Large and small model mixing-based plan management method and device, equipment and medium

Abstract

The invention discloses a plan management method and device based on big and small model mixing, equipment and a medium. The method comprises the following steps: acquiring real-time data of each business field; based on the first model group of each business field, analyzing the real-time data of each business field to obtain a data analysis result of each business field; summarizing the data analysis results of the business fields to obtain a summarized result; based on the second model group, performing comprehensive analysis on the summarized result; and determining a comprehensive analysis result to generate a target plan, and distributing the target plan to each business field. According to the method, data of a single field is analyzed by using a method of combining a large model and a small model, then the target plan is generated by using the large model, and then the plan execution condition is monitored, so that the target plan can be adjusted in time, making of the target plan is more efficient, and the effects of reducing cost and improving efficiency are achieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a planning management method, device, equipment and medium based on mixing of large and small models. Background Art

[0002] Large models are based on massive data in any field, while small models are often based on professional data in vertical fields. Professional data in many industries is private and not open to the public. Therefore, when solving industry problems, a combination of large and small models is needed. Large models have more parameters and stronger fitting capabilities, and can capture complex patterns and details in the data. Small models have fewer parameters and simpler model structures. They can "focus" on learning data in a narrow field and are suitable for specific applications. They require less data and lock in specific scenarios to win with high efficiency. For some complex tasks and large amounts of data, large models may be a good choice, while small models may be more suitable for scenarios with limited resources or that need to be deployed on devices.

[0003] The affairs of an enterprise are completed by multiple departments. Some are internal affairs of a single department, covering a single field with simple logic; some affairs need to be solved collaboratively across multiple departments, which is a full-process collaborative problem in complex scenarios; some require global scenarios across industries and fields, which are more complex. How to mix and match large and small models, how to select appropriate models according to different scenarios, specific tasks, data volume, computing resources, etc., and how to efficiently manage enterprise affairs are technical issues that need to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a planning management method, device, equipment and medium based on a mixture of large and small models, so as to solve the problem of how to select suitable models according to different scenarios, specific tasks, data volume, computing resources, etc., mix large and small models, and efficiently manage corporate affairs.

[0005] In order to solve the above technical problems, a technical solution adopted by the present invention is to provide a planning management method based on a mixture of large and small models, and the method includes the following steps: acquiring real-time data of each business field; based on the first model group of each business field, analyzing the real-time data of each business field to obtain data analysis results of each business field; summarizing the data analysis results of each business field to obtain summary results; based on the second model group, performing comprehensive analysis on the summary results; determining that the result of the comprehensive analysis generates a target plan, and distributing the target plan to each business field.

[0006] In some embodiments, before acquiring the real-time data of each business field, the step is also included: collecting historical data of each business field, and using the historical data of each business field to construct the first model group of each business field and the second cross-domain model group.

[0007] In some embodiments, the data analysis results of each business field include a first analysis result, and the real-time data of each business field is analyzed based on the first model group of each business field to obtain the data analysis results of each business field, including the steps of: using the small model in the first model group of each business field to analyze the real-time data of each business field to obtain the first analysis result.

[0008] In some embodiments, the data analysis results of each business field also include a second analysis result. The first model group based on each business field analyzes the real-time data of each business field to obtain the data analysis results of each business field, and also includes the step of using the large model in the first model group of each business field to analyze the real-time data of each business field to obtain the second analysis result.

[0009] In some embodiments, after distributing the target plan to each business field, the method further includes the step of formulating and executing a field plan for each business field by using the first model group and the target plan for each business field.

[0010] In some embodiments, when executing the field plans of the various business fields, monitor whether the target plan is updated; if the target plan is updated, modify the field plans of the various business fields according to the latest target plan; if the target plan is not updated, continue to monitor whether the target plan is updated.

