined based on the difference between the predicted temperature value and the second theoretical value. The window size of the moving average method is negatively correlated with validity, and the second theoretical value is the temperature data with the closest validity exceeding the validity threshold to the current moment. By using the reciprocal of the container turbulence intensity as the validity index of the temperature data, the reliability of sensor data under different operating conditions can be dynamically identified. The greater the turbulence intensity, the more likely the data fluctuation is caused by interference, resulting in lower validity; conversely, data under stable conditions is more reliable and has higher validity, thus providing quality assurance for subsequent processing.

[0008] Furthermore, the method for obtaining the coverage includes: statistically analyzing the proportion of temperature data existing at each relative time in all temperature sequences, i.e., the coverage at that time.

[0009] Furthermore, the validity of the current temperature data is calculated, including: using a triaxial accelerometer to collect the acceleration values ​​of the enclosure in the x, y, and z directions at a preset frequency; calculating the standard deviation of the magnitude of the acceleration vector between the current and previous moments as the turbulence intensity at that moment; and using the reciprocal of the turbulence intensity as the validity of the current temperature data. By using the standard deviation of the acceleration vector as the turbulence intensity evaluation index, a quantitative judgment on the validity of the current temperature data is achieved.

[0010] Furthermore, the acceleration vector between the current moment and the previous moment is calculated using the following formula: In the formula, This represents the magnitude of the triaxial acceleration vector at the i-th sampling point at time t. , , Let x, y, and z represent the squares of the acceleration values ​​of the i-th sampling point in the x, y, and z directions, respectively, at time t.

[0011] Furthermore, the method for obtaining the matching degree includes: constructing a target window based on the length of the temperature change sequence at the current moment and in the current state; performing a sliding match between the target window and the standard temperature sequence, wherein the sliding step size is 1, and calculating the matching degree between the temperature data of the overlapping part of the target window and the standard temperature sequence after each sliding, and selecting the one with the highest matching degree as the best matching degree at the current moment and in the current state. This sliding window matching mechanism achieves similarity assessment between the temperature change sequence and the standard sequence, maximizing matching accuracy and temporal adaptability, and improving the accuracy of judging whether the current temperature fluctuation conforms to a normal behavioral pattern.

[0012] Furthermore, the matching degree between the target window and the temperature data of the overlapping part of the standard temperature sequence after each slide is calculated, including: using the DTW algorithm to calculate the matching degree between the target window and the temperature data of the overlapping part of the standard temperature sequence after each slide, wherein the matching degree is negatively correlated with the distance value.

[0013] Furthermore, the matching degree between the target window and the temperature data of the overlapping part of the standard temperature sequence after each sliding is calculated, including: the matching degree is not only related to the distance between the target window and the temperature data of the overlapping part of the standard temperature sequence after sliding, but also to the consistency of the changing trend between the two sequences within the overlapping window.

[0014] Furthermore, the final window calculation formula includes: In the formula, This represents the final window corresponding to time t. This represents the minimum value of the window, where the empirical value for the minimum window is 5. This represents the maximum window size, where the empirical value for the maximum window size is 20. Indicates the first The validity of temperature data at any given time. By fusing the matching degree between turbulence intensity and temperature change, and dynamically adjusting the size of the moving average window, accurate adaptation to different disturbance environments and behavioral patterns can be achieved.

[0015] In a second aspect, the present invention also provides a pharmaceutical logistics transportation monitoring system, comprising a processor, a memory, and an alarm device, wherein the memory stores a computer program, and the processor executes the computer program to implement the pharmaceutical logistics transportation monitoring method described in any of the above embodiments.

[0016] Beneficial effects: By combining the matching degree between turbulence intensity and temperature change trend, the size of the moving average window is dynamically adjusted to achieve adaptive smoothing of temperature data. This suppresses noise during turbulence while maintaining sensitivity during real fluctuations, thereby improving prediction accuracy and reducing false alarm rates in anomaly detection. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the monitoring of a transport container according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the sliding matching of a target window according to an embodiment of the present invention. Detailed Implementation

[0018] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] The moving average is a commonly used time series smoothing technique. By averaging data within a certain time window, it reduces the impact of random fluctuations and noise, thus better reflecting the overall trend of the data. Its basic principle is to calculate the arithmetic mean of the data at each time point and the data over the preceding period. The moving window continuously moves forward over time, updating the average in real time. Specifically, a fixed-length window is set, and the sum of all data within the window is divided by the window length to obtain the smoothed value at the current moment. As the window slides, old data is removed, new data is added, and the average is dynamically updated.

