Production management and control method and system based on business intelligence agent

Abstract

The application provides a production management and control method and system based on business intelligence agents, and relates to the technical field of data processing. The order intelligence agent monitors business communication records in real time. If it is detected that a customer has a tendency to place an order and the time is urgent, a virtual order containing the expected production time and the expected delivery time point is generated. The virtual order parameters are pushed to the resource intelligence agent. The resource intelligence agent combines the current idle time period to filter out candidate order intelligence agents that meet the time requirements. According to the evaluation score of the candidate order intelligence agent, the order intelligence agent selects the reservation request and generates a reservation time period. When the customer places an order, the resource intelligence agent determines the production time period according to the real delivery time point, the real production time and the reservation time period, and updates the evaluation score of the order intelligence agent according to the time deviation and the reservation realization rate. The closed-loop optimization of demand prediction and resource scheduling is realized by the cooperative working mode of the two intelligence agents, which significantly reduces the risk of loss of urgent orders.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a production management and control method and system relying on a business intelligence agent. BACKGROUND

[0002] The current mainstream production management and control mode is that an enterprise starts a production plan scheduling process only after formally receiving a customer order: first, the product specifications, quantity and delivery period of the order are confirmed, then the corresponding production resources (such as designated production lines, production equipment allocation, and raw material inventory accounting) are matched, and finally a production task list is generated for sequential execution. This mode can realize the orderly use of production resources and guarantee the delivery efficiency of regular orders in the scenario of stable order demand and sufficient delivery period.

[0003] However, in the current market with frequent demand fluctuations, the probability of emergency orders significantly increases. When an emergency order occurs, the existing production system is often passive, for example, the production resources are already at full capacity or are sequentially occupied, the production line may be continuously operating to process the previous regular orders, the core production equipment has fixed process parameters and is in continuous operation, and even the supporting material supply and labor scheduling are also based on the original production plan.

[0004] If an emergency order is inserted, the enterprise faces a dilemma: on the one hand, if the current production task is forcibly interrupted, the device parameters are adjusted, and the raw materials are replaced, it will cause the production progress of the previous order to be delayed, increase the material loss and equipment debugging cost, and may also cause dissatisfaction of the regular order customers; on the other hand, if the emergency order is not prioritized, the existing production task is completed according to the original plan, and then the urgent production is started, which often exceeds the delivery deadline of the emergency order, the customer may choose other suppliers due to the inability to wait, resulting in an increase in the loss rate of emergency orders, directly affecting the business efficiency of the enterprise. SUMMARY

[0005] The problem to be solved by the present application is that the existing production management and control method has weak processing capacity for emergency orders, which easily leads to an increase in the loss rate of emergency orders and affects the business efficiency of the enterprise.

[0006] To solve the above problems, in a first aspect, the present application provides a production management and control system relying on a business intelligence agent, comprising a plurality of order agents and a plurality of resource agents;

[0007] The order agent is used to obtain the communication record information between the front-end salesperson and the customer, generate an estimated delivery time point and an estimated production duration according to the customer's order intention and time urgency, send the estimated delivery time point and the estimated production duration to the plurality of resource agents, and send a reservation request to the resource agent;

[0008] The resource agent is configured to generate a current idle time period according to an existing production scheduling plan of the resource agent, and to screen a plurality of order agents according to the current idle time period and a predicted delivery time point and a predicted production duration sent by the plurality of order agents, to obtain a plurality of selected order agents.

[0009] The selected order agent is configured to receive an order instruction from a front-end business agent, to generate a real order, a real delivery time point and a real production duration according to communication record information, and to send the real order, the real delivery time point and the real production duration to a corresponding resource agent.

[0010] The resource agent is further configured to determine a production time period according to the real delivery time point, the real production duration and the reserved time period, and to update an evaluation score of the selected order agent according to a time deviation between the production time period and the reserved time period and a reservation redemption rate.

[0011] Optionally, the communication record information between the front-end business agent and the customer is obtained, and a predicted delivery time point and a predicted production duration are generated according to an order tendency of the customer and a time urgency degree, and the method comprises the following steps.

[0012] A preset order word in the communication record information is extracted, and a word frequency is determined, wherein the preset order word is divided into a plurality of tendency grades, and different tendency grades correspond to different frequency thresholds.

[0013] A comprehensive frequency is determined according to the tendency grade of the preset order word, the corresponding word frequency and the corresponding frequency threshold.

[0014] If the word frequency of any one of the preset order words is greater than the corresponding frequency threshold or the comprehensive frequency is greater than a comprehensive threshold, it is determined that the customer will place an order.

[0015] The delivery time, the current time, the product type and the quantity expected by the customer in the communication record information are extracted, and a virtual order, a predicted delivery time point, a predicted production duration and an available duration are determined.

[0016] The available duration is divided by the predicted production duration to determine a time margin.

[0017] If the predicted production duration is less than or equal to a first preset duration and the time margin is less than a first margin threshold, it is determined that the order of the customer is urgent.

[0018] If the predicted production duration is greater than the first preset duration and the time margin is less than a second margin threshold, it is determined that the order of the customer is urgent, wherein the first margin threshold is greater than the second margin threshold.

[0019] According to the predicted delivery time point and the predicted production duration, the latest production time period is determined as the production time period of the imaginary order.

[0020] Optionally, the integrated frequency is

[0021]

[0022] in, To pre-determine the weight of the i-th tendency level of the pre-defined order words, Let the word frequency be the preset order word for the i-th preference level. Let n be the frequency threshold of the preset order words for the i-th preference level, and n be the total number of preference levels of the preset order words.

[0023] Optionally, the step of filtering multiple order agents based on the current idle time period and the estimated delivery time and estimated production duration sent by multiple order agents to obtain candidate order agents includes:

[0024] Before the expected delivery time, analyze whether there are time periods in the current idle time period that are longer than or equal to the expected production time; or whether there are time periods in the current idle time period that are longer than or equal to the expected production time and have a greater overlap with the production time of the hypothetical order than the preset overlap.

[0025] If so, the order agent corresponding to the hypothetical order production period will be selected as the candidate order agent;

[0026] If not, reservation requests from the order agent corresponding to the hypothetical order production period will not be accepted.