[0011] In some embodiments, the method further includes the steps of: based on the second model group, decomposing and arranging the received total task to obtain multiple subtasks with a set execution order; based on the expert model in the first model group of each business field, analyzing the real-time data of each business field, and executing the subtasks to obtain the subtask execution results of each business field; summarizing the subtask execution results of each business field to obtain the summary task results; based on the second model group, comprehensively analyzing the summary task results to obtain a supplementary plan; judging whether the total task is completed, if the total task is not completed, returning to the expert model in the first model group based on the each business field, analyzing the real-time data of each business field, and executing the subtasks to obtain the subtask execution results of each business field; if the total task is completed, distributing the supplementary plan to each business field.

[0012] The present invention also provides a planning management device based on a mixture of large and small models, the device comprising: a data acquisition unit, used to acquire real-time data of each business field; a data analysis unit, used to analyze the real-time data of each business field based on the first model group of each business field, and obtain data analysis results of each business field; an analysis result summary unit, used to summarize the data analysis results of each business field, and obtain summary results; a comprehensive analysis unit, used to perform comprehensive analysis on the summary results based on the second model group; a plan distribution unit, used to determine that the result of the comprehensive analysis generates the target plan, and distribute the target plan to each business field.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0014] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0015] The beneficial effects of the present invention are as follows: the present invention discloses a plan management method, device, equipment and medium based on a mixture of large and small models, the method comprising the following steps: obtaining real-time data of each business field; analyzing the real-time data of each business field based on the first model group of each business field to obtain the data analysis results of each business field; summarizing the data analysis results of each business field to obtain the summary results; based on the second model group, comprehensively analyzing the summary results; determining that the result of the comprehensive analysis generates a target plan, and distributing the target plan to each business field. The method of the present invention can be applied to a variety of enterprise scenarios, including production scheduling plan generation and execution scenarios, sales plan generation and execution, etc., by constructing the first model group of each business field and the second cross-domain model group. Among them, each business field uses the first model group to complete the simple and complex transaction analysis related to the target, as well as the monitoring during the execution of the target plan, and the cross-domain large model in the second model group integrates the data analysis results of the first model group of each business field to generate and adjust the target plan. Using the method of the present invention for plan management makes the internal plan formulation of the enterprise more reasonable and more efficient, and the combination of large and small models is applied to the digital transformation of the enterprise, which has the effect of reducing costs and increasing efficiency relative to the solution of using large models completely. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of an embodiment of a planning management method based on mixing of large and small models of the present invention;

[0017] Figure 2It is a partial flow chart of another embodiment of a planning management method based on mixing of large and small models of the present invention;

[0018] Figure 3 It is a partial flow chart of another embodiment of a planning management method based on mixing of large and small models of the present invention;

[0019] Figure 4 It is a structural block diagram of an embodiment of a planning management device based on mixing of large and small models of the present invention;

[0020] Figure 5 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention;

[0021] Figure 6 It is a schematic block diagram of an embodiment of a computer-readable storage medium of the present invention. DETAILED DESCRIPTION

[0022] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. Preferred embodiments of the present invention are provided in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0023] It should be noted that, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used in this specification includes any and all combinations of one or more related listed items.

[0024] like Figure 1 As shown, a planning management method based on mixing of large and small models of the present invention is shown, and the method comprises the following steps:

[0025] Step S1: Acquire real-time data of each business area.

[0026] Among them, each business field refers to the business fields corresponding to the internal affairs of multiple single departments within the enterprise, and one single department internal affair corresponds to one business field. Real-time data refers to business field data within a certain period of time.

[0027] Furthermore, before step S1, the method further includes the steps of collecting historical data of each business field, and using the historical data to construct a first model group of each business field and a second cross-field model group.

[0028] When constructing the first model group of each business field, for each business field, several small models and several large models of the field are constructed according to various historical data of the field, and these small models and large models together constitute the first model group available in the field. The first model group may also include a model information list (which may be in the form of a file or an open interface), and the list includes information such as the name, function, efficiency, input requirements and output format of each model in this model group.

