Scheduling decision-making method and system based on emotion analysis

Abstract

The invention relates to the technical field of scheduling management, in particular to a scheduling decision-making method and system based on emotion analysis. Comprising the following steps: establishing an emotion prediction model and a plurality of scheduling cycles; generating an emotion change curve of each employee in the current scheduling period according to the emotion prediction model; setting a first-level scheduling strategy of the current scheduling period according to all emotion change curves; and acquiring an emotion monitoring data packet according to the preset feedback time node, and judging whether to generate a correction instruction according to the emotion monitoring data packet. The emotions of all employees are periodically predicted by establishing a plurality of scheduling cycles and an emotion prediction model, and the scheduling strategy is optimized based on the predicted all emotion change curves and the work content of each time period, so that the problem that the work efficiency is reduced due to emotion interference of the employees is avoided, and the scheduling efficiency is improved. The working efficiency and the working satisfaction degree of each employee are improved, potential risks caused by sudden emotion fluctuation of the employees are avoided, and the overall operation and maintenance cost of an enterprise is reduced.

Description

Technical Field

[0001] This application relates to the technical field of shift management, and in particular to a shift scheduling decision method and system based on emotion analysis. Background Art

[0002] Emotion is a state that combines a person's feelings, thoughts, and behaviors, and plays an important role in interpersonal communication. At present, more and more enterprises monitor the emotional states of employees and timely schedule the overall work tasks based on the emotional states of employees, thereby improving the work efficiency of the enterprises.

[0003] When enterprises plan work shifts at present, they often ignore monitoring the emotional changes of employees, resulting in an increased probability of work mistakes by some employees due to emotional interference during the work process, thus interfering with the overall operation efficiency of the enterprise, and even causing safety accidents during the work process, causing serious losses to the enterprise. Summary of the Invention

[0004] The purpose of this application is: to solve the above technical problems, this application provides a shift scheduling decision method and system based on emotion analysis, aiming to improve the shift management efficiency and overall operation efficiency of enterprises.

[0005] In some embodiments of this application, by establishing multiple shift scheduling cycles and emotion prediction models to periodically predict the emotions of each employee, and based on all the predicted emotion change curves and the work content of each time period, optimize the shift scheduling strategy, avoid the problem of decreased work efficiency caused by emotional interference of employees, improve the work efficiency and job satisfaction of each employee, and reduce the work pressure of employees.

[0006] In some embodiments of this application, by setting multiple feedback time nodes, monitoring the real-time emotional states of each employee, generating corresponding emotion deviation values, and timely adjusting the shift scheduling strategy according to the emotion deviation values, avoiding potential risks brought by sudden emotional fluctuations of employees, ensuring the safe progress of the work process, and reducing the overall operation and maintenance costs of the enterprise.

[0007] In some embodiments of this application, a shift scheduling decision method based on emotion analysis is provided, including: Establishing an emotion prediction model and multiple shift scheduling cycles; Generating emotion change curves of each employee in the current shift scheduling cycle according to the emotion prediction model; Setting a primary shift scheduling strategy for the current shift scheduling cycle according to all the emotion change curves; Obtaining an emotion monitoring data packet according to a preset feedback time node, and judging whether to generate a correction instruction according to the emotion monitoring data packet; Wherein, it further includes: Establish a sequence of employees A, A = (a1, a2…a i …a n ), where, b i is the i-th employee; n is the number of employees.

[0008] In some embodiments of the present application, when establishing an emotion prediction model, it includes: Set a i as the target employee in sequence according to the sequence of employees A; Generate a training data packet for the target employee according to historical monitoring data; Establish an emotion sub-model and a confidence evaluation value f for the target employee according to the training data packet; f = µ i *j i; where, θ is the number of evaluation indicators; µ i is the influence factor of the i-th evaluation indicator; j i is the reference value of the i-th evaluation indicator generated based on the training data packet of the target employee; Set the update parameters of the emotion sub-model according to the confidence evaluation value f; Generate the emotion sub-models and confidence evaluation values of each employee in sequence; Establish a sequence of emotion sub-models P, P = (p1, p2…p i …p n ), where, pi is the emotion sub-model of the i-th employee; Establish an emotion prediction model according to the sequence of emotion sub-models P.

