Automatic scheduling method and device, electronic equipment and computer readable medium

Abstract

The invention discloses an automatic scheduling method and device, electronic equipment and a computer readable medium, and relates to the technical field of internet medical treatment. The method comprises the following steps: predicting the inquiry order quantity of each time period in the future by adopting a pre-trained order quantity prediction model; according to the portrait data of each person, the scheduling priority of each person in each time period is calculated; according to the on-duty person efficiency of each time period and the inquiry list amount of each time period, the number of classes of each time period is calculated; and based on the scheduling priority of each person in each time period and the number of shifts in each time period, respectively distributing on-duty persons for each time period. According to the embodiment, the technical problems that class number evaluation is not accurate enough and the quality of an on-duty doctor cannot be guaranteed can be solved.

Description

Technical Field

[0001] The present invention relates to the field of Internet medical technology, and in particular to an automatic scheduling method, device, electronic device and computer-readable medium. Background Art

[0002] With the development of Internet technology, the application of Internet medical care is becoming more and more extensive. Doctors stationed in Internet hospitals provide patients with professional consultation services and medication guidance online through the Internet to ensure the safety of users' medication. In the online consultation scenario, by setting up doctors on duty for each department, it can effectively ensure that patients are immediately seen by doctors after placing an order. However, Internet hospitals have a large number of consultation orders every day, and a huge number of doctors need to be scheduled. If such a large number of doctors are manually scheduled, it will be a huge and long-term task, and the corresponding cost is relatively high. In addition, there are many doctors stationed on the platform, and their abilities vary greatly.

[0003] At present, there are several scheduling modes for Internet hospitals:

[0004] 1) Manually assess the number of shifts required and manually schedule doctors on duty in advance. Manual assessment of the number of shifts is not accurate enough, and scheduling is a long-term job, and manual operations have huge manpower costs.

[0005] 2) Manually assess the number of shifts required, and after setting the number of shifts, notify the doctor to grab the shift. The doctor who grabs the shift is eligible to be on duty. Manual assessment of the number of shifts is not accurate enough, and it leads to a bad doctor experience, and the quality of the doctors on duty cannot be guaranteed. Summary of the invention

[0006] In view of this, embodiments of the present invention provide an automatic scheduling method, device, electronic device and computer-readable medium to solve the technical problems of inaccurate shift assessment and inability to ensure the quality of doctors on duty.

[0007] To achieve the above object, according to one aspect of an embodiment of the present invention, an automatic scheduling method is provided, comprising:

[0008] Use the pre-trained order volume estimation model to predict the number of consultation orders in each period in the future;

[0009] According to the portrait data of each person, the scheduling priority of each person in each time period is calculated respectively;

[0010] Calculate the number of shifts in each time period according to the on-duty personnel efficiency and the number of medical consultation orders in each time period;

[0011] Based on the scheduling priorities of the various personnel in the various time periods and the number of shifts in the various time periods, personnel on duty are respectively allocated to the various time periods.

[0012] Optionally, before using a pre-trained order volume estimation model to predict the number of consultation orders in each future period, the following steps are also included:

[0013] Build training and testing datasets based on the number of consultations in each historical time period;

[0014] The training data set is used to train a time series model, thereby training a single quantity estimation model;

[0015] The test data set is used to test the accuracy of the single quantity estimation model.

[0016] Optionally, the training data set is used to train a time series model, thereby training a single quantity estimation model, including:

[0017] Dividing the training data set into a plurality of sub-training data sets of different magnitudes;

[0018] The multiple sub-training data sets are used to train each time series model respectively, and the time series model with the smallest Akaike Information Criterion AIC value is selected as the single quantity estimation model.

[0019] Optionally, according to the portrait data of each person, the scheduling priority of each person in each time period is calculated respectively, including:

[0020] For each person, a quality score of the person is calculated according to the characteristics of the person, thereby determining a score level of the person according to the quality score of the person;

[0021] For each person, the scheduling priority of the person in each time period is determined according to the person's willingness to be on duty in each time period and the scoring level of each person.

[0022] Optionally, the characteristics of the personnel include one or more of the following: professional title, hospital grade, favorable review rate, number of recent consultation orders, consultation time efficiency, and response time efficiency;

[0023] For each person, determining the scheduling priority of the person in each time period according to the person's willingness to be on duty in each time period and the scoring level of each person, includes:

[0024] The scheduling priorities of the personnel in each time period are determined according to the following rules:

[0025] The priority of those who are willing to work and have a high score level is higher than that of those who are willing to work and have a low score level, the priority of those who are willing to work and have a low score level is higher than that of those who are unwilling to work and have a high score level, and the priority of those who are unwilling to work and have a high score level is higher than that of those who are unwilling to work and have a low score level.