[0029] It should be noted that for each first model group, the large model is established based on the overall data of the industry in which each business field is located, or based on the complex data within the multidimensional enterprise of each business field; the small model is established based on the simple data of each business field within the enterprise.

[0030] As an example, the first model group of each business field includes a production line model group, a logistics model group, a warehouse model group, a human resources model group and a demand model group.

[0031] Among them, the production line model group includes a large production line model and a small production line model. The large production line model is established based on multiple types of data, and the small production line model is established based on one type of data. The production line model group is used to analyze various types of data on the production line, such as equipment operation data, environmental data, production quality data, etc. The large production line model is used for comprehensive analysis of multiple types of data, and the small production line model is used for analysis of one type of data.

[0032] The logistics model group includes a large logistics industry model and a small logistics model within the enterprise. The large logistics industry model is built based on logistics industry data and is used to track cargo status, optimize transportation routes and warehouse layout, predict inventory demand and transportation time, etc. The small logistics model is built based on the enterprise's own logistics data and is used for logistics personnel arrangements, cargo classification and identification, cargo handling arrangements, etc.

[0033] The warehouse model group includes a large warehouse model and a small warehouse model. The large warehouse model and the small warehouse model are constructed based on information such as inventory, cargo movement, and order information. The large warehouse model is used to predict inventory trends based on inventory information, order information, shipping information, and cargo information; the small warehouse model is used to determine the cargo placement and picking path.

[0034] The human resource model group includes a large human resource model and a small human resource model. The large human resource model is established based on personnel capabilities, personnel turnover, personnel work information, etc. for human resource allocation. The small human resource model is constructed based on personnel training, work experience, age, etc. for personnel capability assessment.

[0035] The demand model group includes a large demand analysis model and a small demand analysis model. The large demand analysis model is built based on multi-dimensional information such as market data, sales data, and customer information to determine whether product demand exists. The small demand analysis model includes a small market trend analysis model and a small customer portrait model. A small market trend analysis model is built based on market data to analyze market trends; a small customer portrait model is built based on customer information to determine user demand tendencies.

[0036] Furthermore, when constructing the second model group across business fields, one or more large models across business fields are obtained based on historical data training of multiple business fields within the enterprise to form the second model group. The second model group has the ability to conduct comprehensive analysis of real-time data in each business field. At the same time, the second model group can also access the model information list of the first model group.

[0037] As an example, the second model group includes multiple cross-business domain models such as the demand analysis model and the production scheduling model. The demand analysis model is built based on historical data such as market data, sales data, and customer orders, and the production scheduling model is built based on historical data such as human resources data, warehouse data, logistics data, and production line data.

[0038] In step S1, after acquiring the real-time data of each business field, the real-time data of each business field is respectively transmitted to the first model group of the corresponding field, and the first model group analyzes the real-time data to obtain the key information therein. After that, the key information of each business field can be summarized and transmitted to the second model group across business fields, and the second model group conducts a comprehensive analysis of the key information of each business field.

[0039] Step S2: Based on the first model group of each business field, the real-time data of each business field is analyzed to obtain data analysis results of each business field.

[0040] It should be noted that the data analysis results of each business field include the first analysis results, and the first analysis results are related to the small models in the first model group. Each business field corresponds to an analysis module, that is, multiple business departments within the enterprise correspond to analysis modules of multiple business fields respectively. The analysis modules of each business field correspond one-to-one to the first model group of each business field. The first model group of each business field can be called to analyze the data of the business field to obtain the first analysis results of each business field. Among them, the small models in the first model group of each business field are used to analyze the real-time data of each business field to obtain the first analysis results.

[0041] As an example, the production line analysis module corresponds to the production line model group, calls the small production line model in the production line model group, and analyzes a type of data on the production line in real time, such as equipment operation data / environmental data / production quality data / other types of data. The first analysis result obtained is one of the types of data. The logistics analysis module corresponds to the logistics model group, calls the small logistics model in the logistics model group, and analyzes the company's own logistics data in real time. The first analysis result obtained is logistics personnel arrangement, cargo classification and identification, cargo handling arrangement, and logistics company capacity. Correspondingly, for analysis modules in other business areas, the first analysis result obtained by calling the first model group for analysis corresponds to the function of the small model in the corresponding business area, which will not be repeated here.

