Enterprise intelligent management system and method based on big data

Abstract

The invention discloses an enterprise intelligent management method based on big data, and the method comprises the steps: carrying out the region division of each street, and collecting the data monitored by a camera in each region and the activity data of an industrial enterprise; s2, collecting and sorting the activity level of the camera data directly facing the industrial enterprise obtained in the step S1, and analyzing the activity level of the camera data directly facing the industrial enterprise; s2, establishing a model according to the data sorted in the S2, and performing training optimization so as to simulate the activity level of the industrial enterprise; adjusting the activity state of the industrial enterprise in a period of time in the future through the optimized model; camera monitoring data and industrial enterprise activity data are integrated, in-depth correlation analysis is carried out, a prediction model for simulating the industrial enterprise activity level is established, the response speed of an enterprise to market changes and the perspectiveness of decision making are greatly improved, the enterprise can more accurately predict future market demands, inventory management is optimized, resource waste is reduced, and the economic benefits of the enterprise are improved. The production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to a field related to big data truck traffic analysis, and in particular to an enterprise intelligent management system and method based on big data. Background Art

[0002] In the digital age, the importance of big data enterprise intelligent management is becoming increasingly prominent. Big data enterprise intelligent management can not only help enterprises achieve deep data mining and intelligent analysis, but also help enterprises intuitively understand the operating status through real-time monitoring and intelligent analysis, discover the laws and trends behind the data, and make scientific decisions. This data-based decision-making model enables enterprises to respond to market changes more quickly and improve the foresight and response speed of decision-making.

[0003] However, existing technical solutions often fail to achieve advance adjustments to production plans and logistics arrangements, nor can they optimize inventory management. These problems usually stem from insufficient data collection and analysis, as well as the lack of effective forecasting models to guide decision-making. When faced with the challenges of large, diverse, and complex data volumes, traditional management methods often find it difficult to effectively extract valuable information from massive amounts of data, resulting in the inability to accurately predict market demand, and thus the inability to adjust production and logistics strategies in a timely manner and optimize inventory levels.

[0004] Therefore, it is necessary to improve the intelligent management of big data enterprises in existing technologies. Summary of the invention

[0005] The present invention overcomes the deficiencies of the prior art and provides an enterprise intelligence management system and method based on big data.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a method for enterprise intelligent management based on big data, comprising the following steps:

[0007] S1: Divide each street into regions and collect data monitored by cameras and industrial enterprise activities in each region;

[0008] S2: The camera data obtained in S1 is collecting, collating and analyzing the activity level of industrial enterprises;

[0009] S3: Building a model based on the data collated in S2 and performing training optimization to simulate the activity level of industrial enterprises;

[0010] S4: Adjust the activity status of industrial enterprises in the future by optimizing the completed model.

[0011] In a preferred embodiment of the present invention, in S1, the division is specifically to divide the street into blocks, so as to ensure that the camera can capture key areas and key activities.

[0012] In a preferred embodiment of the present invention, key areas include traffic nodes, public places, industrial enterprises and other areas where truck operation can be collected; key activities include production activities, material procurement activities, material supply activities and other activities involving truck operation.

[0013] In a preferred embodiment of the present invention, in S1, the industrial enterprise activity data is specifically data that can reflect the status of the industrial enterprise's logistics activities such as raw material procurement and product sales.

[0014] In a preferred embodiment of the present invention, in S2, the collection and organization specifically involves arranging the camera data and the enterprise activity level data in chronological order into a collection, and excluding irrelevant data, and then making a one-to-one correspondence.

[0015] In a preferred embodiment of the present invention, the irrelevant data specifically includes the determination and exclusion of passing vehicles and the exclusion of vehicles with special license plates.

[0016] In a preferred embodiment of the present invention, in S2, the analysis is specifically to perform a correlation analysis on the truck flow obtained from the camera data and the industrial enterprise activity level data, so as to analyze the intrinsic relationship between the two.

[0017] In a preferred embodiment of the present invention, in S3, establishing a model specifically involves establishing a suitable model based on the characteristic relationship between vehicle flow and the activity level of industrial enterprises obtained by the analysis in S2.

[0018] In a preferred embodiment of the present invention, in S3, the training optimization specifically includes returning the data sorted in S2 to the established model again, training and optimizing the model and testing the model, and finally obtaining an optimized prediction model.

[0019] In a preferred embodiment of the present invention, in S4, adjusting the future activity state of the industrial enterprise specifically involves intervening and adjusting the activities of the industrial enterprise according to the predicted value output by the model.