[0027] Optionally, the step of selecting the order agent and the reservation time period based on the evaluation scores of multiple resource agents on multiple candidate order agents includes:

[0028] Based on the evaluation time points of multiple resource agents' evaluation scores for a candidate order agent, determine the latest evaluation score and the latest evaluation time point;

[0029] Based on other evaluation scores and corresponding evaluation time points, as well as the latest evaluation score and latest evaluation time point, determine the comprehensive evaluation score of the candidate order agent;

[0030] Based on the comprehensive evaluation score and the preset score threshold, multiple comprehensive evaluation scores are filtered, and the filtered comprehensive evaluation scores are sorted from largest to smallest to obtain the evaluation score sequence.

[0031] Select a preset number of order agents corresponding to the comprehensive evaluation scores from the evaluation score sequence in order, and determine whether the time period to be occupied by the selected order agents will conflict based on the production time and delivery time of the selected order agents.

[0032] If not, the virtual order of the selected order agent is added to the production plan of the resource agent, the reservation time period of the order agent is determined, and the current idle time period is updated;

[0033] If yes, the order agent with a larger comprehensive evaluation score is selected, the order agent with a smaller comprehensive evaluation score is discarded, the virtual order of the selected order agent is added to the production plan of the resource agent, the reservation time period of the order agent is determined, and the current idle time period is updated;

[0034] According to the updated current idle time period, the order agents to be selected are screened to obtain an updated order agent to be selected and an updated evaluation score sequence, until a preset number of order agents are selected or the order agent to be selected in the updated order agent to be selected is zero.

[0035] Optionally, the comprehensive evaluation score is

[0036]

[0037]

[0038]

[0039] wherein, is the latest evaluation score, is the weight of the latest evaluation score, K is the total number of other evaluation scores, is the time difference of the kth evaluation score in the other evaluation scores, is the total time difference, is the kth evaluation score in the other evaluation scores, is the evaluation time point of the kth evaluation score in the other evaluation scores, is the latest evaluation time point.

[0040] Optionally, the production time period is determined according to the real delivery time point, the real production duration and the reservation time period, and the production time period comprises:

[0041] determining whether the real delivery time point is within the reservation time period and whether the real production duration is less than or equal to the reservation time period;

[0042] If yes, the real order is inserted into the reservation time period, the production plan is updated, and it is determined whether the reservation is realized;

[0043] If no, the exceeding duration of the real delivery time point and the real production duration exceeding the reservation time period is analyzed, and it is determined whether the exceeding duration is greater than a preset duration threshold;

[0044] If the exceeding duration is less than or equal to the preset duration threshold, the production plan is adjusted according to the delivery time point of the production order in the production plan and the current idle time period, a production time period is obtained, and it is determined that the appointment is redeemed;

[0045] If the exceeding duration is greater than the preset duration threshold, the corresponding real order is rejected, and it is determined that the appointment is not redeemed.

[0046] Optionally, the updating of the evaluation score of the selected order agent according to the time deviation between the production time period and the appointment time period and the appointment redemption rate comprises:

[0047] determining an appointment deviation rate according to the time deviation and the length of the appointment time period;

[0048] determining the appointment redemption rate according to the total number of appointments within the preset duration and the number of appointment redemptions;

[0049] determining the appointment redemption rate according to the total number of appointments within the preset duration and the number of appointment redemptions;

[0050] Optionally, the updated evaluation score is

[0051]

[0052] wherein, the updated evaluation score is, the time deviation is, the length of the appointment time period is, the total number of appointments is, the number of appointment redemptions is, the existing evaluation score is.

[0053] In a second aspect, the present application further provides a production management method based on a business agent, comprising:

[0054] obtaining communication record information of a front-end business agent and a customer, generating a predicted delivery time point and a predicted production duration according to the customer's order placing tendency and time urgency;

[0055] sending the predicted delivery time point and the predicted production duration to a plurality of resource agents, and sending an appointment request to the resource agents;

[0056] generating a current idle time period according to an existing production scheduling plan of the resource agent;

[0057] screening a plurality of order agents according to the current idle time period and the predicted delivery time point and the predicted production duration sent by the plurality of order agents, to obtain a selected order agent;

[0058] According to the evaluation score of the plurality of candidate order agents by the plurality of resource agents, the selected order agent and the reservation time period are determined;

[0059] The order instruction of the reception front-end clerk is received, the real order, the real delivery time point and the real production duration are generated according to the communication record information, and are sent to the corresponding resource agent;

[0060] According to the real delivery time point, the real production duration and the reservation time period, the production time period is determined;

[0061] According to the time deviation between the production time period and the reservation time period and the reservation realization rate, the evaluation score of the selected order agent is updated.

[0062] The application provides a production management method and system based on business intelligence agents. Compared with the prior art, the application has the following beneficial effects:

[0063] The order agent monitors the business communication record in real time, and when it is detected that the customer has an order placing tendency and the time is urgent, a virtual order containing the expected production duration and the expected delivery time point is automatically generated. These virtual order parameters are pushed to the related resource agent for feasibility verification. The resource agent combines the current production load and the current idle time period to select the candidate order agent that meets the time requirement. According to the evaluation score of the candidate order agent, the resource agent selects the order agent that can accept the reservation request, and generates a reservation time period; when the customer formally places an order, the resource agent determines the production time period according to the real delivery time point, the real production duration and the reservation time period, that is, the real order is flexibly bound with the reserved reservation time period, realizing the smooth transition of the production plan, and further updating the evaluation score of the selected order agent according to the time deviation between the production time period and the reservation time period and the reservation realization rate, improving the accurate screening of the reservation request of the order agent next time. The application can significantly reduce the risk of loss of emergency orders. The virtual order mechanism makes the production resources enter the standby state in advance, and when the real order arrives, the reserved production capacity can be directly enabled. The double-agent cooperative working mode realizes the closed-loop optimization of demand prediction and resource scheduling, ensures the timely delivery of regular orders, and reserves flexible processing space for sudden demand. The application solves the problem of lagging behind in processing emergency orders caused by passive response of the traditional production management system, and has the advantages of actively predicting potential orders, planning production resources in advance and improving the efficiency of processing emergency orders. BRIEF DESCRIPTION OF DRAWINGS

[0064] In order to make the technical solutions of the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0065] Figure 1 A structural schematic diagram of a production management and control system relying on a business intelligent agent is provided for the embodiments of the present application.