[0042] Furthermore, in some fields, the corresponding first model group may not contain a large model, and the data analysis result of the corresponding business field only includes the first analysis result. In other business fields, the corresponding first model group contains one or more large models. For the first model group containing a large model, the data analysis result of the corresponding business field also includes a second analysis result, which is related to the large model in the first model group and is specifically additional information of the business field.

[0043] The large model in the first model group of each business field is used to analyze the real-time data of each business field to obtain the second analysis result.

[0044] As an example, for the large model of the production line in the production line model group, the large model of the production line is used to analyze various types of data on the production line in real time, such as equipment operation data, environmental data, production quality data and other different types of data, and a comprehensive analysis is performed on various types of data, and the second analysis results obtained are various types of data on the production line. For the large model of the logistics industry in the logistics model group, the large model of the logistics industry is used to analyze the logistics industry data in real time, and the second analysis results obtained are tracking the status of goods, optimizing transportation routes and warehouse layouts, predicting inventory requirements and transportation time, etc. Correspondingly, for the first model group in other business areas, the second analysis results obtained by the analysis correspond to the functions of the large models in the corresponding business areas, which will not be repeated here.

[0045] In this embodiment, for each business field, a small model in the first model group of the business field is used to analyze the real-time data of the business field to obtain a first analysis result; for the first model group containing a large model, the large model therein is used to analyze the real-time data of the business field to obtain additional information of the business field as the second analysis result, thereby obtaining key information of each business field, that is, the data analysis results of each business field.

[0046] Step S3: Summarize the data analysis results of each business field to obtain summary results.

[0047] In this process, the data analysis results obtained by each business field based on its own first model group are summarized to obtain a summary result of the key information of all business fields.

[0048] Step S4: Based on the second model group, a comprehensive analysis is performed on the summary results.

[0049] After receiving the summary results, the second model group calls one or more cross-domain large models in the second model group to conduct a comprehensive analysis of the summary results according to time rules or conditional trigger rules.

[0050] As an example, according to the time rule, the summary results are comprehensively analyzed within a certain period of time based on the cross-business domain large model in the second model group, such as analyzing whether the product demand actually exists, so as to determine whether production scheduling is needed. When the product demand actually exists and it is determined that production scheduling is needed, a production scheduling plan is generated.

[0051] It should be noted that while conducting a comprehensive analysis of the summary results, a target plan may be generated, so the next step of judgment operation is required.

[0052] Step S5: Determine whether a target plan has been generated.

[0053] If the target plan is not generated, the process returns to step S1. If the target plan is generated, the process proceeds to step S6.

[0054] Step S6: Based on the generated target plan, distribute the target plan to each business area.

[0055] It should be noted that, for different usage requirements, the target plan includes production scheduling plans, sales plans and other plans for achieving corresponding goals.

[0056] As an example, if the target plan is a production schedule, it is determined whether a production schedule has been generated. If no production schedule has been generated, the process returns to step S3; if a production schedule has been generated, the production schedule is distributed to each business area.

[0057] Furthermore, after generating a target plan based on the results of the comprehensive analysis determined in step S5 and step S6 and distributing the target plan to each business field, the method also includes the steps of: formulating and executing a field plan for each business field using the first model group and the target plan for each business field.

[0058] It should be noted that the target plan is a plan that requires all areas to achieve the corresponding goals in order to achieve the overall goal. For each business area, executing part of the target plan can achieve the corresponding goal. Therefore, for each business area, it is necessary to formulate and implement the field plan corresponding to each business area based on the first model group and the target plan of each business area. The field plan is the plan that the corresponding field needs to achieve the sub-goals of the field.