[0020] It is important to note that the choice of window length has a significant impact on the smoothing effect: a longer window can effectively suppress noise, but the response speed is slower, which may lead to a delayed response to actual changes; a shorter window is more sensitive to the response, but its noise suppression ability is weaker, which may lead to larger fluctuations in the prediction results. Therefore, this embodiment adaptively determines the window length based on the changes in door opening and closing patterns during transportation to improve the accuracy of the prediction results. The specific steps are as follows.

[0021] like Figure 1 As shown, S101: Collects environmental parameters inside the transport container.

[0022] In one embodiment, during the transportation of pharmaceuticals, because these drugs have strict environmental requirements—for example, insulin needs to be stored in a low-temperature environment—precise control of the environmental characteristics (temperature or humidity) within the transport container is necessary to meet the transport requirements. Specifically, temperature sensors are installed on the inner wall of the transport container to monitor the temperature data inside the container in real time. Additionally, an accelerometer is installed on the transport container to obtain its motion status, which is used to calculate the degree of vehicle vibration. An alarm device is installed in the driver's cab of the transport vehicle to warn the driver. The alarm device is an indicator light; when a warning is issued, the indicator light remains illuminated to alert the driver.

[0023] S102: Determine the window size for the closed mode, and use the moving average method and the above window size to predict the temperature, and issue an early warning based on the predicted value.

[0024] In one embodiment, during vehicle operation in closed mode, if the road surface contains potholes, speed bumps, or other uneven areas, the vehicle will experience significant bumps, manifesting as vertical or lateral vibrations, impacts, or swaying. These bumps are transmitted to the transport vehicle compartment and physically disturb various environmental sensors (such as temperature sensors) installed within it. Since the sensors are mounted on the compartment structure, when the vehicle experiences bumps, their bodies may experience short-term displacement, shaking, or poor contact, or their internal measuring elements may be mechanically disturbed, causing instantaneous fluctuations in the electrical signals. These fluctuations appear numerically as abrupt changes or severe jitter, but they do not reflect the actual changes in the compartment's internal environment (such as temperature or humidity); they are merely spurious signal fluctuations caused by mechanical impact. Therefore, to avoid these abnormal fluctuations affecting data accuracy and subsequent judgment, the sensor output data needs to be smoothed. Considering that the greater the intensity of the turbulence, the higher the probability and amplitude of signal fluctuations, a larger sliding window should be used for smoothing to enhance the filtering's ability to suppress impact noise. This will increase the buffering of abrupt data during the filtering process and improve the system's data stability under severe turbulence scenarios.

[0025] S1021: Calculate the turbulence intensity at any given time.

[0026] In one embodiment, a triaxial accelerometer is selected to monitor the motion state of the transport vehicle. In this embodiment, the sampling frequency of the triaxial accelerometer is set to 100Hz, and the sliding window is set to 1 second. For each moment, the turbulence intensity within the previous second is calculated as the turbulence intensity at that moment.

[0027] In one embodiment, a calculation formula is provided to calculate the turbulence intensity at any given time. The formula is as follows: In the formula, This represents the intensity of the turbulence at time t. This represents the magnitude of the triaxial acceleration vector at the i-th sampling point at time t. This indicates the number of sampling points within the window (obtained by multiplying the sampling frequency by the sliding window). This represents the average value of the acceleration modulus within the current window. ( ) represents the normalization function. Here, the standard deviation of the acceleration modulus of all sampling points within the previous second is calculated to measure the severity of acceleration changes during that time period, i.e., the bump intensity of the vehicle at that moment. The standard deviation reflects the dispersion of the data; that is, the more drastic the vehicle acceleration changes, the larger the standard deviation of the three-axis acceleration modulus, and thus the greater the bump intensity at that moment. Then, the reciprocal of the bump intensity at the current moment is calculated and normalized. The normalized value is used as the validity of the temperature data at the current moment; that is, the greater the bump intensity at that moment, the lower the validity of the temperature data at that moment.