[0026] Optionally, the number of shifts in each time period is calculated according to the on-duty personnel efficiency in each time period and the number of medical consultation orders in each time period, including:

[0027] For each time period, the ratio of the number of medical consultation orders in the time period to the on-duty efficiency of the time period is calculated and rounded upward to an integer, thereby obtaining the number of shifts in the time period.

[0028] Optionally, based on the scheduling priorities of the various personnel in the various time periods and the number of shifts in the various time periods, the personnel on duty are respectively assigned to the various time periods, including:

[0029] For each time period, personnel on duty are allocated to the time period according to the number of shifts in the time period and in descending order of the scheduling priorities of the personnel in the time period, so that the number of personnel on duty is equal to the number of shifts.

[0030] In addition, according to another aspect of an embodiment of the present invention, there is provided an automatic scheduling device, comprising:

[0031] The order volume module is used to use the pre-trained order volume estimation model to predict the number of consultation orders in each period in the future;

[0032] A first calculation module is used to calculate the scheduling priority of each person in each time period according to the portrait data of each person;

[0033] The second calculation module is used to calculate the number of shifts in each time period according to the on-duty personnel efficiency and the number of consultation orders in each time period;

[0034] The scheduling module is used to allocate on-duty personnel to each time period based on the scheduling priority of each person in each time period and the number of shifts in each time period.

[0035] Optionally, a training module is also included, for:

[0036] Build training and testing datasets based on the number of consultations in each historical time period;

[0037] The training data set is used to train a time series model, thereby training a single quantity estimation model;

[0038] The test data set is used to test the accuracy of the single quantity estimation model.

[0039] Optionally, the training module is further used to:

[0040] Dividing the training data set into a plurality of sub-training data sets of different magnitudes;

[0041] The multiple sub-training data sets are used to train each time series model respectively, and the time series model with the smallest Akaike Information Criterion AIC value is selected as the single quantity estimation model.

[0042] Optionally, the first calculation module is further used for:

[0043] For each person, a quality score of the person is calculated according to the characteristics of the person, thereby determining a score level of the person according to the quality score of the person;

[0044] For each person, the scheduling priority of the person in each time period is determined according to the person's willingness to be on duty in each time period and the scoring level of each person.

[0045] Optionally, the characteristics of the personnel include one or more of the following: professional title, hospital grade, favorable review rate, number of recent consultation orders, consultation time efficiency, and response time efficiency;

[0046] The first calculation module is also used for:

[0047] The scheduling priorities of the personnel in each time period are determined according to the following rules:

[0048] The priority of those who are willing to work and have a high score level is higher than that of those who are willing to work and have a low score level, the priority of those who are willing to work and have a low score level is higher than that of those who are unwilling to work and have a high score level, and the priority of those who are unwilling to work and have a high score level is higher than that of those who are unwilling to work and have a low score level.

[0049] Optionally, the second calculation module is further used for:

[0050] For each time period, the ratio of the number of medical consultation orders in the time period to the on-duty efficiency of the time period is calculated and rounded upward to an integer, thereby obtaining the number of shifts in the time period.

[0051] Optionally, the scheduling module is further used for:

[0052] For each time period, personnel on duty are allocated to the time period according to the number of shifts in the time period and in descending order of the scheduling priorities of the personnel in the time period, so that the number of personnel on duty is equal to the number of shifts.

[0053] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, including:

[0054] one or more processors;

[0055] a storage device for storing one or more programs,

[0056] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of the above embodiments.

[0057] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0058] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0059] One embodiment of the above invention has the following advantages or beneficial effects: because the number of consultation orders in each time period in the future is predicted by using a pre-trained order estimation model, the scheduling priority of each person in each time period is calculated according to the portrait data of each person, and the number of shifts in each time period is calculated according to the on-duty efficiency and the number of consultation orders in each time period. Based on the scheduling priority of each person in each time period and the number of shifts in each time period, the technical means of assigning on-duty personnel to each time period are used, so the technical problems of inaccurate assessment of the number of shifts and the inability to guarantee the quality of doctors on duty in the prior art are overcome. The embodiment of the present invention reasonably allocates doctor resources through automated scheduling to ensure the quality of the medical and health industry, and makes the scheduling process of the entire Internet medical scenario intelligent and automated, thereby realizing refined management of shifts, saving scheduling manpower costs to a large extent, and improving scheduling efficiency.