[0059] In some embodiments, when executing each business area plan, it is also necessary to monitor whether the target plan is updated; if the target plan is updated, the field plan of each business area is modified according to the latest target plan; if the target plan is not updated, continue to monitor whether the target plan is updated.

[0060] like Figure 2 As shown, the method of the present invention also includes the following steps:

[0061] Step S12: Based on the monitoring model in the first model group of each business field, monitor the real-time data of each business field.

[0062] Among them, the small models in the first model group include a monitoring model, which can monitor whether the real-time data of the corresponding business field has changed.

[0063] Step S13: Summarize the monitoring results of each business area to obtain a summary monitoring result.

[0064] In this process, the monitoring results obtained by each business field based on the monitoring model in the respective first model group are summarized to obtain a summary monitoring result of the real-time data that has changed in all business fields.

[0065] Step S14: Based on the second model group, a comprehensive analysis is performed on the aggregated monitoring results.

[0066] Among them, after receiving the summarized monitoring results, the second model group calls one or more cross-domain large models in the second model group to conduct a comprehensive analysis of the summarized monitoring results to determine whether the target plan needs to be adjusted. If the target plan needs to be adjusted, a new target plan is generated; if the target plan does not need to be adjusted, a new target plan will not be generated.

[0067] Step S15: Determine whether a new target plan is generated.

[0068] If a new target plan is not generated, the process returns to step S12 to re-monitor the real-time data of each business area; if a new target plan is generated, the process proceeds to step S16.

[0069] Step S16: Based on the generated new target plan, distribute the new target plan to each business area.

[0070] It should be noted that the above steps S12 to S16 are executed synchronously with steps S2 to S6, and whether the target plan is updated is determined based on whether a new target plan is generated in step S15.

[0071] As an example, in step S12, the monitoring model synchronously monitors the real-time data of the corresponding business field. In step S13, the monitoring results of each business field are summarized, and then the summarized monitoring results are comprehensively analyzed according to step S14 to determine whether the production schedule needs to be adjusted. If the production schedule needs to be adjusted, a new production schedule is generated. Then according to step S15, it is determined whether a new production schedule has been generated. If a new production schedule has not been generated, return to step S13 and re-summarize the monitoring results of each business field; if a new production schedule has been generated, enter step S16 and distribute the new production schedule to each business field. Finally, the field plan of each business field can be modified according to the latest production schedule.

[0072] In some embodiments, while executing step S1, the overall task to be achieved is also received (corresponding to the generation / monitoring update of the target plan mentioned in the above content).

[0073] Therefore, if Figure 3 As shown, the method of the present invention also includes the following steps:

[0074] Step S21: Based on the second model group, the received overall task is decomposed and arranged to obtain a plurality of subtasks with a set execution order.

[0075] It should be noted that the small models in the first model group include expert models, which are small models built based on expert knowledge and experience.

[0076] Among them, the total task can be a multi-modal input method such as text, image, and audio. After receiving the total task, the second model group performs semantic understanding and task analysis on the received total task. Secondly, since the second model group can access the model information list of the first model group, the second model group can directly access the model information list file of the first model group or call the open interface of the first model group to obtain the model information of the first model group. Then, the second model group decomposes the total task into multiple subtasks corresponding to the multiple expert models of the first model group, and arranges the order of task execution. Arrangement refers to setting the order and dependency of multiple subtasks. For example, assuming there are 4 subtasks, subtask A and subtask B are executed synchronously first. After subtask A and subtask B are executed, subtask C and subtask D are executed synchronously.

[0077] Step S22: Based on the expert model in the first model group of each business field, analyze the real-time data of each business field, execute subtasks, and obtain the subtask execution results of each business field.

[0078] The subtask execution results include the data analysis results mentioned above and the execution information of the corresponding subtasks. For each business domain, according to the received subtasks and arrangements, the expert model in the model group of the business domain is used to analyze the real-time data of the business domain, and the subtasks are completed to obtain the subtask execution results.