[0009] In some embodiments of the present application, setting the primary scheduling strategy for the current scheduling cycle includes: Set multiple time intervals within the current scheduling cycle; Establish a sequence of time intervals T, T = (t1, t2…t i …t m ), where, t i is the i-th time interval within the current scheduling cycle; m is the number of time intervals within the current scheduling cycle; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the operation evaluation values of each initial strategy and establish a sequence of operation evaluation values C, C = (c1, c2…c i …c r ), where, c i is the operation evaluation value of the i-th initial strategy; r is the number of initial strategies; Set the initial strategy corresponding to the maximum value c max in the sequence of operation evaluation values as the primary scheduling strategy.

[0010] In some embodiments of the present application, when establishing a constraint model, it includes: Preset a plurality of constraint conditions; Obtain the work plan within the current shift scheduling period, and generate the work intensity values for each time interval according to the work plan; Set ai as the target employee in sequence according to the employee sequence A; Generate the emotion prediction values of the target employee in each time interval according to the emotion fluctuation curve of the target employee; Establish the emotion prediction value sequence B of the target employee in the current shift scheduling period, B = (b1, b2... b i …b m ), where b i is the emotion prediction value of the target employee in the i-th time interval; Establish the emotion prediction value sequences of each employee in sequence; Establish a constraint model according to the fusion results of all emotion prediction value sequences, the work intensity values of each time interval, and all constraint conditions.

[0011] In some embodiments of the present application, generating the operation evaluation values of each initial strategy includes: Establish the initial strategy sequence W, W = (w1, w2... w i …w r ), where w i is the i-th initial strategy; r is the number of initial strategies; Set wi as the target initial strategy in sequence; Generate the operation evaluation value c of the target initial strategy; c = e1 * Q1 * η i * d i + e2 * Q2 * i * k i ; e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; η i is the influence factor set for the i-th time interval based on the work intensity value; d i is the efficiency evaluation value of the i-th time interval in the target initial strategy; i is the influence factor of the i-th employee set based on the confidence evaluation value; k i is the work fatigue value of the i-th employee in the target initial strategy; Generate the operation evaluation values of each initial strategy in sequence.

[0012] In some embodiments of the present application, determining whether to generate a correction instruction based on the emotion monitoring data packet includes: Based on the time interval sequence T, set the start time node of each time interval as the feedback time node; According to the first-level scheduling strategy, set the employees to be monitored at the current feedback time node; Obtain the emotion monitoring data of all employees to be monitored; Generate an emotion monitoring data packet, and generate an emotion deviation value h at the current feedback time node according to the emotion monitoring data packet; Preset the first emotion deviation value threshold H1; If h > H1, generate a first-level correction instruction at the current feedback time node.

[0013] In some embodiments of the present application, generating the emotion deviation value h at the current feedback time node includes: h = e3 * Q3 * 1i * Y1(i) * (s i - b' i ) 2 + e4 * Q4 * Y2(i) * (s i - b' i + s'i)]; e3 is a preset third weight coefficient; e4 is a preset fourth weight coefficient; Q3 is a preset third fixed coefficient; Q4 is a preset fourth fixed coefficient; n1 is the number of employees to be monitored at the current feedback time node; 1i is the influence factor of the i-th employee to be monitored; s i is the emotion evaluation value of the i-th employee to be monitored generated based on the emotion monitoring data packet; b' i is the emotion prediction value of the i-th employee to be monitored corresponding to the time interval at the current feedback time node; s' i is the emotion fluctuation threshold of the i-th employee to be monitored; Y1(i) is a selection coefficient; if (s i - b' i ) < 0, Y1(i) = 1; if (s i - b' i ) > 0, Y1(i) = 0; Y2(i) is a selection coefficient, if (s i - b' i + s' i ) > 0, Y2(i) = 0; if (s i - b' i + s' i ) < 0, Y2(i) = 1 / (s i - b' i + s' i)。

[0014] In some embodiments of the present application, a scheduling decision-making system based on sentiment analysis is provided, including: A central control unit, configured to establish a sentiment prediction model and multiple scheduling cycles; A monitoring unit, configured to obtain a sentiment monitoring data packet according to a preset feedback time node; The central control unit further includes: A first processing module, configured to establish an employee sequence A, A = (a1, a2…a i …a n ), where b i is the i-th employee; n is the number of employees; A second processing module, configured to establish a sentiment prediction model and generate a sentiment change curve of each employee in the current scheduling cycle according to the sentiment prediction model; A third processing module, configured to set a first-level scheduling strategy for the current scheduling cycle according to all the sentiment change curves; A correction module, configured to determine whether to generate a correction instruction according to the sentiment monitoring data packet.