[0066] Figure 2 A flowchart of a production management and control method relying on a business intelligent agent is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0067] In order to make the technical solutions of the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0068] The emergency order cannot be given priority, which not only causes the customer satisfaction to decrease, but also causes the enterprise to miss business opportunities, directly affecting the operating efficiency. The traditional system lacks the ability to predict potential orders, and cannot plan production resources in advance before the customer formally places an order, so that the production scheduling is always in a passive state. In addition, the prior art cannot effectively evaluate the urgency and credibility of the order, and it is difficult to make reasonable resource allocation decisions among multiple potential orders, further exacerbating the problem of not timely processing of emergency orders.

[0069] In order to solve the above problems, the inventor found that the root cause of production resource conflict is the disconnection between demand prediction and resource scheduling. By analyzing the communication process between the salesperson and the customer, the customer's order placing tendency can be found through specific language characteristics, and the time urgency can be deduced in combination with the possible product order information. Based on this, a double-intelligent-agent collaboration mechanism is proposed, which places demand prediction and resource matching in the order generation stage, and realizes dynamic pre-adjustment of production planning.

[0070] In order to better understand the above technical solutions, the above technical solutions will be described in detail in combination with the accompanying drawings of the specification and specific embodiments.

[0071] As shown in Figure 1 The production management and control system relying on a business intelligent agent provided by the embodiments of the present application includes a plurality of order intelligent agents and a plurality of resource intelligent agents.

[0072] The order agent 10 is configured to acquire communication record information between the front-end salesperson and the customer, generate an estimated delivery time point and an estimated production duration according to the customer's order tendency and time urgency, and send the estimated delivery time point and the estimated production duration to the resource agents and send a reservation request to the resource agents.

[0073] Specifically, the order agent is a computing unit with natural language processing capability, and can specifically extract key intention features in the communication text by using a semantic analysis algorithm. The order agent is configured to convert unstructured communication information into quantifiable production demand parameters. Each order agent is connected to a communication client of one or more front-end salespersons, and can acquire real-time communication records between the front-end salesperson and the customer.

[0074] The resource agent 20 is configured to generate a current idle time period according to an existing production scheduling plan, screen the order agents according to the current idle time period and the estimated delivery time points and the estimated production durations sent by the order agents, and obtain selected order agents, and select an order agent and a reservation time period according to evaluation scores of the resource agents on the selected order agents.

[0075] Specifically, the resource agent is a decision module with production scheduling optimization function, and can specifically analyze idle time periods of equipment or production lines by using a time window matching algorithm, and realize multi-order scheduling optimization by dynamic programming. The resource agent is configured to actively coordinate the space-time matching relationship between production resources and potential orders.

[0076] The selected order agent 10 is configured to receive an order instruction from the front-end salesperson, generate a real order, a real delivery time point and a real production duration according to the communication record information, and send the real order, the real delivery time point and the real production duration to the corresponding resource agent.

[0077] The resource agent 20 is further configured to determine a production time period according to the real delivery time point, the real production duration and the reservation time period, and update the evaluation score of the selected order agent according to a time deviation between the production time period and the reservation time period and a reservation realization rate.

[0078] In the embodiment, the system first monitors the business communication record in real time through the order agent, and when detecting that the customer has a placing order tendency and is in a hurry, a virtual order containing the predicted production duration and the predicted delivery time point is automatically generated. These virtual order parameters are pushed to the relevant resource agent for feasibility verification. The resource agent filters out the candidate order agent that meets the time requirement according to the current production scheduling load and the current idle time period. According to the evaluation score of the candidate order agent, the resource agent selects the order agent that can accept the reservation request, and generates a reservation time period; when the customer formally places an order, the resource agent determines the production time period according to the real delivery time point, the real production duration and the reservation time period, that is, flexibly binds the real order and the reserved reservation time period, realizes the smooth transition of the production plan, and further updates the evaluation score of the selected order agent according to the time deviation between the production time period and the reservation time period and the reservation redemption rate, to improve the accurate screening of the order agent's reservation request next time. The application can significantly reduce the risk of loss of emergency orders. The virtual order mechanism enables the production resources to enter the standby state in advance, and the reserved capacity can be directly enabled when the real order arrives. The dual-agent collaborative working mode realizes the closed-loop optimization of demand prediction and resource scheduling, which guarantees the timely delivery of regular orders while reserving flexible processing space for sudden demand.

[0079] In the optional embodiment of the application, the communication record information of the front-end business personnel and the customer is acquired, and the predicted delivery time point and the predicted production duration are generated according to the placing order tendency and the time urgency of the customer, comprising:

[0080] The preset placing order vocabulary in the communication record information is extracted, and the vocabulary frequency is determined, wherein the preset placing order vocabulary is divided into multiple tendency levels, and different tendency levels correspond to different frequency thresholds.

[0081] Specifically, the preset placing order vocabulary refers to the keywords or phrases related to the customer's placing order, such as "urgent" and "place an order immediately". The keywords such as "place an order immediately" and "urgent" are classified as high tendency level, and the keywords such as "consider buying" and "contact later" are classified as medium tendency level. The natural language processing technology can be used to extract the keywords from the communication record for quantifying the customer's placing order tendency. The tendency level refers to the level divided according to the influence degree of the keywords on the placing order decision, for example, "urgent" is divided into high tendency level, and the corresponding frequency threshold is low, which is convenient for quickly identifying high-intention customers.

[0082] According to the tendency level of the preset placing order vocabulary, the corresponding vocabulary frequency and the corresponding frequency threshold, the comprehensive frequency is determined.

[0083] Specifically, the comprehensive frequency is a comprehensive index obtained by weighted calculation of the vocabulary frequency and the frequency threshold of different tendency levels, and the calculation formula of the comprehensive frequency is as follows.

[0084]

[0085] wherein, is the weight of the i-th tendency level of the preset order word vocabulary, is the word frequency of the preset order word vocabulary of the i-th tendency level, is the frequency threshold of the preset order word vocabulary of the i-th tendency level, and n is the total number of tendency levels of the preset order word vocabulary. This calculation method can more accurately quantify customer intention and improve the accuracy of customer order tendency judgment by weighting and superimposing the influence of different level words, avoiding misjudgment of a single word or a single level.

[0086] If the word frequency of any one of the preset order word vocabularies is greater than the corresponding frequency threshold or the comprehensive frequency is greater than the comprehensive threshold, it is determined that the customer will place an order.