[0079] Step S23: Summarize the execution results of the subtasks in each business area to obtain a summary task result.

[0080] The summary task result includes the summary results mentioned above and the execution information of each subtask. In this process, the subtask execution results obtained by each business field based on its first model group are summarized to obtain the summary task result that summarizes the subtask execution results of all fields.

[0081] Step S24: Based on the second model group, a comprehensive analysis is performed on the summary task results to obtain a supplementary plan.

[0082] Among them, the supplementary plan is a supplement provided when the current target plan cannot achieve the required goal, which is used to ensure the achievement of the required goal. The supplementary plan includes the solution to the corresponding problem that needs to be solved to achieve the current overall task, or the relevant field knowledge required to achieve the current overall task, or the pending operations required to achieve the current overall task. According to the summary task results, a comprehensive analysis is performed based on the cross-business domain large model in the second model group to obtain the supplementary plan required to achieve the current overall task.

[0083] Step S25: Determine whether the overall task is completed.

[0084] If the overall task is not completed, return to step S22; if the overall task is completed, enter step S26.

[0085] Step S26: Based on the obtained supplementary plan, distribute the supplementary plan to each business area.

[0086] It should be noted that steps S21 to S26 correspond synchronously with steps S1 to S6 in sequence.

[0087] As an example, this embodiment can be used in the field of sales. For example, the second model group receives a sales plan generation task, decomposes the sales plan generation task into multiple subtasks, and assigns them to expert models in the fields of sales analysis, sales forecasting, sales comparison, etc. Secondly, the expert model in the sales analysis field analyzes the current sales situation, the expert model in the sales forecasting field generates future sales situation, and the expert model in the sales comparison field compares the predicted future sales situation with the current sales situation to derive problems in sales. Based on the analysis results in these fields, the sales large model in the second model group generates the personnel skills and knowledge, personnel optimization solutions, and new sales solutions required to solve these problems.

[0088] In this embodiment, the model information of the first model group can be perceived by the second model group, and the small models in the first model group include expert models. After the second model group receives the overall task and decomposes the task, the sub-tasks can be executed by expert models in different first model groups, and a comprehensive analysis can be performed based on the execution results of the sub-tasks. Ultimately, domain knowledge, operations to be performed or solutions corresponding to the overall task are generated, making the planning process more efficient.

[0089] Based on the same inventive concept, Figure 4 As shown, the present invention also provides a planning management device based on mixing of large and small models, the device comprising:

[0090] The data collection unit 101 is used to obtain real-time data in various business areas.

[0091] The data analysis unit 102 is used to analyze the real-time data of each business field based on the first model group of each business field to obtain data analysis results of each business field.

[0092] The analysis result summarizing unit 103 is used to summarize the data analysis results of each business field to obtain a summary result.

[0093] The comprehensive analysis unit 104 is used to perform comprehensive analysis on the summary results based on the second model group.

[0094] The plan distribution unit 105 is used to determine that a target plan is generated based on the result of the comprehensive analysis, and distribute the target plan to each business area.

[0095] In the present application, other technical features in the above-mentioned planning management device based on mixing of large and small models are the same as the features disclosed in the above-mentioned method embodiment, and will not be repeated here.

[0096] Based on the same inventive concept, the present application also provides an electronic device, which includes a processor, a memory and a communication circuit, and the processor is connected to the memory and the communication circuit respectively; wherein the communication circuit is used for communication connection, the memory is used to store a computer program, and the processor is used to execute the computer program to implement the above method.

[0097] See also Figure 5 The electronic device described in the embodiment of the present application may specifically include a processor 210 and a memory 220. The memory 220 is coupled to the processor 210.

[0098] The processor 210 is used to control the operation of the electronic device. The processor 210 may also be referred to as a CPU (Central Processing Unit). The processor 210 may be an integrated circuit chip having the ability to process signals. The processor 210 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. A general-purpose processor may be a microprocessor or the processor 210 may also be any conventional processor, etc.