[0020] The present invention provides an enterprise intelligence management system based on big data, comprising:

[0021] Area division module: used to divide streets into blocks to ensure that cameras can cover key areas and activities, including traffic nodes, public places and industrial enterprises;

[0022] Data collection unit: used to collect data monitored by cameras in each area and data on industrial enterprise activities, including data on logistics activities such as raw material procurement and product sales;

[0023] Data sorting and cleaning unit: used to sort the camera data and enterprise activity level data in chronological order and exclude irrelevant data such as passing vehicles and vehicles with special license plates;

[0024] Correlation analysis engine: used to analyze the correlation between truck traffic obtained from camera data and industrial enterprise activity level data, and calculate the correlation coefficient to measure the degree of correlation between the two;

[0025] Model building and training optimization unit: used to build a regression model, such as a polynomial regression model, based on the analysis results, and perform training optimization to capture the nonlinear relationship between truck flow and the activity level of industrial enterprises;

[0026] Seasonal Factor Analysis Module: used to identify seasonal variables and fit them into the established model to take into account the impact of seasonality on truck traffic and industrial enterprise activity levels;

[0027] Prediction model application unit: used to apply the optimized prediction model and adjust the activity status of the industrial enterprise in the future according to the prediction value output by the model;

[0028] Decision management system: used to provide decision support for production planning and logistics arrangements based on model predictions, helping enterprises to adjust production plans and logistics strategies in advance and optimize inventory management.

[0029] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0030] (1) The present invention provides an enterprise intelligent management method based on big data. Through regional division and data collection, it integrates camera monitoring data and industrial enterprise activity data, uses these data to conduct in-depth correlation analysis, and establishes a prediction model that can simulate the activity level of industrial enterprises. The advantage of this method is that it greatly improves the enterprise's response speed to market changes and the foresight of decision-making; by real-time monitoring and analysis of the relationship between truck traffic and the activity level of industrial enterprises, enterprises can more accurately predict future market demand, thereby adjusting production plans and logistics arrangements in advance, optimizing inventory management, reducing resource waste, and improving production efficiency.

[0031] (2) The present invention uses big data to monitor the information data of trucks, and conducts exploratory data analysis to construct the daily change rate of truck traffic and time series analysis, thereby constructing a model of the relationship between the activity level of industrial enterprises and truck traffic that contains seasonal effects. This model can more comprehensively and accurately reflect the changing trend of the activity level of industrial enterprises, and through the application of data mining and prediction models. In addition, by optimizing the model considering seasonal factors, enterprises can respond to seasonal market fluctuations more flexibly and achieve more refined market management. This method also enhances the risk management capabilities of enterprises. By predicting potential market risks and operational problems, enterprises can formulate response strategies in a timely manner to reduce the impact of uncertainty. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 It is a three-dimensional structural diagram of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0036] Application Overview:

[0037] Truck flow analysis based on big data is an analysis method that uses massive truck operation data to reveal truck traffic flow patterns, trends, and influencing factors. This method builds a comprehensive and real-time truck flow monitoring system by collecting and integrating multiple data such as truck positioning, driving speed, and cargo capacity. First, truck flow data is collected through cameras, sensors, GPS devices, and vehicle networking technology to ensure the real-time and accuracy of the data. These data are stored in a high-performance big data processing platform, and distributed computing and storage technologies are used to achieve fast processing and query.

[0038] Subsequently, data analysis tools and methods are used to mine and analyze truck traffic data. For example, through time series analysis, the trend of truck traffic in the future can be predicted; through spatial analysis, the geographical distribution characteristics of truck traffic can be revealed; through association analysis, the relationship between truck traffic and other factors (such as weather and holidays) can be explored. Based on the analysis results, the distribution, trend and influencing factors of truck traffic can be intuitively displayed through visualization technology, providing decision support for traffic management departments, logistics companies and decision-making departments.

[0039] Finally, truck traffic data is not only a reflection of road traffic conditions, but also an indirect reflection of the activity level of industrial enterprises. By analyzing the activity level of industrial enterprises through truck traffic, we can predict the changing trends of enterprises' production, logistics and other activities, and provide forward-looking decision-making basis for enterprises. At the same time, this analysis also has a guiding effect, which can help enterprises optimize production layout, adjust logistics strategies, and improve operational efficiency.

[0040] Exemplary methods:

[0041] like Figure 1 As shown, a method for enterprise intelligent management based on big data includes the following steps:

[0042] S1: Divide each street into regions and collect data monitored by cameras and industrial enterprise activities in each region;

[0043] S2: The camera data obtained in S1 is collecting, collating and analyzing the activity level of industrial enterprises;

[0044] S3: Building a model based on the data collated in S2 and performing training optimization to simulate the activity level of industrial enterprises;

[0045] S4: Adjust the activity status of industrial enterprises in the future by optimizing the completed model.