[0060] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0062] Figure 1 is a flow chart of an automatic scheduling method according to an embodiment of the present invention;

[0063] Figure 2 is a flow chart of an automatic scheduling method according to a reference embodiment of the present invention;

[0064] Figure 3 is a comparison diagram of the prediction results and the test data set according to an embodiment of the present invention;

[0065] Figure 4 is a system architecture diagram for implementing the automatic scheduling method of an embodiment of the present invention;

[0066] Figure 5 is a schematic diagram of an automatic scheduling device according to an embodiment of the present invention;

[0067] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0068] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0070] It should be noted that the collection, collection, updating, analysis, processing, use, transmission, storage and other aspects of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0071] Figure 1 is a flow chart of an automatic scheduling method according to an embodiment of the present invention. As an embodiment of the present invention, Figure 1 As shown, the automatic scheduling method may include:

[0072] Step 101, using a pre-trained order volume estimation model to predict the number of consultation orders in each future time period.

[0073] First, a pre-trained order volume estimation model is used to predict the number of consultation orders at various times in the future, such as predicting the number of consultation orders per hour in the next day. Optionally, a separate order volume estimation model can be trained in advance for each department, and each department uses the pre-trained order volume estimation model to predict the number of consultation orders in each time period in the future.

[0074] Step 102, based on the portrait data of each person, respectively calculate the scheduling priority of each person in each time period.

[0075] In this step, based on the portrait data of each person (such as a doctor), the scheduling priority of each person in each future time period is calculated respectively, for example, the scheduling priority of each person in each hour in the future day is calculated respectively.

[0076] Optionally, step 102 may include: for each person, calculating the quality score of the person according to the characteristics of the person, thereby determining the score level of the person according to the quality score of the person; for each person, determining the scheduling priority of the person in each time period according to the person's willingness to work in each time period and the score level of each person. Optionally, the characteristics of the person include one or more of the following: professional title, hospital grade, praise rate, recent consultation order volume, consultation time efficiency, and response time efficiency.

[0077] First, select the doctor characteristics:

[0078] The basic characteristics of doctors include: doctor's title, hospital grade, department, prescription authority and other characteristics. Through the effectiveness analysis of the characteristics, it is found that the department, prescription authority and other characteristics have little correlation with the doctor's score. Therefore, the characteristics with low correlation are eliminated, and finally the doctor's title, hospital grade and other characteristics that can directly reflect the doctor's quality and ability are retained. In addition, in the scenario of Internet medical care, the digitization of the consultation process can obtain more characteristics, such as the patient's praise rate, the number of consultation orders handled by the doctor, and the timeliness of the consultation. Finally, 6 characteristics were determined as the basic characteristics of the scoring model: title, hospital grade, praise rate, number of consultation orders (such as the past 3 months, the past 6 months, etc.), consultation timeliness, and reply timeliness.

[0079] Then, build the priority model:

[0080] In order to control the calculation cost of the model, the doctor priority model is established using the method of active willingness + direct scoring.

[0081] Active willingness refers to collecting doctors' willingness to be on duty, which has three levels: willing, no, and unwilling.

[0082] The direct scoring method divides and scores the six characteristics into three levels, as shown in the following table.

[0083] Features\scores A (100) points B(75) points C(50) points D(25) points job title Chief Physician Associate Chief Physician Attending Physician Resident Physician Hospital Grade Top three Level 3 II Below Class II Positive rate >97.5 >95 >92.5 <92.5 Number of consultation orders (last 3 months) >750 <750 <500 <250 Time limit for consultation <20s <25s <30s >30s Reply Time <30s <1min <2min >2min

[0084] Using the table above, you can quickly calculate a score for a doctor.

[0085] For example, the six characteristic data of a doctor A are:

[0086] job title Hospital Grade Positive rate Number of consultation orders Time limit for consultation Reply Time Chief Physician Grade 3A 99 1347 26S 22S

[0087] Then the quality score of doctor A is: 100+100+100+100+50+100=550 points.

[0088] For example, the six characteristic data of a doctor B are:

[0089] job title Hospital Grade Positive rate Number of consultation orders Time limit for consultation Reply Time Attending Physician Second Class A 100 107 20S 34S

[0090] Then the quality score of doctor B is: 50+50+100+25+100+75=400 points.

[0091] Pre-configure the scoring strategy, for example, if the quality score is greater than or equal to 480 points, the scoring level is high, otherwise, the scoring level is general, for example, if the quality score is greater than or equal to 400 points, the scoring level is high, otherwise, the scoring level is general. For another example, if the quality score is greater than or equal to 450 points, the scoring level is high, if the quality score is less than 450 points and greater than or equal to 400 points, the scoring level is medium, and if the quality score is less than 400 points, the scoring level is low.

[0092] And so on, the ratings of all doctors are calculated.

