Domestic necessity regulation and control method and equipment based on DeepSeek large model, and medium

Abstract

The embodiment of the invention discloses a living necessity regulation and control method and device based on a DeepSeek large model and a medium, belongs to the technical field of material regulation and control, and solves the problem that an existing living necessity circulation guarantee supply operation system is low in efficiency and reliability. Comprising the following steps: training a DeepSeek large model through historical life necessity regulation and control data, and integrating the trained DeepSeek large model to a life necessity circulation, insurance and supply operation system; inputting the obtained market multi-source data into the trained model, and outputting market demand prediction information of the life necessities; according to the market demand prediction information, generating a regional demand thermodynamic diagram, and determining regulation and control emergency levels corresponding to different regions based on the regional demand thermodynamic diagram; acquiring a corresponding allocation information set based on the regulation and control emergency level; and based on the market demand prediction information and the deployment information set, generating life necessity regulation and control strategies corresponding to different regions.

Description

Technical Field

[0001] This application relates to the field of material control technology, and in particular to a method, equipment and medium for controlling daily necessities based on the DeepSeek large model. Background Technology

[0002] With socio-economic development and the diversification of consumer demand, a stable supply of daily necessities is crucial for people's livelihood and social stability. However, the existing system for ensuring the distribution and supply of daily necessities has many shortcomings in dealing with complex and ever-changing market environments.

[0003] On the one hand, traditional systems have limited data processing capabilities, leading to significant deviations in market demand forecasts for daily necessities. This results in the supply side being unable to respond promptly to changes in consumer demand, causing not only resource waste but also potential shortages in some areas. On the other hand, in the emergency supply and demand matching process, existing resource allocation methods struggle to monitor real-time inventory, transportation capacity, and regional demand differences. Consequently, problems such as allocation delays and unreasonable route planning frequently occur during resource allocation, leading to a paradoxical situation of inventory backlog and untimely supply. Consequently, the efficiency and reliability of existing systems for ensuring the circulation and supply of daily necessities are relatively low. Summary of the Invention

[0004] This application provides a method, device, and medium for regulating daily necessities based on the DeepSeek large model, which is used to solve the following technical problem: Existing methods for regulating daily necessities have low efficiency and reliability due to large errors in market demand forecasting and lag in material allocation.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides a method for regulating essential consumer goods based on a DeepSeek big data model. The method includes: training a DeepSeek big data model using historical essential consumer goods regulation data; integrating the trained DeepSeek big data model into an essential consumer goods distribution and supply guarantee system; inputting acquired multi-source market data into the trained DeepSeek big data model to output market demand forecast information for essential consumer goods; generating regional demand heat maps based on the market demand forecast information; determining the corresponding emergency response levels for different regions based on the regional demand heat maps; obtaining corresponding allocation information sets based on the emergency response levels for different regions; wherein different emergency response levels correspond to different dimensions of allocation information; and generating essential consumer goods regulation strategies for different regions based on the market demand forecast information and the allocation information sets.

[0007] This application's embodiments utilize the DeepSeek large-scale model to process multi-source heterogeneous data, obtaining consumer behavior patterns and market trends. Trained with historical control data, the model can adapt to changes in the market environment, continuously optimizing the prediction algorithm and reducing prediction biases compared to traditional methods. Furthermore, by visually displaying risk distribution through regional demand heatmaps, it supports decision-makers in proactively deploying prevention and control measures. Based on regional demand heatmaps and emergency levels, the system automatically matches corresponding allocation information, integrating multi-dimensional data such as logistics, inventory, consumption, and public opinion to improve prediction and scheduling accuracy, enabling dynamic cross-regional allocation of resources and alleviating local supply-demand imbalances.

[0008] In one implementation of this application, a DeepSeek model is trained using historical essential goods regulation data, and the trained DeepSeek model is integrated into an essential goods distribution and supply guarantee system. Specifically, this includes: obtaining historical essential goods regulation data through the essential goods distribution and supply guarantee system; wherein the historical essential goods regulation data includes at least historical sales data, consumer behavior data, market price fluctuation data, and supplier information; preprocessing the historical essential goods regulation data and using the preprocessed data as training samples to train the DeepSeek model; wherein the data preprocessing includes at least data cleaning, data organization, and data labeling; and integrating the trained DeepSeek model with the essential goods distribution and supply guarantee system.