[0028] Following on from the above, the formula for calculating the triaxial acceleration vector of the i-th sampling point at time t is: In the formula, This represents the magnitude of the triaxial acceleration vector at the i-th sampling point at time t. , , These represent the squares of the acceleration values ​​of the i-th sampling point in the x, y, and z directions at time t, respectively. The bumps experienced by a vehicle during transportation often come from multiple directions (up / down, front / back, left / right) simultaneously. Considering only a single direction (such as the X-axis) would miss disturbances in other directions; by calculating the acceleration modulus, the overall vibration intensity experienced by the vehicle at that moment can be comprehensively reflected.

[0029] S1022: Determine the window size corresponding to the sliding moving average method for the closed mode.

[0030] In one embodiment, the moving average method can be used to predict the temperature at the next moment. Specifically, a window size is set, and the temperature at the next moment is predicted based on the temperature value within the window. A larger window can effectively suppress abrupt changes and sensor fluctuations, while a smaller window can detect real changes and quickly capture real-time changes. Therefore, the window size can directly affect the accuracy of the prediction. Furthermore, the window size can be adaptively adjusted according to the temperature change trend at the current moment to improve the accuracy of the prediction.

[0031] In one embodiment, the formula for calculating the window size is as follows: In the formula, This represents the final window corresponding to time t. This represents the minimum value of the window, where the empirical value for the minimum window is 5. This represents the maximum window size, where the empirical value for the maximum window size is 20. Indicates the first The validity of the temperature data at any given time. The larger the value, the more significant the th... The less reliable the real-time temperature data, the greater the likelihood of noise. A larger window should be used to reduce the impact of noise. In other words, when the turbulence is severe and the data reliability is poor, the sliding window can be adaptively enlarged to enhance noise suppression.

[0032] S1023: Obtain the temperature prediction value for the next moment based on the moving average method, and determine whether to issue an early warning based on the difference between the temperature prediction value and the second theoretical value.

[0033] In one embodiment, if the transport vehicle is in a closed state, the validity of the temperature data at the current moment is calculated, where validity is the reciprocal of the bump intensity of the container at that moment. The predicted temperature value for the next moment is obtained based on the moving average method. Whether to issue an alarm is determined based on the difference between the predicted temperature value and the second theoretical value. If the absolute value of the difference is greater than the warning threshold, an indicator light is kept on to warn the driver and remind them to perform maintenance. Otherwise, the temperature inside the container continues to be monitored. The warning threshold is 1 degree Celsius, and the difference is the absolute value of the difference between the predicted temperature value and the first theoretical value. The window size of the moving average method is negatively correlated with validity. The second theoretical value is the temperature data with the closest validity greater than the validity threshold to the current moment. The empirical value of the validity threshold is 0.7. For example, if the temperature values ​​from the current moment to the previous K moments are 3.9, 3.6, 3.5, and 3.55, their validity values ​​are 0.3, 0.7, 0.8, and 0.8 respectively. That is, the validity of the moment corresponding to the temperature value of 3.5 is greater than 0.8, so the temperature value of 3.5 is taken as the second theoretical value.

[0034] S103: Determine the window size for the door opening mode, and use the moving average method and the above window size to predict the temperature, and issue an early warning based on the predicted value.

[0035] In one embodiment, during operations such as temporary unloading, checking the condition of medicines, or customs clearance, logistics personnel may open the vehicle door, resulting in a short-term (tens of seconds to minutes) rapid increase in temperature followed by a return to normal. This situation typically lasts only tens of seconds to minutes, and the temperature fluctuation is a real fluctuation, not constituting a true cold chain disruption. Therefore, when using the moving average method to predict the temperature at the next moment, a smaller window should be used to ensure the predicted value keeps pace with the actual temperature changes, reflects real-time dynamics, avoids misjudgments or delayed responses, and improves the sensitivity (response speed) and real-time performance of responding to short-term real fluctuations.

[0036] S1031: Calculate the degree of matching between the current temperature change sequence and the standard temperature sequence.