[0093] Finally, for each doctor, the scheduling priority of the doctor in each time period is determined according to the doctor's willingness to work in each time period and the doctor's score level. Optionally, the scheduling priority of the personnel in each time period is determined according to the following rules: the priority of the personnel willing to work and with a high score level is higher than the priority of the personnel willing to work and with a low score level, the priority of the personnel willing to work and with a low score level is higher than the priority of the personnel unwilling to work and with a high score level, and the priority of the personnel unwilling to work and with a high score level is higher than the priority of the personnel unwilling to work and with a low score level.

[0094] The scheduling priority is determined as follows:

[0095] Willing and high score> Willing and average score> Not willing and high score> Not willing and average score> Unwilling and high score> Unwilling and average score.

[0096] Step 103, calculating the number of shifts in each time period according to the on-duty personnel efficiency in each time period and the number of medical consultation orders in each time period.

[0097] For each time period, the number of consultation orders in the time period calculated in step 101 is divided by the on-duty efficiency of the time period and the integer is rounded up to obtain the number of shifts in the time period.

[0098] It should be noted that the on-duty efficiency of each time period can be obtained by taking the average of the on-duty efficiency of each time period in the past period, or by dividing the number of medical consultation orders processed in the past day (24 time periods) by 24, as the on-duty efficiency of each time period.

[0099] Step 104 , assigning personnel on duty to each time period based on the scheduling priorities of each person in each time period and the number of shifts in each time period.

[0100] For each time period, according to the scheduling priorities of each person in the time period obtained in step 102, and in descending order of priority, the on-duty personnel are assigned to the time period, so that the number of on-duty personnel is equal to the number of shifts in the time period calculated in step 103.

[0101] According to the various embodiments described above, it can be seen that the embodiments of the present invention use a pre-trained order volume estimation model to predict the number of consultation orders in each time period in the future, calculate the scheduling priority of each person in each time period according to the portrait data of each person, calculate the number of shifts in each time period according to the on-duty efficiency and the number of consultation orders in each time period, and allocate the on-duty personnel to each time period based on the scheduling priority of each person in each time period and the number of shifts in each time period. The technical means solve the technical problems of inaccurate assessment of the number of shifts and the inability to guarantee the quality of doctors on duty in the prior art. The embodiments of the present invention reasonably allocate doctor resources through automated scheduling to ensure the quality of the medical and health industry, and make the scheduling process of the entire Internet medical scenario intelligent and automated, thereby realizing refined management of shifts, saving scheduling manpower costs to a large extent, and improving scheduling efficiency.

[0102] Figure 2 is a flow chart of an automatic scheduling method according to a reference embodiment of the present invention. As another embodiment of the present invention, Figure 2 As shown, the automatic scheduling method may include:

[0103] Step 201 : construct a training data set and a test data set according to the number of consultation orders in each historical time period.

[0104] First, determine the characteristic factors of the number of consultation orders:

[0105] The consultation data is continuous and cyclical, and the consultation volume of different departments at different times also varies greatly. For example, the demand for influenza consultations is greater in winter, so the number of consultations in the respiratory department is greater in winter, and more respiratory doctors need to be scheduled. Specifically, at the granularity of days and hours, there are characteristics such as greater demand for consultations on holidays and greater demand for consultations during non-working hours. After comprehensive analysis, the characteristic factors with greater influencing factors are:

[0106] Season, whether it is a holiday, whether it is working hours (8 - 12, 14 - 18), night (23:00 - 7:00), etc.

[0107] After determining the impact factor, classify and count the historical consultation order volume data, and mark it. The results are as follows:

[0108]

[0109]

[0110] For example, the training data set and test data set of a certain period in this department are as follows in the table:

[0111]

[0112] Among them, count represents the number of non - missing values, mean represents the average value, std represents the standard deviation, min represents the minimum value, 25% represents the first quartile (25% quantile), 50% represents the median (the second quartile, that is, 50% quantile), 75% represents the third quartile (75% quantile), and max represents the maximum value.

[0113] Use the ADF statistic to monitor the data stationarity. The test results are as follows:

[0114]

[0115]

[0116] ADF Statistic: The ADF statistic is used to test whether time - series data has a unit root. If the value of the ADF statistic is significantly less than the critical value, the null hypothesis can be rejected, that is, it is considered that the time - series data is stationary. A smaller ADF statistic indicates stronger evidence to support the stationarity of the time - series data.

[0117] p - value: The p - value is an index used to judge whether the null hypothesis holds. In the ADF test, the null hypothesis is that the time - series data has a unit root (non - stationary). If the p - value is less than the significance level (usually 0.05), the null hypothesis can be rejected, that is, it is considered that the time - series data is stationary.

[0118] Critical Values: The critical values are the thresholds used to judge the significance of the ADF statistic. If the value of the ADF statistic is less than the critical value, the null hypothesis can be rejected, that is, it is considered that the time - series data is stationary.