[0009] In one implementation of this application, the acquired multi-source market data is input into a trained DeepSeek model to output market demand forecast information for daily necessities. Specifically, this includes: inputting multi-source market data into a trained DeepSeek model; wherein the multi-source market data includes at least historical consumption data, seasonal fluctuation data, and public opinion information; outputting market demand forecast information for daily necessities corresponding to different regions through the trained DeepSeek model; and constructing a time series set of market demand forecasts for daily necessities based on the pre-set market demand forecast information for daily necessities corresponding to different future consecutive time periods.

[0010] In one implementation of this application, a regional demand heatmap is generated based on market demand forecast information. The corresponding emergency response levels for different regions are then determined based on the regional demand heatmap. Specifically, this includes: mapping and associating market demand forecast information with geographical regions to construct a spatial dataset; assigning weights to pre-defined multiple indicators, and weighting and fusing the weights with the multi-indicator data to obtain a multi-indicator fusion emergency index; determining a color mapping strategy based on the fusion emergency index; constructing a regional demand heatmap based on the time dimension and the color mapping strategy corresponding to each geographical region; and determining the corresponding emergency response level for each geographical region based on the color changes in the regional demand heatmap.

[0011] In one implementation of this application, based on the corresponding emergency response levels for different regions, a corresponding allocation information set is obtained. Specifically, this includes: determining the required allocation information from a pre-set allocation information table based on the emergency response level; acquiring data based on the required allocation information to construct the allocation information set; wherein, when the emergency response level is Level 1, the allocation information set includes at least inventory data, transportation capacity data, production capacity data, personnel allocation data, regional cooperation data, and financial data; when the emergency response level is Level 2, the allocation information set includes at least inventory data, transportation capacity data, personnel allocation data, and regional cooperation data; and when the emergency response level is Level 3, the allocation information set includes at least inventory data, community demand data, and price data.

[0012] In one implementation of this application, based on market demand forecast information and allocation information set, a regulation strategy for daily necessities corresponding to different regions is generated. Specifically, this includes: outputting an emergency material allocation plan based on the allocation information set and a trained DeepSeek model; determining multiple material transportation routes based on the emergency material allocation plan; obtaining traffic information corresponding to each of the multiple material transportation routes, and determining the transportation time corresponding to each of the multiple material transportation routes through the traffic information; taking the material transportation route with the shortest transportation time as the target route; and obtaining the regulation strategy for daily necessities based on the emergency material allocation plan and the target route.

[0013] In one implementation of this application, after generating the daily necessities regulation strategies corresponding to different regions, the method further includes: identifying the demand questions sent by users based on the trained DeepSeek large model, and responding to the demand questions; generating improvement suggestions based on the demand questions and user feedback information, and pushing the improvement suggestions to the supply chain management module.

[0014] In one implementation of this application, generating control strategies for essential goods in different regions specifically includes: when using unmanned delivery equipment for material transportation, inputting market demand forecast information, allocation information set, and unmanned delivery equipment information into the DeepSeek large model; outputting unmanned transportation strategies corresponding to the unmanned delivery equipment through the DeepSeek large model; wherein, the unmanned transportation strategies include load allocation and route planning; and dynamically adjusting the unmanned transportation strategies based on real-time weather information.

[0015] This application provides a device for regulating essential consumer goods based on a DeepSeek big data model, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to: train the DeepSeek big data model using historical essential consumer goods regulation data, and integrate the trained DeepSeek big data model into an essential consumer goods distribution and supply guarantee system; input acquired multi-source market data into the trained DeepSeek big data model to output market demand forecast information for essential consumer goods; generate regional demand heat maps based on the market demand forecast information, and determine the regulation and emergency response levels corresponding to different regions based on the regional demand heat maps; obtain corresponding allocation information sets based on the regulation and emergency response levels corresponding to different regions; wherein different regulation and emergency response levels correspond to allocation information of different dimensions; and generate essential consumer goods regulation strategies corresponding to different regions based on the market demand forecast information and the allocation information sets.