[0015] In some embodiments of the present application, the second processing module is further configured to: Successively set a i as the target employee according to the employee sequence A; Generate a training data packet of the target employee according to historical monitoring data; Establish a sentiment sub-model and a confidence evaluation value f of the target employee according to the training data packet; f = µ i *j i; where θ is the number of evaluation indicators; µ i is the influence factor of the i-th evaluation indicator; j i is the reference value of the i-th evaluation indicator generated based on the training data packet of the target employee; Set update parameters of the sentiment sub-model according to the confidence evaluation value f; Successively generate sentiment sub-models and confidence evaluation values of each employee; Establish a sentiment sub-model sequence P, P = (p1, p2…p i …p n ), where pi is the sentiment sub-model of the i-th employee; Establish a sentiment prediction model according to the sentiment sub-model sequence P.

[0016] In some embodiments of the present application, the third processing module is further configured to: Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T = (t1, t2…t i …t m ), where t i is the i-th time interval within the current scheduling period; m is the number of time intervals within the current scheduling period; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the running evaluation values of each initial strategy and establish a running evaluation value sequence C, C = (c1, c2…c i …c r ), where c i is the running evaluation value of the i-th initial strategy; r is the number of initial strategies; Set the initial strategy corresponding to the maximum value c max in the running evaluation value sequence as the first-level scheduling strategy; Among them, generating the running evaluation values of each initial strategy includes: Establish an initial strategy sequence W, W = (w1, w2…w i …w r ), where w i is the i-th initial strategy; r is the number of initial strategies; Successively set wi as the target initial strategy; Generate the running evaluation value c of the target initial strategy; c = e1 * Q1 * η i * d i + e2 * Q2 * i * k i ; e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; η i is the influence factor for setting the i-th time interval based on the work intensity value; d i is the efficiency evaluation value of the i-th time interval in the target initial strategy; i is the influence factor for setting the i-th employee based on the confidence evaluation value; k i is the work fatigue value of the i-th employee in the target initial strategy; Successively generate the running evaluation values of each initial strategy.

[0017] Compared with the prior art, the beneficial effect of the scheduling decision method and system based on emotion analysis in the embodiments of the present application is that: By establishing multiple scheduling cycles and emotion prediction models, the emotions of each employee are predicted periodically. Based on the entire predicted emotion change curves and the work content at each time period, the scheduling strategy is optimized to avoid the problem of decreased work efficiency caused by emotional interference of employees, improve the work efficiency and job satisfaction of each employee, and reduce the work pressure of employees.

[0018] By setting multiple feedback time nodes, the real-time emotion states of each employee are monitored, and corresponding emotion deviation values are generated. According to the emotion deviation values, the scheduling strategy is adjusted in a timely manner to avoid potential risks brought by sudden emotional fluctuations of employees, ensure the safe progress of the work process, and reduce the overall operation and maintenance costs of the enterprise. Description of the Drawings

[0019] Figure 1 It is a schematic flowchart of a scheduling decision-making method based on emotion analysis in a preferred embodiment of an embodiment of the present application. Detailed Embodiments

[0020] The following will further describe in detail the specific embodiments of the present application in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0021] In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present application.

[0022] The terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "multiple" is two or more.

[0023] In the description of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0024] As shown Figure 1 in the figure, a scheduling decision-making method based on emotion analysis according to a preferred embodiment of an embodiment of the present application includes: S101: Establish an emotion prediction model and multiple scheduling cycles; S102: Generate an emotion change curve of each employee in the current scheduling cycle according to the emotion prediction model; S103: Set a primary scheduling strategy for the current scheduling cycle according to all the emotion change curves; S104: Obtain an emotion monitoring data packet according to a preset feedback time node, and determine whether to generate a correction instruction according to the emotion monitoring data packet; Among them, it further includes: Establish an employee sequence A, A = (a1, a2... a i ... a n ), where b i is the i-th employee; n is the number of employees.