[0119] From the statistic results, it can be seen that ADF Statistic < Critical Values and p - value < 0.05. Therefore, it is judged that the training data is stationary.

[0120] For example, the data in the training data set are the data of the number of consultation orders in each period from September to October, and the data in the test data set are the data of the number of consultation orders in each period from October 1st to 5th. For another example, the data in the training data set are the data of the number of consultation orders in each period from July to December, and the data in the test data set are the data of the number of consultation orders in each period from January 1st to 31st.

[0121] Step 202: Use the training data set to train a time series model, thereby training a single quantity estimation model.

[0122] Due to the continuity and periodicity of the number of consultation orders, a time series algorithm is used as an algorithm for constructing a single quantity estimation model. In addition, due to the development of business and the increase in the application rate of Internet medical care, the number of consultation orders has a trend of continuous growth, so a model with trend and seasonal components is selected. For example, the SARIMA model is suitable for time series data with obvious seasonal components. The embodiment of the present invention uses the SARIMA model to construct a single quantity estimation model.

[0123] Optionally, step 202 may include: dividing the training data set into a plurality of sub-training data sets of different magnitudes; using the plurality of sub-training data sets to respectively train each time series model, and selecting the time series model with the smallest AIC value of the Akaike Information Criterion as the single quantity estimation model. The embodiment of the present invention divides the training data set into a plurality of sub-training data sets for automatic fitting training, compares the fitting effects, and selects the optimal model. Since there is noise in the data, a large amount of data noise is more complicated. The noise processing process can be simplified by using data of different magnitudes, and low-noise data can be obtained to establish a model.

[0124] Optionally, the training can be divided into three sub-training data sets: the latest month (720 items), the latest three months (2160 items), and all, and automatic fitting training can be performed. The fitting results can be compared to obtain the optimal model parameters, thereby forming a single quantity estimation model:

[0125]

[0126] AIC: Akaike Information Criterion, a commonly used model selection criterion, is used to measure the goodness of fit and complexity of a statistical model, so as to select the best model among multiple models. The smaller the AIC value, the better, indicating that the model can better fit the data.

[0127] Order, seasonal_order: The two parameters of the SARIMA model, order and seasonal_order parameters, determine the order and seasonal order of the model.

[0128] The order parameter is a tuple (p, d, q) where:

[0129] p represents the order of the autoregressive (AR) model;

[0130] d represents the order of differencing, which is used to deal with non-stationarity;

[0131] q represents the order of the moving average (MA) model.

[0132] The seasonal_order parameter is a tuple (P,D,Q,s) where:

[0133] P represents the order of the seasonal autoregressive (Seasonal AR) model;

[0134] D represents the order of seasonal difference;

[0135] Q represents the order of the seasonal moving average (Seasonal MA) model;

[0136] s represents the length of the seasonal cycle.

[0137] Among the above training models, the Akaike Information Criterion AIC value of the model trained with the most recent 720 data is the smallest, so this model parameter is used. The specific model is:

[0138] SARIMA(data, order=(0,0,1), seasonal_order=(1,0,1,24)).

[0139] Step 203: Use the test data set to test the accuracy of the order quantity estimation model.

[0140] After selecting the time series model with the smallest AIC value of Akaike Information Criterion as the order quantity estimation model, the accuracy of the order quantity estimation model is tested using the test data set. Specifically, the order quantity estimation model is used to predict the number of consultation orders in each time period in the test data set, and the prediction results are compared with the actual number of consultation orders in each time period in the test data set. When comparing, the season corresponding to each time period, whether it is a holiday, whether it is working hours, whether it is nighttime, etc. are combined to make a judgment. If the estimation effect of the order quantity estimation model is good, such as Figure 3 As shown, step 204 is executed.

[0141] Since the number of consultation orders in different departments is different in different seasons and time periods, in order to improve the prediction accuracy of the model, each department trains its own order estimation model. Each department uses the pre-trained order estimation model to predict the number of consultation orders in the department in each future time period.

[0142] Step 204, using a pre-trained order volume estimation model to predict the number of consultation orders in each future time period.

[0143] According to the training results of step 202, the Akaike Information Criterion AIC value of the model trained with the most recent 720 data is the smallest. Therefore, in step 204, the prediction step is set and SARIMA (data, order = (0,0,1), seasonal_order = (1,0,1,24)) is used to predict the number of medical consultations in each time period in the future, for example, predicting the number of medical consultations in 24 time periods on a specified date.

[0144] Step 205, calculating the scheduling priority of each person in each time period according to the portrait data of each person.