[0099] The memory 220 is used to store computer programs, and may be a RAM, a ROM, or other types of storage terminals. Specifically, the memory 220 may include one or more computer-readable storage media, which may be non-transitory or transient. The memory 220 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage terminals, flash memory storage terminals. In some embodiments, the non-transitory computer-readable storage medium in the memory 220 is used to store at least one program code.

[0100] The processor 210 is used to execute the computer program stored in the memory 220 to implement the methods described in the various method embodiments of the present application.

[0101] In some embodiments, the electronic device may further include: a peripheral terminal interface 230 and at least one peripheral terminal. The processor 210, the memory 220 and the peripheral terminal interface 230 may be connected via a bus or a signal line. Each peripheral terminal may be connected to the peripheral terminal interface 230 via a bus, a signal line or a circuit board. Specifically, the peripheral terminal includes: at least one of a radio frequency circuit 240, a display screen 250, an audio circuit 260 and a power supply 270.

[0102] The peripheral terminal interface 230 may be used to connect at least one peripheral terminal related to I / O (Input / output) to the processor 210 and the memory 220. In some embodiments, the processor 210, the memory 220, and the peripheral terminal interface 230 are integrated on the same chip or circuit board; in some other implementations, any one or two of the processor 210, the memory 220, and the peripheral terminal interface 230 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0103] The radio frequency circuit 240 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 240 communicates with the communication network and other IoT devices through electromagnetic signals, and the radio frequency circuit 240 is the communication circuit of the electronic device. The radio frequency circuit 240 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 240 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, an operator identity module card, and the like. The radio frequency circuit 240 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: the World Wide Web, a metropolitan area network, an intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 240 may also include circuits related to NFC (Near Field Communication), which is not limited in this application.

[0104] The display screen 250 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 250 is a touch display screen, the display screen 250 also has the ability to collect touch signals on the surface or above the surface of the display screen 250. The touch signal can be input to the processor 210 as a control signal for processing. At this time, the display screen 250 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 250 can be one, set on the front panel of the electronic device; in other embodiments, the display screen 250 can be at least two, respectively set on different surfaces of the electronic device or in a folding design; in other embodiments, the display screen 250 can be a flexible display screen, set on a curved surface or a folding surface of the electronic device. Even, the display screen 250 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 250 can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode, organic light-emitting diode).

[0105] The audio circuit 260 may include a microphone and a speaker. The microphone is used to collect sound waves from the operator and the environment, and convert the sound waves into electrical signals and input them into the processor 210 for processing, or input them into the radio frequency circuit 240 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged in different parts of the electronic device. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 210 or the radio frequency circuit 240 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 260 may also include a headphone jack.

[0106] The power supply 270 is used to power various components in the electronic device. The power supply 270 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 270 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged through a wired line, and a wireless rechargeable battery is a battery that is charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0107] For a detailed description of the functions and execution processes of each functional module or component in the electronic device embodiment of the present application, reference can be made to the description in the above-mentioned method embodiments of the present application, which will not be repeated here.

[0108] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the various embodiments of the electronic devices described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0109] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0110] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0111] Based on the same inventive concept, the present application also provides a computer-readable storage medium storing a computer program, which can be executed by a processor to implement the above method.

[0112] See also Figure 6 , if the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium 300. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions / computer programs to enable an IoT device (which can be a personal computer, server, or network terminal, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical disks, and electronic terminals such as computers, mobile phones, laptops, tablet computers, cameras, etc. having the above-mentioned storage media.

[0113] The description of the execution process of the program data in the computer-readable storage medium can refer to the description in the above-mentioned method embodiments of the present application, and will not be repeated here.