[0025] Specifically, establish an employee sequence according to all the personnel who need to be scheduled Specifically, when establishing the emotion prediction model, it includes: Set a i as the target employee in sequence according to the employee sequence A; Generate a training data packet of the target employee according to the historical monitoring data; Establish an emotion sub-model and a confidence evaluation value f of the target employee according to the training data packet; f = µ i *j i; Among them, θ is the number of evaluation indicators; µ i is the influence factor of the i-th evaluation indicator; j i is the reference value of the i-th evaluation indicator generated based on the training data packet of the target employee; Set the update parameter of the emotion sub-model according to the confidence evaluation value f; Generate the emotion sub-models and confidence evaluation values of each employee in sequence; Establish an emotion sub-model sequence P, P = (p1, p2... p i ... p n ), where pi is the emotion sub-model of the i-th employee; Establish an emotion prediction model according to the emotion sub-model sequence P.

[0026] Specifically, training data packets for each employee are generated based on historical data. The data in the training data packet is the continuous monitoring data of the corresponding employee. By analyzing the data in the training data packet, it is determined whether there are periodic change characteristics in the employee's emotions, thereby generating corresponding emotion sub-models.

[0027] Specifically, the evaluation indicators include, but are not limited to, the significance of the periodic change characteristics of the employee's emotions, the evaluation value of the fluctuation of the employee's work efficiency, the correlation degree between the emotion change and the work efficiency, etc.

[0028] Specifically, the higher the confidence evaluation value, the more significant the periodic characteristics of the current employee's emotion fluctuation, the simpler the emotion prediction, and the smaller the degree of interference of the emotion fluctuation on the work efficiency of the employee.

[0029] Specifically, the lower the confidence evaluation value, the faster the update and optimization frequency of the corresponding emotion sub-model, thereby improving the prediction accuracy of the employee's emotion corresponding to each emotion sub-model.

[0030] In the preferred embodiment of the present application, the first-level scheduling strategy for the current scheduling period is set, including: Set multiple time intervals within the current scheduling period; Establish a time interval sequence T, T = (t1, t2... t i ... t m ), where t i is the i-th time interval within the current scheduling period; m is the number of time intervals within the current scheduling period; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the operation evaluation values of each initial strategy, and establish an operation evaluation value sequence C, C = (c1, c2... c i ... c r ), where c i is the operation evaluation value of the i-th initial strategy; r is the number of initial strategies; Set the initial strategy corresponding to the maximum value c max in the operation evaluation value sequence as the first-level scheduling strategy.

[0031] Specifically, a single time interval is a scheduling period. The duration of the scheduling period and the duration of a single time interval can be set according to historical parameters.

[0032] Specifically, the larger the operation evaluation value, the higher the overall feasibility of the scheduling plan corresponding to the current initial strategy, the higher the average work efficiency of the employees, and the smaller the possibility of work mistakes.

[0033] Specifically, when establishing the constraint model, it includes: Preset multiple constraint conditions; Obtain the work plan within the current scheduling cycle, and generate the work intensity values for each time interval according to the work plan; Set ai as the target employee in sequence according to the employee sequence A; Generate the emotion prediction values of the target employee in each time interval according to the emotion fluctuation curve of the target employee; Establish the emotion prediction value sequence B of the target employee in the current scheduling cycle, B = (b1, b2…b i …b m ), where b i is the emotion prediction value of the target employee in the i-th time interval; Establish the emotion prediction value sequences of each employee in sequence; Establish a constraint model based on the fusion results of all emotion prediction value sequences, the work intensity values of each time interval, and all constraint conditions.

[0034] Specifically, the preset constraint conditions include, but are not limited to, the number of on-duty employees in a single time interval, the minimum emotion prediction value threshold that employees need to meet under each work intensity value, the upper and lower limits of the on-duty duration of employees in a single scheduling cycle, and other condition parameters that affect scheduling.

[0035] Specifically, by solving the constraint model, generate multiple initial strategies, where a single initial strategy includes the personnel required to complete on-duty corresponding to each time interval.