[0145] Specifically, for each person, the quality score of the person is calculated according to the characteristics of the person, and the score level of the person is determined according to the quality score of the person; for each person, the priority of the person in the scheduling of each time period is determined according to the person's willingness to work in each time period and the score level of each person. Optionally, the characteristics include one or more of the following: professional title, hospital grade, praise rate, recent consultation order volume, consultation time efficiency, and response time efficiency.

[0146] Step 206, calculating the number of shifts in each time period according to the on-duty personnel efficiency in each time period and the number of medical consultation orders in each time period.

[0147] Optionally, step 206 may include: for each time period, calculating the ratio of the number of medical consultation orders in the time period to the on-duty personnel efficiency in the time period and rounding up to an integer, thereby obtaining the number of shifts in the time period.

[0148] The order volume estimation model predicts the number of consultation orders in 24 time periods on a specified date. If a doctor in the department completes 6 consultation orders within an hour according to historical data analysis, the doctor's duty efficiency is 6. The estimated number of shifts in the department during this period is the ratio of the number of consultation orders to the doctor's duty efficiency and rounded up. For example, if the estimated number of consultation orders in a department from 9:00 to 10:00 is 243, the estimated number of shifts is ceil (243 / 6) = 41 shifts, and the corresponding department should have 41 doctors on duty from 9:00 to 10:00.

[0149] Step 207 , based on the scheduling priorities of the personnel in the various time periods and the number of shifts in the various time periods, assigning personnel on duty to the various time periods.

[0150] After determining the estimated number of shifts corresponding to each department and each time period, the doctors in each department are arranged to the shifts in the corresponding time period according to the scheduling priority of each doctor in each time period, thus completing the entire automatic scheduling process.

[0151] Optionally, step 207 may include: for each time period, allocating on-duty personnel to the time period according to the number of shifts in the time period and in descending order of the scheduling priorities of the personnel in the time period, so that the number of on-duty personnel is equal to the number of shifts.

[0152] In addition, the specific implementation content of the automatic scheduling method in a reference embodiment of the present invention has been described in detail in the automatic scheduling method described above, so the repeated content will not be described again here.

[0153] Figure 4 is a system architecture diagram for implementing the automatic scheduling method of the embodiment of the present invention. Figure 4 As shown in the figure, the automatic scheduling system includes two modules: intelligent scheduling and intelligent monitoring. These two modules jointly support the overall process of doctors and patients completing consultations. Among them, the intelligent scheduling module analyzes the consultation order data through characteristic factors of multiple dimensions, and establishes a single quantity estimation model with department and time period as the granularity, and then calculates the actual number of shifts to be scheduled based on the efficiency of the scheduled doctors on duty; secondly, doctors are assigned to corresponding shifts according to the scheduling priority based on the doctor's portrait. The intelligent monitoring module is responsible for monitoring the number of consultation orders and the idle status of doctors during the duty period, and triggers dynamic shift adjustments in real time according to the set thresholds.

[0154] The intelligent monitoring module is responsible for obtaining data such as the number of pending orders and the idle status of doctors in real time, and automatically triggering shift adjustments when the threshold is reached. Specifically, first, an evaluation indicator is constructed for the adaptive ability of shifts: the health of the shift. The health indicator of the shift is comprehensively evaluated by multiple indicators such as the backlog of consultation orders, the number of idle doctors, and the time efficiency of consultation. For example, if there is a backlog of real-time consultation orders, it means that the number of doctors on duty is less than the demand. At this time, the number of shifts should be expanded, and more doctors should be arranged to join the duty to handle the backlog of consultation orders in time. When the actual number of consultation orders is less than the estimated number of consultation orders, it means that the doctors on duty are idle, and a certain number of shifts can be released to save doctor resources. At the same time, if the time efficiency of the doctor's consultation can also reflect the doctor's duty pressure, a new doctor on duty is needed at this time.

[0155] For example, if the total score of the scheduling health is set to 100, if the consultation forms are backlogged, the corresponding score will be deducted. The more the backlog, the more the score will be deducted. Similarly, if the time efficiency of the available doctors decreases, the corresponding score will be deducted. The intelligent monitoring module performs the following operations:

[0156] 1. Collect various indicator data of shift health in real time and calculate the total score.

[0157] 2. When the schedule health is lower than the threshold, an abnormal alarm is sent to notify the operator and trigger the processing task.

[0158] 3. Processing tasks identify various indicators and execute corresponding expansion or reduction of shift times. When adding shifts or releasing shifts for doctors, doctors are notified through messages.

[0159] 4. After the execution is completed, the processing result alarm is sent to the operator.

[0160] In summary, the entire automated scheduling solution is implemented through the establishment of a single quantity estimation model, the establishment of doctor portraits, automatic scheduling by the scheduling engine, and shift adaptation.