[0016] This application provides a non-volatile computer storage medium storing computer-executable instructions. These instructions are configured to: train a DeepSeek model using historical data on essential consumer goods regulation, and integrate the trained DeepSeek model into an essential consumer goods distribution and supply guarantee system; input acquired multi-source market data into the trained DeepSeek model to output market demand forecast information for essential consumer goods; generate regional demand heat maps based on the market demand forecast information, and determine the corresponding regulatory emergency levels for different regions based on the regional demand heat maps; obtain corresponding allocation information sets based on the regulatory emergency levels for different regions; wherein different regulatory emergency levels correspond to different dimensions of allocation information; and generate essential consumer goods regulation strategies for different regions based on the market demand forecast information and the allocation information sets.

[0017] The at least one technical solution adopted in this application embodiment can achieve the following beneficial effects: This application embodiment, through the DeepSeek large model, can process multi-source heterogeneous data and determine consumer behavior patterns and market change trends. Through training with historical control data, the model can adapt to changes in the market environment, continuously optimize the prediction algorithm, and reduce prediction biases of traditional methods. Furthermore, by intuitively displaying risk distribution through regional demand heat maps, it supports decision-makers in deploying prevention and control measures in advance. Based on regional demand heat maps and emergency levels, the system automatically matches allocation information of corresponding dimensions, integrates multi-dimensional data such as logistics, inventory, consumption, and public opinion, improves prediction and scheduling accuracy, realizes dynamic cross-regional scheduling of materials, and alleviates local supply and demand imbalances. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0019] Figure 1 A flowchart of a method for regulating daily necessities based on a DeepSeek large model, provided in this application embodiment;

[0020] Figure 2 This is a schematic diagram of a daily necessities control device based on a DeepSeek large model, provided as an embodiment of this application.

[0021] Figure label:

[0022] 200: Essential goods control device based on DeepSeek large model; 201: Processor; 202: Memory. Detailed Implementation

[0023] This application provides a method, device, and medium for regulating daily necessities based on a DeepSeek large model.

[0024] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0025] The technical solutions proposed in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0026] Figure 1 A flowchart of a method for regulating daily necessities based on a DeepSeek large model, as provided in this application embodiment, is shown below. Figure 1 As shown, the method for regulating daily necessities based on the DeepSeek large model includes the following steps:

[0027] S101. Train the DeepSeek big model using historical data on the regulation of daily necessities, and integrate the trained DeepSeek big model into the daily necessities circulation and supply guarantee system.

[0028] In one implementation of this application, historical essential goods regulation data is obtained through an essential goods distribution and supply guarantee system. This historical essential goods regulation data includes at least historical sales data, consumer behavior data, market price fluctuation data, and supplier information. The historical essential goods regulation data is preprocessed, and the preprocessed data is used as training samples to train the DeepSeek large-scale model. The data preprocessing includes at least data cleaning, data organization, and data labeling. The trained DeepSeek large-scale model is then integrated with the essential goods distribution and supply guarantee system.

[0029] Specifically, the essential goods distribution and supply guarantee system collects data through various channels, and cleans, organizes, and labels the data to meet the input requirements of the DeepSeek large-scale model. The preprocessed data is then input into the DeepSeek large-scale model for training, continuously adjusting model parameters and optimizing model performance to accurately predict market demand and provide decision-making suggestions. The trained DeepSeek large-scale model is then integrated with the essential goods distribution and supply guarantee system to achieve real-time data interaction and automated business processing.

[0030] S102. Input the acquired multi-source market data into the trained DeepSeek model to output market demand forecast information for daily necessities.

[0031] In one implementation of this application, multi-source market data is input into a trained DeepSeek model; this multi-source market data includes at least historical consumption data, seasonal fluctuation data, and public opinion information. The trained DeepSeek model outputs market demand forecasts for essential goods in different regions. Based on these pre-set market demand forecasts for essential goods at consecutive future times, a time-series set of market demand forecasts for essential goods is constructed.