[0036] Specifically, generate the operation evaluation values of each initial strategy, including: Establish the initial strategy sequence W, W=(w1, w2…w i …w r ), where w i is the i-th initial strategy; r is the number of initial strategies; Set wi as the target initial strategy in sequence; Generate the operation evaluation value c of the target initial strategy; c = e1 * Q1 * η i * d i + e2 * Q2 * i * k i ; e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; η i is the influence factor set for the i-th time interval based on the work intensity value; d i is the efficiency evaluation value of the i-th time interval in the target initial strategy; iis the influence factor of the i-th employee set based on the confidence evaluation value; k i is the work fatigue value of the i-th employee in the target initial strategy; Generate the operation evaluation values of each initial strategy in sequence.

[0037] Specifically, normalize all the parameters in the model by presetting the first fixed coefficient and the second fixed coefficient, so that each parameter in the model is within the same value range.

[0038] Specifically, the efficiency evaluation value of the time interval is analyzed according to the work intensity value and the predicted emotion value of the employee in the current time interval, combined with the confidence evaluation value corresponding to the employee. The larger the efficiency evaluation value, the higher the work efficiency of the corresponding time interval under the current initial strategy, and the higher the possibility of work mistakes.

[0039] Specifically, the work fatigue value is comprehensively analyzed according to the total on-duty duration and continuous on-duty duration of the employee in the current scheduling cycle, the dispersion degree of the on-duty time period, the work intensity corresponding to the on-duty, and the on-duty status during the emotion peak period and emotion trough period. The larger the work fatigue value, the greater the work pressure of the current employee.

[0040] It can be understood that in the above embodiments, by establishing multiple scheduling cycles and emotion prediction models to periodically predict the emotions of each employee, and based on all the predicted emotion change curves and the work content of each time period, optimize the scheduling strategy, avoid the problem of the decline in work efficiency caused by employees' emotional interference, improve the work efficiency and job satisfaction of each employee, and reduce the work pressure of employees.

[0041] In the preferred embodiment of the present application, determining whether to generate a correction instruction according to the emotion monitoring data packet includes: Set the start time node of each time interval as the feedback time node based on the time interval sequence T; Set the employees to be monitored at the current feedback time node according to the primary scheduling strategy; Obtain the emotion monitoring data of all employees to be monitored; Generate an emotion monitoring data packet, and generate the emotion deviation value h at the current feedback time node according to the emotion monitoring data packet; Preset the first emotion deviation value threshold H1; If h > H1, generate a primary correction instruction at the current feedback time node.

[0042] Specifically, the employees to be monitored are the employees who need to be on duty in the time interval corresponding to the current feedback time node.

[0043] Specifically, the first-level correction instruction refers to the drastic fluctuation of the emotions of the on-duty employees in the current time interval, which seriously affects their own work efficiency and requires timely optimization and adjustment of the subsequent shift scheduling strategy.

[0044] Specifically, generating the emotion deviation value h at the current feedback time node includes: h = e3 * Q3 * 1i * Y1(i) * (s i - b' i ) 2 + e4 * Q4 * Y2(i) * (s i - b' i + s'i)]; e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; n1 is the number of personnel to be monitored at the current feedback time node; 1i is the influence factor of the i-th employee to be monitored; s i is the emotion evaluation value of the i-th employee to be monitored generated based on the emotion monitoring data packet; b' i is the emotion prediction value of the i-th employee to be monitored corresponding to the time interval at the current feedback time node; s' i is the emotion fluctuation threshold of the i-th employee to be monitored; Y1(i) is the selection coefficient; if (s i - b' i ) < 0, Y1(i) = 1; if (s i - b' i ) > 0, Y1(i) = 0; Y2(i) is the selection coefficient, if (s i - b' i + s' i ) > 0, Y2(i) = 0; if (s i - b' i + s' i ) < 0, Y2(i) = 1 / (s i - b' i + s' i ).

[0045] Specifically, all parameters in the model are normalized by presetting the third fixed coefficient, the fourth fixed coefficient, and the fifth fixed coefficient, so that each parameter in the model is within the same value range.