[0161] Therefore, the automatic scheduling method provided by the embodiment of the present invention has the following beneficial effects:

[0162] (1) Automatically estimate the shift demand of each department at each time period on the current day and in the next few days without manual intervention; (2) Collect the doctors' on-duty intentions at each time period in each department; (3) Automatically schedule doctors according to priority based on factors such as doctors' on-duty intentions, abilities, and preferences; (4) Automatically identify and expand shifts when the number of consultation orders increases suddenly and notify more doctors to join the duty.

[0163] Figure 5 Schematic diagram of an automatic scheduling device according to an embodiment of the present invention. Figure 5 As shown, the automatic scheduling device 500 includes an order quantity module 501, a first calculation module 502, a second calculation module 503 and a scheduling module 504; wherein the order quantity module 501 is used to use a pre-trained order quantity estimation model to predict the number of consultation orders in each future time period; the first calculation module 502 is used to calculate the scheduling priority of each person in each time period according to the portrait data of each person; the second calculation module 503 is used to calculate the number of shifts in each time period according to the on-duty efficiency and the number of consultation orders in each time period; the scheduling module 504 is used to assign on-duty personnel to each time period based on the scheduling priority of each person in each time period and the number of shifts in each time period.

[0164] Optionally, a training module is also included, for:

[0165] Build training and testing datasets based on the number of consultations in each historical time period;

[0166] The training data set is used to train a time series model, thereby training a single quantity estimation model;

[0167] The test data set is used to test the accuracy of the single quantity estimation model.

[0168] Optionally, the training module is further used to:

[0169] Dividing the training data set into a plurality of sub-training data sets of different magnitudes;

[0170] The multiple sub-training data sets are used to train each time series model respectively, and the time series model with the smallest Akaike Information Criterion AIC value is selected as the single quantity estimation model.

[0171] Optionally, the first calculation module 502 is further configured to:

[0172] For each person, a quality score of the person is calculated according to the characteristics of the person, thereby determining a score level of the person according to the quality score of the person;

[0173] For each person, the scheduling priority of the person in each time period is determined according to the person's willingness to be on duty in each time period and the scoring level of each person.

[0174] Optionally, the characteristics of the personnel include one or more of the following: professional title, hospital grade, favorable review rate, number of recent consultation orders, consultation time efficiency, and response time efficiency;

[0175] The first calculation module is also used for:

[0176] The scheduling priorities of the personnel in each time period are determined according to the following rules:

[0177] The priority of those who are willing to work and have a high score level is higher than that of those who are willing to work and have a low score level, the priority of those who are willing to work and have a low score level is higher than that of those who are unwilling to work and have a high score level, and the priority of those who are unwilling to work and have a high score level is higher than that of those who are unwilling to work and have a low score level.

[0178] Optionally, the second calculation module 503 is further used for:

[0179] For each time period, the ratio of the number of medical consultation orders in the time period to the on-duty efficiency of the time period is calculated and rounded upward to an integer, thereby obtaining the number of shifts in the time period.

[0180] Optionally, the scheduling module 504 is further used for:

[0181] For each time period, personnel on duty are allocated to the time period according to the number of shifts in the time period and in descending order of the scheduling priorities of the personnel in the time period, so that the number of personnel on duty is equal to the number of shifts.

[0182] It should be noted that the specific implementation content of the automatic scheduling device of the present invention has been described in detail in the automatic scheduling method described above, so the repeated content will not be described again here.

[0183] Figure 6 An exemplary system architecture 600 is shown to which an automatic scheduling method or an automatic scheduling device according to an embodiment of the present invention can be applied.

[0184] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0185] Users can use terminal devices 601, 602, and 603 to interact with server 605 through network 604 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 601, 602, and 603, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0186] The terminal devices 601 , 602 , and 603 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0187] The server 605 may be a server that provides various services, such as a backend management server (for example only) that provides support for shopping websites browsed by users using the terminal devices 601, 602, and 603. The backend management server may analyze and process the received data such as the item information query request, and feed back the processing results to the terminal device.

[0188] It should be noted that the automatic scheduling method provided in the embodiment of the present invention is generally executed by the server 605, and accordingly, the automatic scheduling device is generally set in the server 605. The automatic scheduling method provided in the embodiment of the present invention can also be executed by the terminal devices 601, 602, and 603, and accordingly, the automatic scheduling device can be set in the terminal devices 601, 602, and 603.

[0189] It should be understood that Figure 6 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0190] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device suitable for implementing an embodiment of the present invention. Figure 7The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0191] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0192] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0193] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0194] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0195] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0196] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes a single quantity module, a first calculation module, a second calculation module and a scheduling module, wherein the names of these modules do not constitute a limitation on the modules themselves in some cases.