[0032] Specifically, the DeepSeek large-scale model is trained with multi-source market data, including historical consumption data, seasonal fluctuation data, and public opinion information. The model performs in-depth data mining and feature extraction on this data to output market demand forecasts for essential consumer goods in different regions. Furthermore, based on pre-defined future timeframes, such as days, weeks, and months, the market demand forecasts for each time point are systematically integrated to construct a time-series set of market demand forecasts for essential consumer goods. This time-series set can comprehensively present the dynamic trends of market demand for essential consumer goods in different regions over a future period.

[0033] S103. Based on market demand forecast information, generate regional demand heat maps, and determine the corresponding emergency control levels for different regions based on the regional demand heat maps.

[0034] In one implementation of this application, market demand forecast information is mapped and associated with geographical regions to construct a spatial dataset. Pre-defined multiple indicators are weighted and then weighted and fused with the indicator data to obtain a multi-indicator fusion emergency index. A color mapping strategy is determined based on the fusion emergency index. A regional demand heatmap is constructed based on the time dimension and the color mapping strategy corresponding to each geographical region. The emergency response level for each geographical region is determined based on the color changes in the regional demand heatmap.

[0035] Specifically, market demand forecasts for daily necessities, such as demand volume and supply-demand gaps for each category, are linked to specific geographical areas, such as cities, districts, and streets, giving the demand data geospatial attributes. Through mapping and association, the originally scattered demand data is integrated into a structured dataset with location information.

[0036] Furthermore, based on market demand forecasts, several key indicators are selected, such as the supply-demand gap ratio, price volatility, and inventory turnover days. Historical experience is used to assign corresponding weights to each indicator to reflect its importance in assessing regional demand risk. Through weighted summation, the data from each indicator and its weights are integrated to calculate a multi-indicator fusion emergency index that comprehensively reflects the degree of regional demand risk.

[0037] Furthermore, this application embodiment pre-designs a color mapping rule based on the numerical range of the multi-indicator fusion emergency index. The emergency index is divided into several intervals, each interval corresponding to a color; for example, low-risk intervals correspond to green, medium-risk intervals to yellow, and high-risk intervals to red. Through this color mapping strategy, risk values ​​are intuitively converted into visual colors, facilitating users to quickly identify the differences in risk levels across different areas.

[0038] Furthermore, the emergency response index of each geographical region at each time point is visualized using a color mapping strategy, arranged chronologically. The color-coded geographical regions are then overlaid on a map to create a regional demand heatmap, which reflects the demand risk status of different regions at different times.

[0039] Furthermore, by using the colors of each geographical region in the regional demand heatmap, the degree of demand risk in that region can be determined, and the corresponding emergency response level can be established accordingly. For example, areas shown in red correspond to extremely high risk and require the activation of the highest level of emergency response; areas shown in green correspond to low risk and can be addressed with conventional control measures.

[0040] S104. Based on the corresponding emergency response levels for different regions, obtain the corresponding allocation information sets.

[0041] In one implementation of this application, an emergency supplies dispatch engine is built based on the DeepSeek large-scale model. This engine dynamically generates optimal delivery routes and emergency supplies allocation plans by combining real-time traffic data, inventory distribution, and demand forecasts. At emergency supplies transfer stations, the large-scale model optimizes sorting priorities and transfer routes, reducing response time by more than 30%. When a sudden disaster is detected, the system automatically triggers an emergency mode, coordinating with last-mile delivery points to quickly allocate supplies and notifying residents to collect them via an intelligent outbound calling system.

[0042] In one implementation of this application, a precise sales forecasting model is generated by analyzing historical consumption data, seasonal fluctuations, and public opinion information using the DeepSeek large-scale model, and inventory strategies are dynamically adjusted. In the reserve of essential consumer goods, the model is used to predict demand for the next 7 days, optimizing reserve size and replenishment cycles, reducing inventory backlog risk by 20%. Real-time analysis of sales data from thousands of retail outlets allows for dynamic adjustments to promotional strategies and replenishment plans.