[0046] It can be understood that in the above embodiments, by setting multiple feedback time nodes, the real-time emotional states of each employee are monitored, and corresponding emotional deviation values are generated. According to the emotional deviation values, the scheduling strategy is adjusted in a timely manner to avoid potential risks brought by sudden emotional fluctuations of employees, ensure the safe progress of the work process, and reduce the overall operation and maintenance costs of the enterprise.

[0047] Based on another preferred embodiment of a scheduling decision-making method based on emotion analysis in any of the above preferred embodiments, a scheduling decision-making system based on emotion analysis is provided in this preferred embodiment, including: A central control unit, configured to establish an emotion prediction model and multiple scheduling cycles; A monitoring unit, configured to obtain emotion monitoring data packets according to preset feedback time nodes; The central control unit further includes: A first processing module, configured to establish a sequence of employees A, A = (a1, a2…a i …a n ), where b i is the i-th employee; n is the number of employees; A second processing module, configured to establish an emotion prediction model and generate an emotion change curve of each employee in the current scheduling cycle according to the emotion prediction model; A third processing module, configured to set a primary scheduling strategy for the current scheduling cycle according to all the emotion change curves; A correction module, configured to determine whether to generate a correction instruction according to the emotion monitoring data packet.

[0048] Specifically, the second processing module is further configured to: Set a i as the target employee in sequence according to the sequence of employees A; Generate a training data packet of the target employee according to historical monitoring data; Establish an emotion sub-model and a confidence evaluation value f of the target employee according to the training data packet; f = µ i *j i; where θ is the number of evaluation indicators; µ i is the influence factor of the i-th evaluation indicator; j i is the reference value of the i-th evaluation indicator generated based on the training data packet of the target employee; Set the update parameters of the emotion sub-model according to the confidence evaluation value f; Generate the emotion sub-models and confidence evaluation values of each employee in sequence; Establish a sequence of emotion sub-models P, P = (p1, p2…p i …p n), where pi is the emotion sub - model of the i - th employee; Establish an emotion prediction model according to the emotion sub - model sequence P.

[0049] In the preferred embodiment of the present application, the third processing module is further configured to: Set multiple time intervals within the current scheduling cycle; Establish a time - interval sequence T, T=(t1, t2…t i …t m ), where t i is the i - th time interval within the current scheduling cycle; m is the number of time intervals within the current scheduling cycle; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the operation evaluation values of each initial strategy and establish an operation evaluation value sequence C, C=(c1, c2…c i …c r ), where c i is the operation evaluation value of the i - th initial strategy; r is the number of initial strategies; Set the initial strategy corresponding to the maximum value c max in the operation evaluation value sequence as the first - level scheduling strategy; Among them, generating the operation evaluation values of each initial strategy includes: Establish an initial - strategy sequence W, W=(w1, w2…w i …w r ), where w i is the i - th initial strategy; r is the number of initial strategies; Set wi as the target initial strategy in turn; Generate the operation evaluation value c of the target initial strategy; c = e1*Q1* η i *d i +e2*Q2* i *k i ; e1 is a preset first weight coefficient; e2 is a preset second weight coefficient; Q1 is a preset first fixed coefficient; Q2 is a preset second fixed coefficient; η i is the influence factor for setting the i - th time interval based on the work intensity value; d i is the efficiency evaluation value of the i - th time interval in the target initial strategy; i is the influence factor for setting the i - th employee based on the confidence evaluation value; k i is the work fatigue value of the i - th employee in the target initial strategy; Generate the operation evaluation values of each initial strategy in sequence.

[0050] According to the first concept of the present application, the emotions of each employee are periodically predicted by establishing multiple scheduling cycles and emotion prediction models, and based on all the predicted emotion change curves and the work content of each period, the scheduling strategy is optimized to avoid the problem of decreased work efficiency caused by employee emotional interference, improve the work efficiency and job satisfaction of each employee, and reduce the work pressure of employees.

[0051] According to the second concept of the present application, by setting multiple feedback time nodes, the real-time emotion state of each employee is monitored, and the corresponding emotion deviation value is generated. According to the emotion deviation value, the scheduling strategy is adjusted in a timely manner to avoid potential risks brought by sudden emotional fluctuations of employees, ensure the safe progress of the work process, and reduce the overall operation and maintenance cost of the enterprise.