[0087] Extract the expected delivery time, current time, product type and quantity of the customer in the communication record information, and determine the virtual order, expected delivery time point, expected production duration and available duration.

[0088] Divide the available duration by the expected production duration to determine the time margin.

[0089] Specifically, the time margin refers to the ratio of the available duration to the expected production duration. For example, if the available duration is 10 days and the expected production duration is 8 days, the time margin is 1.25, which is used to measure the urgency of production arrangement.

[0090] If the expected production duration is less than or equal to the first preset duration and the time margin is less than the first margin threshold, it is determined that the customer's order is urgent.

[0091] If the expected production duration is greater than the first preset duration and the time margin is less than the second margin threshold, it is determined that the customer's order is urgent, wherein the first margin threshold is greater than the second margin threshold.

[0092] Specifically, the first preset duration can be set to 5 days or 10 days, etc., and the first margin threshold and the second margin threshold can be dynamically adjusted values, for example, 1.5 and 1.2 respectively, which are used to distinguish the urgency judgment criteria under different production duration conditions.

[0093] According to the expected delivery time point and the expected production duration, the latest production time period is determined as the hypothetical order production time period.

[0094] Specifically, when the order agent obtains the communication record, first, the preset order vocabulary is extracted through text analysis, for example, it is identified that the customer mentions "must complete delivery this week" multiple times. According to the tendency level to which the vocabulary belongs and its frequency, for example, "must" belongs to a high tendency level and its frequency exceeds the corresponding threshold, the system determines that the customer has a high probability of ordering. Then, combined with the expected delivery time proposed by the customer and the current time difference, for example, delivery is expected after 5 days, the available time is calculated as 5 days. According to the product category and quantity, the production time is estimated, for example, 3 days are needed to complete production, so the time margin is 5 / 3≈1.67. If the first preset time length is 3 days and the first margin threshold is 1.5, since the production time is equal to the first preset time length and the time margin 1.67 is greater than the threshold 1.5, the system determines that the order is not urgent; otherwise, if the production time is 4 days and the second margin threshold is 1.2, the time margin 1.25 is still greater than the threshold, so it is determined as an urgent order. Finally, according to the expected delivery time point and the production time, it is predicted that production needs to be started in 2 days at the latest, and this time period is marked as a hypothetical order production period for the resource agent to arrange production in advance.

[0095] Compared with the prior art, the existing method only starts production arrangement after receiving a formal order, while the present scheme predicts the customer's demand and generates a hypothetical production period in advance by analyzing the frequency of vocabulary and time parameters in the communication record before the customer places an order. For example, the prior art cannot handle the case where the customer expresses urgent demand in the communication but has not submitted an order, resulting in production resources being occupied by other orders; while the present scheme can identify such potential urgent orders in advance, providing a pre-screening basis for the resource agent, thereby reducing the loss of emergency orders due to resource conflicts.

[0096] In this embodiment, high-probability orders and their urgency can be automatically identified based on communication content before the customer places a formal order, and corresponding hypothetical production periods are generated. For example, when the customer frequently uses high-tendency-level vocabulary and the time margin is insufficient, the system reserves the production period for the order in advance to avoid resource occupation by non-urgent orders. Thus, the resource conflict problem caused by the lag of production arrangement in the prior art is solved, and the production resource allocation efficiency and order fulfillment rate are improved.

[0097] In an optional embodiment of the present application, the method for screening a plurality of order agents according to the current idle time period and the predicted delivery time point and the predicted production time length sent by the plurality of order agents to obtain the selected order agent includes:

[0098] Before the predicted delivery time point, analyze whether there is a time period on the current idle time period whose length is greater than or equal to the predicted production time length; or whether there is a time period on the current idle time period whose length is greater than or equal to the predicted production time length and whose coincidence degree with the hypothetical order production period is greater than a preset coincidence degree.

[0099] If yes, the order agent corresponding to the hypothetical order production period is taken as a candidate order agent.

[0100] If no, the reservation request of the order agent corresponding to the hypothetical order production period is not received.

[0101] Specifically, the current idle time period refers to a time interval in which the resource agent is not occupied by a production task, and can be determined by the time gap of the arranged tasks in the production scheduling plan. The hypothetical order production period refers to a hypothetical production time period generated by the order agent according to customer demand, which can be, for example, a time range obtained by reversing the production duration from the expected delivery time point. The preset coincidence degree refers to a threshold of the overlap ratio of the hypothetical order production period and the idle time period, which can be, for example, 70% or 80%, and is used to judge the matching degree of the two time windows.

[0102] When the resource agent screens the order agent, it first traverses all the idle time periods to determine whether there is a continuous idle time that meets the production duration requirement before the expected delivery time point. If yes, the order agent is directly listed as a candidate object. If no, it is further checked whether the coincidence degree of the hypothetical order production period and the idle time period is the preset coincidence degree. For example, if the preset coincidence degree is 80%, when the overlapping part of the hypothetical order production period and a certain idle time period accounts for 85% of the total length of the hypothetical period, and the duration of the certain idle time period is greater than or equal to the expected production duration, the order agent can still be selected as a candidate object because a small delay can be solved by negotiation with the customer. Through the double screening conditions, both the production duration requirement and the matching of the time window are guaranteed, and the subsequent production plan conflict caused by time misalignment is avoided.

[0103] In this embodiment, the matching accuracy of the idle time period and the order demand can be improved, and the risk of order execution failure caused by time window misalignment can be reduced. For example, when the idle time period of the resource agent partially overlaps with the hypothetical order production period, the order reservation request that can be adjusted is screened through the preset coincidence degree threshold, which not only retains the potentially executable reservation request, but also avoids the loss of orders caused by too strict full matching conditions, thereby improving the utilization rate of production resources.

[0104] In an optional embodiment of the present application, the selecting the order agent and the reservation time period according to the evaluation scores of the plurality of resource agents on the plurality of candidate order agents comprises:

[0105] According to the evaluation time points of the evaluation scores of the plurality of resource agents on one candidate order agent, the latest evaluation score and the latest evaluation time point are determined.

[0106] Specifically, the latest evaluation score refers to the score of the multiple resource agents' latest evaluation on the candidate order agent, and the evaluation time point can be recorded by using a time stamp, so as to ensure the timeliness of the evaluation result by using the latest data.

[0107] According to the other evaluation scores and the corresponding evaluation time points, and the latest evaluation score and the latest evaluation time point, a comprehensive evaluation score of the candidate order agent is determined.