[0037] Specifically, to determine whether the temperature change inside the transport container after the door is opened is a true fluctuation, it is necessary to obtain the standard temperature sequence under the current state and calculate the matching degree between the temperature change sequence after the door is opened and the standard temperature sequence. If the matching degree is larger, the temperature change inside the transport container after the door is opened is more likely to be a true fluctuation. In this case, a smaller window should be used to retain short-term changes to reflect the actual state, thereby improving the sensitivity and real-time performance of the response to short-term true temperature fluctuations and avoiding misjudgments caused by prediction lag or over-smoothing. Conversely, if the matching degree is small, it indicates that the current temperature change does not conform to the standard change pattern and may be caused by sensor noise, sudden disturbances, or non-true fluctuations. In this case, a larger sliding window should be used to smooth the data.

[0038] In one embodiment, a method for obtaining a standard temperature sequence under the current state is provided, comprising: determining the current temperature state of the transport container, wherein the temperature state is the temperature difference between the inside and outside of the container, and taking the current temperature difference at each moment as the current state at each moment. Obtaining the historical temperature difference segment where the current temperature difference is located, wherein the temperature difference segment consists of multiple temperature differences within a temperature threshold range above and below the current temperature difference, for example, current temperature difference ± ΔT, where T is the temperature threshold, and the temperature threshold is 1 degree. In other embodiments, the temperature threshold can be 0.5 degrees or 0.7 degrees. For any temperature difference segment, collecting a temperature sequence composed of the duration of the container door being opened and all temperature data within that duration in chronological order, and calculating the standard temperature sequence corresponding to that temperature difference segment based on all temperature sequences within that segment, wherein the frequency of collecting temperature data after each door opening is equal. It should be noted that because the duration of each door opening is different, the length of each temperature sequence may be different, therefore, the standard temperature sequence needs to be calculated based on the length of each temperature sequence. Specifically, firstly, the proportion of temperature data existing at each relative moment in all temperature sequences is statistically analyzed, i.e., the coverage at that moment. Coverage indicates the proportion of historical temperature sequences that have valid records at a given moment, meaning temperature data was collected at that moment. A coverage threshold of 80% is set. If the coverage at any moment is below this threshold, it is considered insufficient, and all data from that moment onwards are excluded from the construction of the standard temperature sequence. The earliest relative moment on the timeline where coverage falls below the threshold is used as the cutoff node. Temperature data corresponding to all relative moments before this node are extracted, and the average of all temperature values ​​at each moment is calculated. The average values ​​of all relative moments within this time period are arranged chronologically, and the resulting temperature sequence is the standard temperature sequence under the current temperature difference state. The number of moments in the standard temperature sequence is used as the target window size for subsequent calculations of the matching degree.

[0039] For example, the current temperature difference of the transport container is 10°C, corresponding to a historical temperature difference range of [missing information]. Retrieve all door opening records for this historical temperature range, as follows: Door opening record 1: [4.0, 4.2, 4.5, 4.8, 5.0, —, —], where "—" indicates no temperature data was collected; Door opening record 2: [4.1, 4.3, 4.6, 5.0, 5.3, —, —]; Door opening record 3: [4.0, 4.1, 4.3, 4.6, 4.9, 5.0, 5.2]. Calculate the coverage rate of temperature data occurring at the same time. The temperature data coverage was 100% at all 5 time points, but the coverage was 33% at t=5. At this point, only [the data was] retained. The data for these five time points were truncated due to insufficient data for other time points. Then, the mean of all temperature data at the same time point for these five time points was calculated. The sequence of the mean values ​​of all temperature data in chronological order is the standard temperature sequence.

[0040] In another embodiment, if the length of the standard temperature sequence does not meet the requirements, the longest duration of historical door opening can be obtained. More data can be obtained through multiple trials to obtain a standard temperature sequence of sufficient length.

[0041] In one embodiment, when the enclosure is opened, the matching degree between the temperature change sequence in the current state and the standard temperature sequence in the same state is calculated. Specifically, a target window is constructed based on the length of the temperature change sequence in the current state at the current moment, where the temperature data at the current moment is at the end of the target window. The target window is slid across the standard temperature sequence with a step size of 1. Each time the window is slid, the matching degree between the temperature data of the overlapping part of the target window and the standard temperature sequence is calculated, thereby obtaining multiple matching degrees. The maximum value is taken as the best matching degree at the current moment. The sliding process is as follows: the last data of the target window is matched with the first data of the standard temperature sequence, and the window is slid along the time sequence with a step size of 1.