[0052] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art in the technical field of the present application, without departing from the technical principle of the present application, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application.

Claims

1. A scheduling decision method based on sentiment analysis, characterized in that: include: Build sentiment prediction models and multiple scheduling cycles; Generate the emotion change curve of each employee in the current shift cycle based on the emotion prediction model; Set the first-level scheduling strategy for the current scheduling cycle based on all sentiment change curves; Obtaining an emotion monitoring data packet according to a preset feedback time node, and determining whether to generate a correction instruction according to the emotion monitoring data packet; Among them, it also includes: Create an employee sequence A, A=(a1,a2…a i …a n ), where b i is the i-th employee; n is the number of employees.

2. The scheduling decision method based on sentiment analysis as claimed in claim 1, characterized in that: When building a sentiment prediction model, include: According to the employee number sequence A, set a i For target employees; Generate training data packages for target employees based on historical monitoring data; Establish the target employee's emotion sub-model and confidence evaluation value f according to the training data package; f= µ i *j i; Among them, θ is the number of evaluation indicators; µ i is the influencing factor of the i-th evaluation index; j i is the reference value of the i-th evaluation indicator generated based on the training data package of the target employee; Set the update parameters of the emotion sub-model according to the confidence evaluation value f; Generate the emotion sub-model and confidence evaluation value of each employee in turn; Establish the emotion sub-model series P, P=(p1,p2…p i …p n ), where pi is the emotion sub-model of the i-th employee; The emotion prediction model is established according to the emotion sub-model sequence P.

3. The scheduling decision method based on sentiment analysis as claimed in claim 2, characterized in that: Set the first-level scheduling strategy for the current scheduling cycle, including: Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T=(t1, t2…t i …t m ), where t i is the i-th time interval in the current scheduling cycle; m is the number of time intervals in the current scheduling cycle; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the operation evaluation value of each initial strategy and establish the operation evaluation value sequence C, C = (c1, c2…c i …c r ), where c i is the running evaluation value of the i-th initial strategy; r is the number of initial strategies; Set the maximum value c in the running evaluation value series max The corresponding initial strategy is the first-level scheduling strategy.

4. The scheduling decision method based on sentiment analysis as claimed in claim 3, characterized in that: When building a constraint model, include: Preset multiple constraints; Get the work plan in the current shift schedule, and generate work intensity values ​​for each time interval based on the work plan; According to the employee sequence A, set ai as the target employee in sequence; Generate the target employee's emotion prediction value in each time interval according to the target employee's emotion fluctuation curve; Establish a target employee’s emotion prediction value series B in the current shift cycle, B=(b1,b2…b i …b m ), where b i is the predicted emotion value of the target employee in the i-th time interval; Establish a series of predicted emotion values ​​for each employee in turn; A constraint model is established based on the fusion results of all emotion prediction value series, work intensity values ​​in each time interval and all constraint conditions.

5. The scheduling decision method based on sentiment analysis as claimed in claim 3, characterized in that: Generate operational evaluation values ​​for each initial strategy, including: Establish the initial strategy sequence W, W=(w1, w2…w i …w r ), where w i is the i-th initial strategy; r is the number of initial strategies; Set wi as the target initial strategy in turn; Generate an operational evaluation value c of the target initial strategy; c=e1*Q1*[ η i *d i ]+e2*Q2*[ i *k i ]; e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; η i To set the impact factor of the i-th time interval based on the work intensity value; d i is the efficiency evaluation value of the i-th time interval in the target initial strategy; i is the influence factor of the ith employee set based on the confidence evaluation value; k i is the work fatigue value of the i-th employee in the target initial strategy; Generate the running evaluation value of each initial strategy in turn.

6. The scheduling decision method based on sentiment analysis as claimed in claim 5, characterized in that: Determine whether to generate a correction instruction based on the emotion monitoring data packet, including: Based on the time interval sequence T, the start time node of each time interval is set as the feedback time node; Set the employees to be monitored at the current feedback time node according to the first-level scheduling strategy; Obtain the emotion monitoring data of all employees to be monitored; Generate an emotion monitoring data packet, and generate an emotion deviation value h at the current feedback time node based on the emotion monitoring data packet; Preset the first emotion deviation value threshold H1; If h>H1, the current feedback time node generates a first-level correction instruction.