[0108] Specifically, the comprehensive evaluation score is a weighted calculation result of the historical evaluation score and the latest evaluation score, which reflects the dynamic performance of the order agent by balancing the historical data and the latest data.

[0109] The comprehensive evaluation score is

[0110]

[0111]

[0112]

[0113] wherein, is the latest evaluation score, is the weight of the latest evaluation score, K is the total number of the other evaluation scores, is the time difference of the kth evaluation score in the other evaluation scores, is the total time difference, is the kth evaluation score in the other evaluation scores, is the evaluation time point of the kth evaluation score in the other evaluation scores, is the latest evaluation time point. By introducing the time difference weight mechanism, the influence of the historical evaluation data gradually weakens over time, so as to more accurately reflect the current state of the order agent, reduce the misjudgment caused by the out-of-date historical data, and improve the timeliness and accuracy of the order agent evaluation.

[0114] According to the comprehensive evaluation score and the preset score threshold, the multiple comprehensive evaluation scores are screened, the screened comprehensive evaluation scores are sorted from large to small, and an evaluation score sequence is obtained.

[0115] Specifically, the evaluation score sequence refers to a queue formed by arranging the screened comprehensive evaluation scores in descending order, so as to facilitate the quick selection of the order with high priority by using the ordered queue.

[0116] The order agent corresponding to the preset number of comprehensive evaluation scores is selected in sequence from the evaluation score sequence, and according to the production duration and the delivery time of the selected order agent, it is judged whether the time period to be occupied by the selected order agent conflicts.

[0117] Specifically, the preset number refers to an upper limit of the number of orders that the resource agent can select in the same time period, which can be set according to the production capacity or equipment load capacity of the resource agent. By limiting the number, resource overload or excessive concentration of orders on one resource agent can be avoided.

[0118] If there is no conflict, the virtual order of the selected order agent is added to the production plan of the resource agent, the reservation time period of the order agent is determined, and the current idle time period is updated.

[0119] If there is a conflict, the order agent with a higher comprehensive evaluation score is selected, the order agent with a lower comprehensive evaluation score is discarded, the virtual order of the selected order agent is added to the production plan of the resource agent, the reservation time period of the order agent is determined, and the current idle time period is updated.

[0120] According to the updated current idle time period, the to-be-selected order agent is screened to obtain an updated to-be-selected order agent and an updated evaluation score sequence, until a preset number of order agents or the to-be-selected order agent in the updated to-be-selected order agent is zero.

[0121] Specifically, when the resource agent receives reservation requests of multiple to-be-selected order agents, the latest evaluation score and the corresponding time point are extracted from the historical evaluation record of each order. The comprehensive evaluation score is calculated by weighting the historical score according to the time difference and combining the latest score to form a dynamic evaluation result of the order agent. Then, orders higher than a preset score threshold are screened, and a sorting queue is generated according to the order of the scores. A preset number of orders are selected from the queue in order, and it is checked whether there is a conflict in the production time period. If there is no conflict, the production plan is arranged and the idle time period is updated; if there is a conflict, the high-score order is retained and the low-score order is discarded. This process is repeated until a preset number of orders are selected or there is no remaining to-be-selected order, so that the resource agent can quickly respond to high-priority orders and dynamically adjust.

[0122] In this embodiment, by introducing a dynamic comprehensive evaluation mechanism and a conflict resolution strategy, the resource agent can update the order priority in real time, prioritize high-score orders in production arrangement under limited resources, and continuously optimize the production arrangement plan through a cyclic screening mechanism, thereby ensuring the stability of the production arrangement plan and improving the resource utilization rate.

[0123] In an optional embodiment of the present application, the production time period is determined according to the real delivery time point, the real production time length and the reservation time period.

[0124] It is judged whether the real delivery time point is in the reservation time period and whether the real production time length is less than or equal to the reservation time period.

[0125] Specifically, the real delivery time point refers to the actual delivery time node required by the customer, which can be specifically implemented by parsing the order data field generated by the business system, and is used to verify the consistency of the actual demand and the reservation plan. The real production duration refers to the production cycle required to complete the real order, which can be specifically calculated by the order product type, quantity and standard working hours of the production line, and is used to judge the actual production resource occupation. The reservation time period refers to the production plan period reserved by the resource agent for the order agent in advance.

[0126] If yes, the real order is inserted into the reservation time period, the production plan is updated, and it is determined whether the reservation is realized.

[0127] If no, the exceeding duration of the real delivery time point and the real production duration exceeding the reservation time period is analyzed, and it is determined whether the exceeding duration is greater than the preset duration threshold.

[0128] Specifically, the exceeding duration refers to the time offset of the real production demand and the reservation time period. When the resource agent receives the real order, it first verifies whether the delivery time is within the reservation range and the production duration does not exceed the reservation capacity. If the conditions are met, the order is directly inserted into the production scheduling plan, and the original reservation realization state is maintained. If there is an exceeding situation, the exceeding duration is calculated and compared with the preset duration threshold.

[0129] If the exceeding duration is less than or equal to the preset duration threshold, the production plan is adjusted according to the delivery time point of the production order in the production plan and the current idle time period, the production time period is obtained, and it is determined whether the reservation is realized.

[0130] If the exceeding duration is greater than the preset duration threshold, the corresponding real order is rejected, and it is determined that the reservation is not realized.

[0131] Specifically, when the exceeding amount is within an acceptable range, the production scheduling order of adjacent orders is dynamically adjusted or the idle time period is split to reallocate resources; when the exceeding amount exceeds the threshold, the order is rejected to avoid resource crowding. For example, the preset duration threshold can be set to 10% of the length of the reservation time period, and exceeding the threshold triggers the protection mechanism. Through this process, timely processing of high-priority orders can be ensured, and overall production scheduling chaos caused by excessive adjustment can be avoided.

[0132] In the embodiment, the accurate matching and dynamic response to the real order production demand are realized, the success rate of inserting the urgent order is improved to the quantifiable control level on the premise of ensuring the continuity of the production scheduling plan. In the order confirmation stage, the resource occupation commitment is established, and the elastic adjustment rule is set, so that the production system can maintain the stability of the scheduling and adapt to the reasonable range of demand changes, effectively balancing the contradiction between the plan rigidity and the execution flexibility. It can not only avoid the waste of resources caused by small-scale adjustment, but also prevent the production interruption caused by large-scale plan change, significantly reducing the risk of customer loss caused by temporary order conflict.