[0043] In one implementation of this application, based on the emergency response level, the corresponding required allocation information is determined from a pre-set allocation information table. Data is acquired based on the required allocation information to construct the allocation information set. Specifically, when the emergency response level is Level 1, the allocation information set includes at least inventory data, transportation capacity data, production capacity data, personnel allocation data, regional cooperation data, and financial data. When the emergency response level is Level 2, the allocation information set includes at least inventory data, transportation capacity data, personnel allocation data, and regional cooperation data. When the emergency response level is Level 3, the allocation information set includes at least inventory data, community demand data, and price data. Different emergency response levels correspond to different dimensions of allocation information.

[0044] Specifically, in this embodiment, a pre-set allocation information table is provided, which records the allocation information categories corresponding to different emergency response levels. Once the emergency response level for a region is determined, this level is used as an index to perform a matching query in the pre-set allocation information table, thereby obtaining the required allocation information categories for that level. Based on these categories, targeted data collection is performed from multiple data sources, including internal databases and external data interfaces.

[0045] Furthermore, a Level 1 emergency response level indicates that the region faces extremely high demand risks, requiring comprehensive and intensive resource allocation. A Level 2 emergency response level carries a slightly lower risk than Level 1, but still necessitates coordinated resource allocation across multiple sectors. A Level 3 emergency response level indicates that the region is in a low-to-medium risk state, and resource allocation efforts focus more on refined management.

[0046] S105. Based on market demand forecast information and allocation information set, generate corresponding regulation strategies for daily necessities in different regions.

[0047] In one implementation of this application, an emergency supplies allocation plan is output based on the allocation information set and a trained DeepSeek model. Based on the emergency supplies allocation plan, multiple supply transportation routes are determined. Traffic information corresponding to each of the multiple supply transportation routes is obtained, and the transportation time corresponding to each of the multiple supply transportation routes is determined using the traffic information. The supply transportation route with the shortest transportation time is selected as the target route. Based on the emergency supplies allocation plan and the target route, a strategy for regulating essential goods is derived.

[0048] Specifically, the constructed allocation information set, including multi-dimensional data such as inventory, transportation capacity, and regional demand, is input into the trained DeepSeek model. This model outputs corresponding emergency material allocation plans, clarifying the types and quantities of materials to be allocated to different regions. Based on the material allocation situation in each region within the emergency material allocation plan, multiple potential material transportation routes are generated, starting from the material reserve location and ending at the demand region.

[0049] Furthermore, for each determined material transportation route, its corresponding traffic information is acquired in real time, including real-time road conditions, road construction status, and traffic control information. This traffic information is combined with the route's basic parameters, such as distance and speed limits, to calculate the estimated transportation time required for each material transportation route under current traffic conditions. The calculated transportation times of multiple material transportation routes are compared, and the route with the shortest transportation time is selected.

[0050] Furthermore, by combining the emergency supplies distribution plan with target routes, a strategy for regulating daily necessities is formed. This strategy clarifies the distribution destination, quantity, and transportation routes of supplies, ensuring that supplies are delivered accurately and efficiently to the areas most in need, while also optimizing the transportation process to ensure that the demand for daily necessities in various regions is met in the shortest possible time, thus achieving effective regulation of the supply of daily necessities.

[0051] In one implementation of this application, the trained DeepSeek model is used to identify user-submitted requests and respond to them. Based on the requests and user feedback, improvement suggestions are generated and pushed to the supply chain management module.

[0052] Specifically, this application integrates the natural language processing capabilities of the DeepSeek large model to build an intelligent customer service and satisfaction survey system, enabling automatic identification of customer needs and rapid response to problems. The system automatically calls back users through a telemarketing outbound call system to collect feedback on delivery timeliness, product quality, etc., generating improvement suggestions and pushing them to the supply chain management module.