[0042] For example, such as Figure 2 As shown, X is the target window constructed from the temperature change sequence at the current time and in the current state. The size of the target window is the same as the length of the standard temperature sequence. B is the standard temperature sequence corresponding to the current time and in the current state. The first slide matches the temperature data at the end of the target window with the first temperature data of the standard temperature sequence. The second slide moves in chronological order, matching the last two temperature data in the target window with the first two temperature data of the standard temperature sequence.

[0043] In one embodiment, the DTW algorithm is used to calculate the distance between the target window and the temperature data of the overlapping part of the standard temperature sequence after each slide, and the reciprocal of the distance value is used as the matching degree between the target window and the temperature data of the overlapping part of the standard temperature sequence; the larger the distance value, the smaller the matching degree, and the smaller the distance value, the larger the matching degree.

[0044] In another embodiment, a formula for calculating the matching degree is also provided: In the formula, This indicates the degree of matching between the temperature change sequence in the current state after the sliding motion and the standard temperature sequence in the same state. This represents the o-th temperature value in the temperature change sequence within the overlapping portion after sliding matching. This represents the o-th temperature value in the standard temperature sequence within the overlapping portion after sliding matching. This indicates the total number of temperature data points overlapping between the target window and the standard temperature sequence after this sliding motion. This indicates that the temperature data within the overlapping region of the target window and the standard temperature series exhibits the same trend as the standard temperature data. Hyperparameters The purpose is to avoid the inability to calculate when the denominator is 0.

[0045] As mentioned above, the trend of change is consistent. The method for obtaining the data is as follows: acquire the temperature data of the overlapping portion between the target window and the standard temperature series. Calculate the ratio of temperature values ​​at the same time point to obtain the temperature ratios at each time point, and use the variance of all temperature ratios as the consistency of the trend. It should be noted that when there is only one data point in the overlapping portion between the target window and the standard temperature series, the variance cannot be calculated. To avoid affecting subsequent calculations, the consistency of the trend is used as the basis for the calculation. The value is 0.

[0046] S1032: Obtain the temperature prediction value for the next moment based on the moving average method, and determine whether to issue an early warning based on the difference between the temperature prediction value and the first theoretical value.

[0047] Specifically, the predicted temperature value for the next moment is obtained based on the moving average method. Whether to issue a warning is determined by the difference between the predicted temperature value and the first theoretical value. The difference is the absolute value of the difference between the predicted temperature value and the first theoretical value. If the absolute value of the difference is greater than the warning threshold, the indicator light will remain on to warn the driver and remind them to perform maintenance. Otherwise, the temperature inside the container continues to be monitored. The warning threshold is 1 degree Celsius. It is important to note that the first theoretical value is determined based on a standard temperature sequence. That is, the standard temperature value for the next moment in the standard temperature sequence can be used as the first theoretical value for the next moment. The window size of the moving average method is from the moment the door is opened to the current moment. Using temperature data after the door is opened for prediction can improve the accuracy of the prediction. For example, if the next moment is t3, then the standard temperature value at moment t3 in the standard temperature sequence is used as the first theoretical value for the next moment t3. This completes the monitoring of the drug transportation process.

[0048] The present invention also provides a pharmaceutical logistics transportation monitoring system. The system includes a processor and a memory, the memory storing computer program instructions, which, when executed by the processor, implement a pharmaceutical logistics transportation monitoring method according to the first aspect of the present invention.

[0049] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and therefore will not be described in detail here.

[0050] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained on such a computer-readable medium.