[0133] The updating the evaluation score of the selected order agent according to the time deviation between the production time period and the appointment time period and the appointment fulfillment rate comprises:

[0134] According to the time deviation and the length of the appointment time period, the appointment deviation rate is determined.

[0135] Specifically, the time deviation refers to the time difference between the real production time period and the appointment time period, which can be determined by calculating the difference between the start time of the production time period and the start time of the appointment time period, and the sum of the difference between the end time of the production time period and the end time of the appointment time period. The appointment deviation rate refers to the proportion of the time deviation to the length of the appointment time period, which is used to evaluate the relative degree of the time estimation deviation of the order agent.

[0136] According to the total number of appointments and the number of appointment fulfillments within a preset time period, the appointment fulfillment rate is determined.

[0137] Specifically, the appointment fulfillment rate refers to the proportion of the number of successful appointments of the order agent within a preset time period to the total number of appointments, which can be calculated by the ratio of the number of appointment fulfillments to the total number of appointments within a statistical period, and is used to measure the reliability of the historical performance of the order agent.

[0138] According to the appointment deviation rate, the appointment fulfillment rate and the existing evaluation score, the updated evaluation score is obtained.

[0139] Specifically, the updated evaluation score is

[0140]

[0141] Wherein, is the updated evaluation score, is the time deviation, is the length of the appointment time period, is the total number of appointments, is the number of appointment fulfillments, The calculation formula realizes dynamic optimization of the credit evaluation of the order agent, so that the resource agent can preferentially select orders with good fulfillment records for production scheduling. This credit decay mechanism based on historical performance can prevent low-credit orders that frequently change production plans from occupying key resources, thereby ensuring that emergency orders can obtain reliable production resource supply.

[0142] When the resource agent receives a real order, the actual production time period is compared with the reservation time period. For example, if the actual production start time is delayed by two hours from the reservation time, and the actual production duration is one hour longer than the reservation duration, the time deviation is three hours. At this time, the reservation deviation rate can be calculated as three hours divided by the total length of the original reservation time period. At the same time, the total number of reservations and the number of successful redemptions of the order agent in the past thirty days are counted, for example, the total number of reservations is twenty times, and the number of successful redemptions is fifteen times, so the reservation redemption rate is 75%. Then, the reservation deviation rate, the reservation redemption rate, and the existing evaluation score are substituted into the preset mathematical model, for example, a weighted calculation method is used, to generate an updated evaluation score. In this way, the resource agent can dynamically adjust the priority ranking of the order agent and optimize the subsequent production scheduling plan.

[0143] In this embodiment, by quantifying the reservation deviation rate and the reservation redemption rate, an evaluation score updating mechanism is established, and the time estimation accuracy and the fulfillment stability of the order agent are considered comprehensively when making production scheduling decisions, so that the resource agent can identify high-credibility order agents in advance, reduce resource conflicts caused by temporary adjustment of production plans, and thereby reduce the emergency order loss rate.

[0144] As shown in Figure 2 The production management method provided by the embodiment of the present application comprises the following steps:

[0145] S1: Obtain the communication record information of the front-end business personnel and the customer, and generate the predicted delivery time point and the predicted production duration according to the customer's order placing tendency and the time urgency degree.

[0146] S2: Send the predicted delivery time point and the predicted production duration to a plurality of resource agents, and send a reservation request to the resource agents.

[0147] S3: Generate the current idle time period according to the existing production scheduling plan of the resource agent.

[0148] S4: According to the current idle time period and the predicted delivery time point and the predicted production duration sent by the plurality of order agents, the plurality of order agents are screened to obtain the selected order agents.

[0149] S5: According to the evaluation scores of the plurality of resource agents on the plurality of selected order agents, the order agent and the reservation time period are selected.

[0150] S6: receiving the order instruction of the front-end service personnel, generating a real order, a real delivery time point and a real production duration according to the communication record information, and sending them to the corresponding resource agent.

[0151] S7: determining a production time period according to the real delivery time point, the real production duration and the reservation time period.

[0152] S8: updating the evaluation score of the selected order agent according to the time deviation between the production time period and the reservation time period and the reservation redemption rate.

[0153] In an optional embodiment of the present application, the communication record information between the front-end service personnel and the customer is obtained, and the expected delivery time point and the expected production duration are generated according to the customer's order tendency and the time urgency, comprising:

[0154] extracting the preset order word in the communication record information to determine the word frequency, wherein the preset order word is divided into multiple tendency levels, and different tendency levels correspond to different frequency thresholds;

[0155] determining the comprehensive frequency according to the tendency level of the preset order word, the corresponding word frequency and the corresponding frequency threshold;

[0156] If the word frequency of any one of the preset order words is greater than the corresponding frequency threshold or the comprehensive frequency is greater than the comprehensive threshold, it is determined that the customer will place an order;

[0157] extracting the expected delivery time, the current time, the product type and the quantity of the customer in the communication record information to determine the virtual order, the expected delivery time point, the expected production duration and the available duration;

[0158] determining the time margin using the available duration divided by the expected production duration;

[0159] If the expected production duration is less than or equal to the first preset duration and the time margin is less than the first margin threshold, it is determined that the customer's order is urgent;

[0160] If the expected production duration is greater than the first preset duration and the time margin is less than the second margin threshold, it is determined that the customer's order is urgent, wherein the first margin threshold is greater than the second margin threshold;

[0161] determining the latest production time period as the hypothetical order production time period according to the expected delivery time point and the expected production duration.

[0162] In an optional embodiment of the present application, the plurality of order agents are screened according to the current idle time period and the expected delivery time point and the expected production duration sent by the plurality of order agents to obtain the selected order agent, comprising:

[0163] whether there is a time period on the current idle time period with a time length greater than or equal to the predicted production time length; or whether there is a time period on the current idle time period with a time length greater than or equal to the predicted production time length and a coincidence degree greater than a preset coincidence degree with the hypothetical order production time period.

[0164] If yes, the order agent corresponding to the hypothetical order production time period is selected as a candidate order agent.

[0165] If no, the order agent corresponding to the hypothetical order production time period is not received.

[0166] In an optional embodiment of the present application, the selecting an order agent and a reservation time period according to the evaluation scores of the plurality of resource agents on the plurality of candidate order agents comprises:

[0167] According to the evaluation time point of the evaluation score of the plurality of resource agents on one candidate order agent, a latest evaluation score and a latest evaluation time point are determined.