[0114] It can be seen that the present invention discloses a plan management method, device, equipment and medium based on a mixture of large and small models, the method comprising the following steps: obtaining real-time data of each business field; analyzing the real-time data of each business field based on the first model group of each business field to obtain the data analysis results of each business field; summarizing the data analysis results of each business field to obtain the summary results; based on the second model group, comprehensively analyzing the summary results; determining that the result of the comprehensive analysis generates a target plan, and distributing the target plan to each business field. The method of the present invention can be applied to a variety of enterprise scenarios, including production scheduling plan generation and execution scenarios, sales plan generation and execution, etc., by constructing the first model group of each business field and the second cross-domain model group. Among them, each business field uses the first model group to complete the simple and complex transaction analysis related to the target, as well as the monitoring during the execution of the target plan, and the cross-domain large model in the second model group integrates the data analysis results of the first model group of each business field to generate and adjust the target plan. Using the method of the present invention for plan management makes the internal plan formulation of the enterprise more reasonable and more efficient, and the combination of large and small models is applied to the digital transformation of the enterprise, which has the effect of reducing costs and increasing efficiency relative to the solution of using large models completely.

[0115] The above are only embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent structural transformations made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are included in the protection scope of the present invention.

Claims

1. A planning management method based on a mixture of large and small models, characterized in that: The method comprises the following steps: Obtain real-time data in various business areas; Based on the first model group of each business field, the real-time data of each business field is analyzed to obtain data analysis results of each business field; Summarize the data analysis results of each business field to obtain summary results; Based on the second model group, a comprehensive analysis is performed on the summary results; The results of the comprehensive analysis are determined to generate a target plan, which is distributed to each business area.

2. The planning management method according to claim 1, characterized in that: Before obtaining the real-time data of each business field, the following steps are also included: The historical data of each business field is collected, and the historical data of each business field is used to construct the first model group of each business field and the second model group across fields.

3. The planning management method according to claim 1, characterized in that: The data analysis results of each business field include a first analysis result. The first model group based on each business field analyzes the real-time data of each business field to obtain the data analysis results of each business field, including the steps of: The small models in the first model group of each business field are used to analyze the real-time data of each business field to obtain a first analysis result.

4. The planning management method according to claim 3, characterized in that: The data analysis results of each business field also include a second analysis result, wherein the first model group based on each business field analyzes the real-time data of each business field to obtain the data analysis results of each business field, and further includes the steps of: The large model in the first model group of each business field is used to analyze the real-time data of each business field to obtain a second analysis result.

5. The planning management method according to claim 1, characterized in that: After distributing the target plan to each business area, the following steps are also included: A field plan for each business field is formulated and executed by using the first model group for each business field and the target plan.

6. The planning management method according to claim 5, characterized in that: When executing the field plans of the business fields, monitoring whether the target plans are updated; if the target plans are updated, modifying the field plans of the business fields according to the latest target plans; If the target plan is not updated, continue to monitor whether the target plan is updated.

7. The planning management method according to claim 1, characterized in that: The method further comprises the steps of: Based on the second model group, the received total task is decomposed and arranged to obtain a plurality of subtasks with a set execution order; Based on the expert models in the first model group of each business field, analyzing the real-time data of each business field, and executing the subtasks to obtain the subtask execution results of each business field; Summarize the execution results of the subtasks in each business area to obtain a summary task result; Based on the second model group, comprehensively analyzing the summary task results to obtain a supplementary plan; Determine whether the overall task is completed. If the overall task is not completed, return to the expert model in the first model group based on the business fields, analyze the real-time data of the business fields, and execute the subtasks to obtain the execution results of the subtasks in each business field; if the overall task is completed, distribute the supplementary plan to each business field.

8. A planning management device based on a mixture of large and small models, characterized in that: The device comprises: Data collection unit, used to obtain real-time data in various business areas; A data analysis unit, configured to analyze the real-time data of each business field based on the first model group of each business field to obtain a data analysis result of each business field; An analysis result summary unit, used to summarize the data analysis results of each business field to obtain a summary result; A comprehensive analysis unit, configured to perform a comprehensive analysis on the summary results based on the second model group; The plan distribution unit is used to determine that the result of the comprehensive analysis generates a target plan, and distribute the target plan to each business field.

9. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images