[0051] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for monitoring pharmaceutical logistics and transportation, characterized in that, in, An alarm device is installed in the driver's cab of the transport vehicle, and the monitoring method includes: Collect temperature data inside the container on the transport vehicle; In response to the container being in the open position on the transport vehicle, a standard temperature sequence is determined; The temperature prediction value for the next moment is obtained based on the moving average method. The difference between the temperature prediction value and the first theoretical value is used to determine whether to issue an alarm. The first theoretical value is determined based on a standard temperature sequence, and the window size of the moving average method is from the moment the door was opened to the current moment. The method for obtaining the standard temperature sequence includes: obtaining the duration of each opening of the box in the current state and the temperature sequence composed of all temperature data within that duration from historical data; calculating the coverage of temperature data at each relative moment in all temperature sequences; taking the earliest moment when the coverage is less than a threshold as the cutoff node; calculating the average temperature at the same moment before the cutoff node; and arranging all the average temperature values ​​in chronological order to obtain the standard temperature sequence.

2. The method for monitoring pharmaceutical logistics and transportation according to claim 1, characterized in that, In response to the transport vehicle being in a closed state, the validity of the temperature data at the current moment is calculated, wherein the validity is the reciprocal of the turbulence intensity of the container at that moment; The temperature prediction value for the next moment is obtained based on the moving average method. Whether to issue an early warning is determined based on the difference between the temperature prediction value and the second theoretical value. The window size and effectiveness of the moving average method are negatively correlated. The second theoretical value is the temperature data with the nearest effective value that is greater than the effectiveness threshold at the current moment.

3. The method for monitoring pharmaceutical logistics and transportation according to claim 1, characterized in that, The method for obtaining the coverage includes: The proportion of temperature data existing at each relative time point in all temperature series is counted, i.e., the coverage rate at that time point.

4. The method for monitoring pharmaceutical logistics and transportation according to claim 1, characterized in that, Calculate the validity of the current temperature data, including: The acceleration values ​​of the box in the x, y, and z directions are collected using a triaxial accelerometer at a preset frequency; Calculate the standard deviation of the magnitude of the acceleration vector between the current moment and the previous moment, and use it as the turbulence intensity at that moment. Use the reciprocal of the turbulence intensity as the validity of the temperature data at the current moment.

5. The method for monitoring pharmaceutical logistics and transportation according to claim 4, characterized in that, The magnitude of the acceleration vector between the current moment and the previous moment is calculated using the following formula: ； In the formula, This represents the magnitude of the triaxial acceleration vector at the i-th sampling point at time t. , , Let x, y, and z represent the squares of the acceleration values ​​of the i-th sampling point in the x, y, and z directions, respectively, at time t.

6. The method for monitoring pharmaceutical logistics and transportation according to claim 1, characterized in that, Methods for obtaining matching degree include: Construct a target window based on the length of the temperature change sequence under the current state at the current moment; The target window is matched by sliding movement with a standard temperature sequence, where the sliding step size is 1. The matching degree between the temperature data of the overlapping part of the target window and the standard temperature sequence is calculated after each sliding, and the matching degree with the largest matching degree is selected as the best matching degree at the current time and in the current state.

7. The method for monitoring pharmaceutical logistics and transportation according to claim 6, characterized in that, Calculate the degree of matching between the target window and the temperature data of the overlapping portion of the standard temperature sequence after each slide, including: The DTW algorithm is used to calculate the matching degree between the temperature data of the overlapping part of the target window and the standard temperature sequence after each sliding. The matching degree is negatively correlated with the distance value.

8. The method for monitoring pharmaceutical logistics and transportation according to claim 6, characterized in that, Calculate the degree of matching between the target window and the temperature data of the overlapping portion of the standard temperature sequence after each slide, including: The matching degree is related not only to the distance between the temperature data of the overlapping part of the target window and the standard temperature sequence after sliding, but also to the consistency of the changing trend between the two sequences within the overlapping window.

9. The method for monitoring pharmaceutical logistics and transportation according to claim 2, characterized in that, The final window calculation formula includes: ； In the formula, This represents the final window corresponding to time t. This represents the minimum value of the window, where the empirical value for the minimum window is 5. This represents the maximum window size, where the empirical value for the maximum window size is 20. Indicates the first The validity of the temperature data at any given time.

10. A pharmaceutical logistics and transportation monitoring system, comprising a processor, a memory, and an alarm device, characterized in that, The memory stores a computer program, and the processor executes the computer program to implement the pharmaceutical logistics and transportation monitoring method as described in any one of claims 1-9.

Images