[0168] According to the other evaluation scores and the corresponding evaluation time points, and the latest evaluation score and the latest evaluation time point, a comprehensive evaluation score of the candidate order agent is determined.

[0169] According to the comprehensive evaluation score and a preset score threshold, the plurality of comprehensive evaluation scores are screened, the screened comprehensive evaluation scores are sorted from large to small to obtain an evaluation score sequence.

[0170] The order agents corresponding to a preset number of comprehensive evaluation scores are selected in sequence from the evaluation score sequence, and whether the selected order agents will occupy a time period that conflicts is determined according to the production time length and the delivery time of the selected order agents.

[0171] If no, a virtual order of the selected order agent is added to the production plan of the resource agent, a reservation time period of the order agent is determined, and the current idle time period is updated.

[0172] If yes, an order agent with a larger comprehensive evaluation score is selected preferentially, an order agent with a smaller comprehensive evaluation score is discarded, a virtual order of the selected order agent is added to the production plan of the resource agent, a reservation time period of the order agent is determined, and the current idle time period is updated.

[0173] According to the updated current idle time period, the candidate order agents are screened to obtain updated candidate order agents and an updated evaluation score sequence, until a preset number of order agents are selected or the candidate order agents in the updated candidate order agents are zero.

[0174] In an optional embodiment of the present application, the determining of the production time period according to the real delivery time point, the real production time length and the reservation time period comprises:

[0175] determining whether the real delivery time point is within the reservation time period and the real production time length is less than or equal to the reservation time period.

[0176] If yes, the real order is inserted into the reservation time period, the production plan is updated, and it is determined that the reservation is redeemed.

[0177] If no, the exceeding time length of the real delivery time point and the real production time length beyond the reservation time period is analyzed, and it is determined whether the exceeding time length is greater than a preset time length threshold.

[0178] If the exceeding time length is less than or equal to the preset time length threshold, the production plan is adjusted according to the delivery time point of the production order in the production plan and the current idle time period, the production time period is obtained, and it is determined that the reservation is redeemed.

[0179] If the exceeding time length is greater than the preset time length threshold, the corresponding real order is rejected, and it is determined that the reservation is not redeemed.

[0180] The present application solves the problem of lag in processing of emergency orders caused by passive response of the traditional production control system by predicting potential orders and planning production resources in advance before the customer places an order, and has the advantages of active prediction of potential orders, advance planning of production resources and improvement of emergency order processing efficiency.

[0181] An electronic device provided in an embodiment of the present application comprises a memory and a processor; the memory is configured to store a computer program; and the processor is configured to implement the production control method relying on a business intelligent agent when the computer program is executed.

[0182] A computer readable storage medium provided in an embodiment of the present application has a computer program stored thereon, and the computer program, when executed by a processor, implements the production control method relying on a business intelligent agent.

[0183] In the present embodiment, the electronic device and the computer readable storage medium have similar advantages to those of the production control system relying on a business intelligent agent and the production control method relying on a business intelligent agent, and thus will not be described herein.

[0184] An electronic device that can be a server or a client of the present application will now be described, which is an example of a hardware device that can be applied to aspects of the present application. The electronic device is intended to represent a wide variety of digital electronic computing devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computing devices. The electronic device can also represent a wide variety of mobile devices, such as personal digital assistant devices, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0185] The electronic device includes a computing unit that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The computing unit, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0186] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), or a random access memory (RAM). In the present application, the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application. In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0187] It is to be noted that, in the present text, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0188] The above examples are merely intended for describing the technical solutions of the present application, but not for limiting the same; even though the present application has been described in detail with reference to the foregoing examples, it should be understood by those skilled in the art that the technical solutions recorded in the foregoing examples can be modified, or some of the technical features can be replaced equivalently; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A production management system based on a business intelligence agent, characterized by, The system comprises a plurality of order agents and a plurality of resource agents; The order agent is configured to acquire communication record information of a front-end business staff and a customer, generate an estimated delivery time point and an estimated production duration according to an order placing tendency and a time urgency degree of the customer, and send the estimated delivery time point and the estimated production duration to the plurality of resource agents and send a reservation request to the resource agents; The resource agent is configured to generate a current idle time period according to an existing production scheduling plan of the resource agent, screen the plurality of order agents according to the current idle time period and the estimated delivery time point and the estimated production duration sent by the plurality of order agents, and obtain order agents to be selected; The resource agent is further configured to select an order agent and a reservation time period according to evaluation scores of the plurality of resource agents on the plurality of order agents to be selected; The selected order agent is configured to receive an order placing instruction of the front-end business staff, generate a real order, a real delivery time point and a real production duration according to the communication record information, and send the real order, the real delivery time point and the real production duration to the corresponding resource agent; The resource agent is further configured to determine a production time period according to the real delivery time point, the real production duration and the reservation time period, and update the evaluation score of the selected order agent according to a time deviation between the production time period and the reservation time period and a reservation realization rate; The method comprises the following steps: determining whether the real delivery time point is within the reservation time period and whether the real production duration is less than or equal to the reservation time period; if yes, inserting the real order into the reservation time period, updating a production plan, and determining that the reservation is realized; if no, analyzing an exceeding time period of the real delivery time point and the real production duration exceeding the reservation time period, and determining whether the exceeding time period is greater than a preset time threshold; if the exceeding time period is less than or equal to the preset time threshold, adjusting the production plan according to a delivery time point of a production order in the production plan and the current idle time period, obtaining the production time period, and determining that the reservation is realized; if the exceeding time period is greater than the preset time threshold, rejecting the corresponding real order, and determining that the reservation is not realized; The method comprises the following steps: determining a reservation deviation rate according to the time deviation and a length of the reservation time period; determining the reservation realization rate according to a total number of reservations within a preset time period and a number of realized reservations; obtaining an updated evaluation score according to the reservation deviation rate, the reservation realization rate and the existing evaluation score; The updated evaluation score is The method comprises the following steps: wherein, is an updated evaluation score, is a time offset, is a length of the appointment time period, is a total number of appointments, is a number of appointments redeemed, is an existing evaluation score.