[0197] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by a device, the device implements the following method: using a pre-trained order quantity estimation model to predict the number of consultation orders in each future time period; calculating the scheduling priority of each person in each time period according to the portrait data of each person; calculating the number of shifts in each time period according to the on-duty efficiency and the number of consultation orders in each time period; assigning on-duty personnel to each time period based on the scheduling priority and the number of shifts in each time period.

[0198] As another aspect, an embodiment of the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.

[0199] According to the technical solution of the embodiment of the present invention, because the number of consultation orders in each time period in the future is predicted by using a pre-trained order volume estimation model, the scheduling priority of each person in each time period is calculated according to the portrait data of each person, and the number of shifts in each time period is calculated according to the on-duty efficiency and the number of consultation orders in each time period. Based on the scheduling priority of each person in each time period and the number of shifts in each time period, the technical means of assigning on-duty personnel to each time period are used, so the technical problems of inaccurate assessment of the number of shifts and the inability to guarantee the quality of doctors on duty in the prior art are overcome. The embodiment of the present invention reasonably allocates doctor resources through automated scheduling to ensure the quality of the medical and health industry, and makes the scheduling process of the entire Internet medical scenario intelligent and automated, thereby realizing refined management of shifts, saving scheduling manpower costs to a large extent, and improving scheduling efficiency.

[0200] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic scheduling method, It is characterized in that include: Use the pre-trained order volume estimation model to predict the number of consultation orders in each period in the future; According to the portrait data of each person, the scheduling priority of each person in each time period is calculated respectively; Calculate the number of shifts in each time period according to the on-duty personnel efficiency and the number of medical consultation orders in each time period; Based on the scheduling priorities of the various personnel in the various time periods and the number of shifts in the various time periods, personnel on duty are respectively allocated to the various time periods.

2. The method according to claim 1, It is characterized in that Before using the pre-trained order volume estimation model to predict the number of consultation orders in each future period, it also includes: Build training and testing datasets based on the number of consultations in each historical time period; The training data set is used to train a time series model, thereby training a single quantity estimation model; The test data set is used to test the accuracy of the single quantity estimation model.

3. The method according to claim 2, It is characterized in that The training data set is used to train a time series model, thereby training a single quantity estimation model, including: Dividing the training data set into a plurality of sub-training data sets of different magnitudes; The multiple sub-training data sets are used to train each time series model respectively, and the time series model with the smallest Akaike Information Criterion AIC value is selected as the single quantity estimation model.

4. The method according to claim 1, It is characterized in that According to the profile data of each person, the scheduling priority of each person in each time period is calculated respectively, including: For each person, a quality score of the person is calculated according to the characteristics of the person, thereby determining a score level of the person according to the quality score of the person; For each person, the scheduling priority of the person in each time period is determined according to the person's willingness to be on duty in each time period and the scoring level of each person.

5. The method according to claim 4, It is characterized in that The characteristics of the personnel include one or more of the following: professional title, hospital grade, favorable comment rate, number of recent consultation orders, consultation time efficiency, and response time efficiency; For each person, determining the scheduling priority of the person in each time period according to the person's willingness to be on duty in each time period and the scoring level of each person, includes: The scheduling priorities of the personnel in each time period are determined according to the following rules: The priority of those who are willing to work and have a high score level is higher than that of those who are willing to work and have a low score level, the priority of those who are willing to work and have a low score level is higher than that of those who are unwilling to work and have a high score level, and the priority of those who are unwilling to work and have a high score level is higher than that of those who are unwilling to work and have a low score level.

6. The method according to claim 1, It is characterized in that According to the on-duty personnel efficiency and the number of consultation orders in each time period, the number of shifts in each time period is calculated respectively, including: For each time period, the ratio of the number of medical consultation orders in the time period to the on-duty efficiency of the time period is calculated and rounded upward to an integer, thereby obtaining the number of shifts in the time period.

7. The method according to claim 1, It is characterized in that Based on the scheduling priorities of the personnel in the various time periods and the number of shifts in the various time periods, the personnel on duty are respectively assigned to the various time periods, including: For each time period, personnel on duty are allocated to the time period according to the number of shifts in the time period and in descending order of the scheduling priorities of the personnel in the time period, so that the number of personnel on duty is equal to the number of shifts.

8. An automatic scheduling device, It is characterized in that include: The order volume module is used to use the pre-trained order volume estimation model to predict the number of consultation orders in each period in the future; A first calculation module is used to calculate the scheduling priority of each person in each time period according to the portrait data of each person; The second calculation module is used to calculate the number of shifts in each time period according to the on-duty personnel efficiency and the number of consultation orders in each time period; The scheduling module is used to allocate on-duty personnel to each time period based on the scheduling priority of each person in each time period and the number of shifts in each time period.

9. An electronic device, It is characterized in that include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images