7. The scheduling decision method based on sentiment analysis according to claim 6, characterized in that: Generate the emotion deviation value h of the current feedback time node, including: h=e3*Q3*[ 1i *Y1(i)*(s i -b' i ) 2 ]+e4*Q4*[ Y2(i)*(s i -b' i +s'i)]; e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; n1 is the number of personnel to be monitored at the current feedback time node; 1i is the influencing factor of the i-th employee to be monitored; s i Generates the emotion evaluation value of the i-th employee to be monitored based on the emotion monitoring data packet; b' i is the predicted emotion value of the i-th monitored employee in the time interval corresponding to the current feedback time node; s' i is the emotional fluctuation threshold of the i-th employee to be monitored; Y1(i) is the selection coefficient; if (s i -b' i )<0, Y1(i)=1; if (s i -b' i )>0, Y1(i)=0; Y2(i) is the selection coefficient, if (s i -b' i +s' i )>0,Y2(i)=0; if (s i -b' i +s' i )<0,Y2(i)=1 / (s i -b' i +s' i ).

8. A scheduling decision system based on sentiment analysis, using the scheduling decision method based on sentiment analysis described in any one of claims 1 to 7, characterized in that: include: Central control unit, used to build emotion prediction models and multiple scheduling cycles; A monitoring unit, used to obtain an emotion monitoring data packet according to a preset feedback time node; The central control unit also includes: The first processing module is used to create an employee number sequence A, A=(a1, a2…a i …a n ), where b i is the i-th employee; n is the number of employees; The second processing module is used to establish an emotion prediction model and generate an emotion change curve of each employee in the current scheduling cycle according to the emotion prediction model; The third processing module is used to set the first-level scheduling strategy of the current scheduling cycle according to all emotion change curves; The correction module is used to determine whether to generate a correction instruction based on the emotion monitoring data packet.

9. The scheduling decision system based on sentiment analysis as claimed in claim 8, characterized in that: The second processing module is further used for: According to the employee number sequence A, set a i For target employees; Generate training data packages for target employees based on historical monitoring data; Establish the target employee's emotion sub-model and confidence evaluation value f according to the training data package; f= µ i *j i; Among them, θ is the number of evaluation indicators; µ i is the influencing factor of the i-th evaluation index; j i is the reference value of the i-th evaluation indicator generated based on the training data package of the target employee; Set the update parameters of the emotion sub-model according to the confidence evaluation value f; Generate the emotion sub-model and confidence evaluation value of each employee in turn; Establish the emotion sub-model series P, P=(p1,p2…p i …p n ), where pi is the emotion sub-model of the i-th employee; The emotion prediction model is established according to the emotion sub-model sequence P.

10. The scheduling decision system based on sentiment analysis according to claim 9, characterized in that: The third processing module is also used for: Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T=(t1, t2…t i …t m ), where t i is the i-th time interval in the current scheduling cycle; m is the number of time intervals in the current scheduling cycle; 22 Establish a constraint model and generate multiple initial strategies according to the constraint model; Generate the operation evaluation value of each initial strategy and establish the operation evaluation value sequence C, C = (c1, c2…c i …c r ), where c i is the running evaluation value of the i-th initial strategy; r is the number of initial strategies; Set the maximum value c in the running evaluation value series max The corresponding initial strategy is the first-level scheduling strategy; Among them, the operation evaluation values ​​of each initial strategy are generated, including: Establish the initial strategy sequence W, W=(w1, w2…w i …w r ), where w i is the i-th initial strategy; r is the number of initial strategies; Set wi as the target initial strategy in turn; Generate an operational evaluation value c of the target initial strategy; c=e1*Q1*[ η i *d i ]+e2*Q2*[ i *k i ]; e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; η i To set the impact factor of the i-th time interval based on the work intensity value; d i is the efficiency evaluation value of the i-th time interval in the target initial strategy; i is the influence factor of the ith employee set based on the confidence evaluation value; k i is the work fatigue value of the i-th employee in the target initial strategy; Generate the running evaluation value of each initial strategy in turn.

Images