[0053] In one implementation of this application, when using unmanned delivery equipment for material transportation, market demand forecast information, allocation information set, and unmanned delivery equipment information are input into the DeepSeek large model. The DeepSeek large model then outputs an unmanned transportation strategy corresponding to the unmanned delivery equipment; this strategy includes load allocation and route planning. Real-time weather information is acquired, and the unmanned transportation strategy is dynamically adjusted based on this information.

[0054] Specifically, DeepSeek's large-scale models are used to optimize the path planning and load allocation of unmanned delivery equipment, such as autonomous vehicles and drones, and edge computing is combined to achieve low-latency decision-making. The large-scale models are used to schedule mobile cold storage and self-service shelves in real time, reducing human intervention and improving delivery efficiency by 40%. Geographic information systems and weather data are combined to dynamically adjust the operating strategies of unmanned equipment to cope with the impact of extreme weather.

[0055] In one implementation of this application, a knowledge graph-based data platform for supply assurance is constructed, and the DeepSeek large-scale model is used to clean, correlate, and visualize multi-source heterogeneous data. Within the supply assurance management and scheduling platform, the large-scale model generates a "supply-demand balance index" and a "regional risk heat map" to assist the government in dynamically adjusting its reserve strategies. Effectiveness evaluations are conducted on supply assurance capacity projects for wholesale enterprises, generating a return on investment forecast report.

[0056] Figure 2 This is a schematic diagram of a daily necessities control device based on a DeepSeek large model, provided as an embodiment of this application. Figure 2As shown, the essential goods regulation device 200 based on the DeepSeek big data model includes: at least one processor 201; and a memory 202 communicatively connected to the at least one processor 201. The memory 202 stores instructions executable by the at least one processor 201, which, when executed, enable the at least one processor 201 to: train the DeepSeek big data model using historical essential goods regulation data and integrate the trained DeepSeek big data model into the essential goods circulation and supply guarantee system; input acquired multi-source market data into the trained DeepSeek big data model to output market demand forecast information for essential goods; generate regional demand heat maps based on the market demand forecast information and determine the corresponding regulation emergency levels for different regions based on the regional demand heat maps; obtain corresponding allocation information sets based on the regulation emergency levels for different regions; wherein different regulation emergency levels correspond to different dimensions of allocation information; and generate essential goods regulation strategies for different regions based on the market demand forecast information and the allocation information sets.

[0057] This application provides a non-volatile computer storage medium storing computer-executable instructions. These instructions are configured to: train a DeepSeek model using historical data on essential consumer goods regulation, and integrate the trained DeepSeek model into an essential consumer goods distribution and supply guarantee system; input acquired multi-source market data into the trained DeepSeek model to output market demand forecast information for essential consumer goods; generate regional demand heat maps based on the market demand forecast information, and determine the corresponding regulatory emergency levels for different regions based on the regional demand heat maps; obtain corresponding allocation information sets based on the regulatory emergency levels for different regions; wherein different regulatory emergency levels correspond to different dimensions of allocation information; and generate essential consumer goods regulation strategies for different regions based on the market demand forecast information and the allocation information sets.

[0058] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0059] The above descriptions are merely embodiments of this application and are not intended to limit the scope of this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions in the embodiments of this application.

Claims

1. A method for regulating daily necessities based on a DeepSeek large model, characterized in that, The method includes: The DeepSeek big data model was trained using historical data on the regulation of daily necessities, and the trained DeepSeek big data model was integrated into the daily necessities circulation and supply guarantee system. The acquired multi-source market data is input into the trained DeepSeek model to output market demand forecast information for daily necessities. Based on the market demand forecast information, a regional demand heat map is generated, and the corresponding emergency control level for different regions is determined based on the regional demand heat map. Based on the control and emergency response levels corresponding to different regions, a corresponding allocation information set is obtained; wherein, different control and emergency response levels correspond to allocation information of different dimensions; Based on the market demand forecast information and the allocation information set, the corresponding daily necessities regulation strategies for different regions are generated.

2. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, The process of training the DeepSeek model using historical data on the regulation of daily necessities and integrating the trained DeepSeek model into the daily necessities circulation and supply guarantee system specifically includes: The historical data on the regulation and control of essential consumer goods is obtained through the system for ensuring the supply and distribution of essential consumer goods; wherein, the historical data on the regulation and control of essential consumer goods includes at least historical sales data, consumer behavior data, market price fluctuation data, and supplier information; The historical data on the regulation of daily necessities is preprocessed, and the preprocessed data is used as training samples to train the DeepSeek large model; wherein, the data preprocessing includes at least data cleaning, data organization, and data labeling. The trained DeepSeek large model is integrated with the essential goods distribution and supply guarantee system.

3. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, The process of inputting the acquired multi-source market data into the trained DeepSeek model to output market demand forecast information for daily necessities specifically includes: The multi-source market data is input into the trained DeepSeek model; wherein, the multi-source market data includes at least historical consumption data, seasonal fluctuation data, and public opinion information; The trained DeepSeek model outputs market demand forecasts for daily necessities in different regions. Based on the market demand forecast information of the daily necessities corresponding to the pre-set future continuous time periods, a time series set of market demand forecasts for daily necessities is constructed.

4. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, The step of generating a regional demand heat map based on the market demand forecast information, and determining the corresponding emergency response level for different regions based on the regional demand heat map, specifically includes: By mapping and associating market demand forecast information with geographical regions, a spatial dataset is constructed. The pre-set multiple indicators are weighted and then the weights are weighted and fused with the indicator data to obtain the multi-indicator fusion emergency index. Based on the multi-indicator fusion emergency index, a color mapping strategy is determined; Based on the time dimension and the color mapping strategy corresponding to each geographical region, a heat map of regional demand is constructed. Based on the color changes in the heatmap of regional demand, the corresponding emergency response level for each geographical region is determined.

5. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, The acquisition of corresponding allocation information sets based on the different emergency response levels for different regions specifically includes: Based on the aforementioned emergency response level, the corresponding required allocation information is determined from the pre-set allocation information table; Data is acquired based on the required allocation information to construct the allocation information set; When the emergency response level is Level 1, the allocation information set includes at least inventory data, transportation capacity data, production capacity data, personnel allocation data, regional cooperation data, and financial data. When the emergency response level is Level II, the allocation information set includes at least inventory data, transportation capacity data, personnel allocation data, and regional cooperation data. When the emergency response level is Level III, the allocation information set includes at least inventory data, community demand data, and price data.

6. The method for regulating daily necessities based on the DeepSeek large model according to claim 5, characterized in that, The process of generating essential goods regulation strategies for different regions based on the market demand forecast information and the allocation information set specifically includes: Based on the aforementioned allocation information set and the trained DeepSeek model, an emergency supplies allocation plan is output. Based on the aforementioned emergency supplies distribution plan, multiple supply transportation routes were determined; Obtain traffic information corresponding to multiple material transportation routes, and determine the transportation time corresponding to each of the multiple material transportation routes based on the traffic information; The material transportation route corresponding to the minimum transportation time is taken as the target route; Based on the emergency supplies allocation plan and the target route, the essential goods control strategy is obtained.

7. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, After generating the daily necessities regulation strategies for the different regions, the method further includes: Based on the trained DeepSeek model, the system identifies the user's requests and responds to them. Based on the aforementioned needs and user feedback, improvement suggestions are generated and pushed to the supply chain management module.

8. The method for regulating daily necessities based on the DeepSeek large model according to claim 1, characterized in that, The generation of the daily necessities regulation strategies for the different regions specifically includes: When using unmanned delivery equipment for material transportation, the market demand forecast information, the allocation information set, and the unmanned delivery equipment information are input into the DeepSeek large model; The DeepSeek large model is used to output the unmanned transportation strategy corresponding to the unmanned delivery equipment; wherein, the unmanned transportation strategy includes load allocation and path planning. The system acquires real-time weather information and dynamically adjusts the unmanned transportation strategy based on this information.

9. A device for regulating daily necessities based on a DeepSeek large-scale model, characterized in that, The device includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to perform the method described in any one of claims 1-8.

10. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are capable of performing the method described in any one of claims 1-8.

Images