2. The business intelligence agent-enabled production management system of claim 1, wherein, extracting preset order placing vocabularies in the communication record information, and determining a vocabulary frequency, wherein the preset order placing vocabularies are divided into a plurality of tendency grades, and different tendency grades correspond to different frequency thresholds; determining a comprehensive frequency according to the tendency grades of the preset order placing vocabularies, the corresponding vocabulary frequencies and the corresponding frequency thresholds; if the vocabulary frequency of any one of the preset order placing vocabularies is greater than the corresponding frequency threshold or the comprehensive frequency is greater than a comprehensive threshold, it is determined that the customer will place an order. ​ extracting a delivery time expected by the customer, a current time, and a product category and quantity in the communication record information, determining a virtual order, a predicted delivery time point, a predicted production duration, and an available duration; determining a time margin by dividing the available duration by the predicted production duration; if the predicted production duration is less than or equal to a first preset duration and the time margin is less than a first margin threshold, determining that the order of the customer is urgent; if the predicted production duration is greater than the first preset duration and the time margin is less than a second margin threshold, determining that the order of the customer is urgent, wherein the first margin threshold is greater than the second margin threshold; determining a latest production time period as a hypothetical order production time period according to the predicted delivery time point and the predicted production duration.

3. The business intelligence agent-enabled production management system of claim 2, wherein, The comprehensive frequency is wherein, is a weight of the i-th tendency level of the preset lower-order vocabulary, is a word frequency of the preset lower-order vocabulary of the i-th tendency level, is a frequency threshold of the preset lower-order vocabulary of the i-th tendency level, and n is a total number of tendency levels of the preset lower-order vocabulary.

4. The business intelligence agent-enabled production management system of claim 2, wherein, The screening of the plurality of order agents according to the current idle time period and the predicted delivery time point and the predicted production duration sent by the plurality of order agents includes: analyzing whether there is a time period with a duration greater than or equal to the predicted production duration on the current idle time period before the predicted delivery time point, or whether there is a time period with a duration greater than or equal to the predicted production duration and a coincidence degree greater than a preset coincidence degree with the hypothetical order production time period on the current idle time period; if yes, the order agent corresponding to the hypothetical order production time period is taken as a candidate order agent; if no, the order agent corresponding to the hypothetical order production time period does not receive a reservation request.

5. The business intelligence agent enabled production management system of claim 2, wherein, The selection of the order agent and the reservation time period according to the evaluation scores of the plurality of resource agents on the plurality of candidate order agents includes: determining a latest evaluation score and a latest evaluation time point according to the evaluation time point of the evaluation score of one candidate order agent by the plurality of resource agents; determining a comprehensive evaluation score of the candidate order agent according to the other evaluation scores and the corresponding evaluation time points, and the latest evaluation score and the latest evaluation time point; screening the plurality of comprehensive evaluation scores according to the comprehensive evaluation scores and a preset score threshold, and sorting the screened comprehensive evaluation scores from large to small to obtain an evaluation score sequence; selecting a preset number of order agents corresponding to the comprehensive evaluation scores in sequence from the evaluation score sequence, and judging whether a time period to be occupied by the selected order agents conflicts according to the production duration and the delivery time of the selected order agents; if no, adding the virtual order of the selected order agent to the production plan of the resource agent, determining the reservation time period of the order agent, and updating the current idle time period; if yes, preferentially selecting an order agent with a larger comprehensive evaluation score, discarding an order agent with a smaller comprehensive evaluation score, adding the virtual order of the selected order agent to the production plan of the resource agent, determining the reservation time period of the order agent, and updating the current idle time period; screening the candidate order agents according to the updated current idle time period to obtain updated candidate order agents and an updated evaluation score sequence, until a preset number of order agents are selected or the candidate order agents in the updated candidate order agents are zero.

6. The business intelligence agent-enabled production management system of claim 5, wherein, The comprehensive evaluation score is wherein, is the latest evaluation score, is the weight of the latest evaluation score, K is the total number of other evaluation scores, is the time difference of the kth evaluation score among the other evaluation scores, is the sum of the time differences, is the kth evaluation score among the other evaluation scores, is the evaluation time point of the kth evaluation score among the other evaluation scores, is the latest evaluation time point.

7. A production management method based on a business intelligence agent, characterized by, It includes: Obtain communication record information of a front-end service personnel and a customer, generate an estimated delivery time point and an estimated production duration according to an order placing tendency and a time urgency of the customer; Send the estimated delivery time point and the estimated production duration to a plurality of resource agents, and send a reservation request to the resource agents; Generate a current idle time period according to an existing production scheduling plan of the resource agents; Screen a plurality of order agents according to the current idle time period and the estimated delivery time point and the estimated production duration sent by the plurality of order agents, and obtain to-be-selected order agents; Select an order agent and a reservation time period according to evaluation scores of the plurality of resource agents on the plurality of to-be-selected order agents; Receive an order placing instruction of the front-end service personnel, generate a real order, a real delivery time point and a real production duration according to the communication record information, and send them to a corresponding resource agent; Determine a production time period according to the real delivery time point, the real production duration and the reservation time period; Update the evaluation score of the selected order agent according to a time deviation between the production time period and the reservation time period and a reservation realization rate; The determination of the production time period according to the real delivery time point, the real production duration and the reservation time period comprises: Determine whether the real delivery time point is within the reservation time period and whether the real production duration is less than or equal to the reservation time period; If yes, insert the real order into the reservation time period, update the production plan, and determine that the reservation is realized; If no, analyze an exceeding duration of the real delivery time point and the real production duration exceeding the reservation time period, and determine whether the exceeding duration is greater than a preset duration threshold; If the exceeding duration is less than or equal to the preset duration threshold, adjust the production plan according to a delivery time point of a production order in the production plan and the current idle time period, obtain the production time period, and determine that the reservation is realized; If the exceeding duration is greater than the preset duration threshold, reject the corresponding real order, and determine that the reservation is not realized; The updating of the evaluation score of the selected order agent according to the time deviation between the production time period and the reservation time period and the reservation realization rate comprises: Determine a reservation deviation rate according to the time deviation and a length of the reservation time period; Determine the reservation realization rate according to a total number of reservations within a preset duration and a number of realized reservations; Obtain an updated evaluation score according to the reservation deviation rate, the reservation realization rate and the existing evaluation score; The updated evaluation score is wherein, is an updated evaluation score, is a time offset, is a length of the appointment time period, is a total number of appointments, is a number of appointments redeemed, is an existing evaluation score.

Images