[0046] Divide each street into zones and collect data from camera surveillance and industrial enterprise activities in each zone;

[0047] In a preferred embodiment of the present invention, in S1, the division is specifically to divide the street into blocks, so as to ensure that the camera can capture key areas and key activities.

[0048] In a preferred embodiment of the present invention, key areas include traffic nodes, public places, industrial enterprises and other areas where truck operation can be collected; key activities include production activities, material procurement activities, material supply activities and other activities involving truck operation.

[0049] In a preferred embodiment of the present invention, in S1, the industrial enterprise activity data is specifically data that can reflect the status of the industrial enterprise's logistics activities such as raw material procurement and product sales.

[0050] It should be noted that the division of each street is specifically to divide the streets into blocks, so that the cameras in each area can capture the key activities and key areas in the area. The key activities include production activities, material procurement activities, material supply activities and other activities involving the operation of trucks and vehicles. The key areas include production line areas, warehousing areas and other areas involving the operation of trucks and vehicles.

[0051] Regarding the data collected by the cameras in each area, specifically, based on the results of the area division, the cameras that can collect effective information are determined, mainly cameras that can cover important transportation nodes, public places, industrial enterprises and other key areas and can collect the operation of trucks. In this way, real-time information on traffic flow, personnel flow, industrial enterprise activities, etc. in the area can be obtained to provide support for subsequent data analysis.

[0052] The specific method of collecting industrial enterprise information is to obtain data through the enterprise's information system or database, so as to obtain the enterprise's production data, sales data, financial data, employee data and other data, so as to comprehensively reflect the enterprise's operating conditions, economic benefits and market demand related information, so as to further understand the industrial enterprise's raw material procurement, product sales and other logistics activities.

[0053] The camera data obtained in S1 is collecting, collating and analyzing the activity levels of industrial enterprises;

[0054] In a preferred embodiment of the present invention, in S2, the collection and organization specifically involves arranging the camera data and the enterprise activity level data in chronological order into a collection, and excluding irrelevant data, and then making a one-to-one correspondence.

[0055] In a preferred embodiment of the present invention, the irrelevant data specifically includes the determination and exclusion of passing vehicles and the exclusion of vehicles with special license plates.

[0056] In a preferred embodiment of the present invention, in S2, the analysis is specifically to perform a correlation analysis on the truck flow obtained from the camera data and the industrial enterprise activity level data, so as to analyze the intrinsic relationship between the two.

[0057] It should be noted that the information collected in S1 is sorted, different information is recorded as different sets, and several cameras are recorded as p1, p2, p3...p i , the truck information captured by several cameras, where the truck information includes vehicle type, license plate number, capture time t i Etc., thereby forming a snapshot information set C = (c1, c2, c3, ..., c n ), where C i This means that the i-th item is in p i At the same time, the collected operation data of industrial enterprises are recorded as set A, A = (a1, a2, a3...a m ), where a m It represents the activity level of industrial enterprises at the mth time point, such as shipment volume.

[0058] It should be noted that in the process of sorting the captured information set C and the industrial enterprise set A, there is a lot of information irrelevant to the industrial enterprise operation data that needs to be excluded. Specifically: for each vehicle V in the camera captured information set C, calculate the residence time T s The residence time is the departure time minus the entry time, that is, T s =T 离 -T 进 , then set the threshold to 3min, when T s If the time is less than the threshold, it means that the vehicle is a passing vehicle and needs to be excluded. At the same time, special license plate vehicles are set up, and special license plate vehicles also need to be excluded. Special license plate vehicles include but are not limited to police cars, training cars, fire trucks and other vehicles. In this way, when sorting the camera capture information and industrial enterprise operation information, data information or unreasonable information that is irrelevant to the operation of industrial enterprises such as time less than the threshold and special license plate vehicles are excluded.

[0059] In this way, the information collected by the irrelevant or unreasonable industrial enterprises and the information collected by the camera are synchronized in chronological order, that is, the information set A of the activity level collected by the industrial enterprise and the information set C captured by each camera are arranged in chronological order, and then the elements a in the two sets are synchronized. mand element c n A one-to-one correspondence is made according to the closest time, thereby forming a synchronous data set D associated with truck flow and enterprise operation status, where D = (c1a1, c2a2, c3a3, ... c n a m ).

[0060] After the data is organized into related synchronized data sets, for the elements in set C and set A 实 Specifically, in the captured information set C, a time period T is set, and the truck flow F is the number of trucks passing through a camera p within the time period T. i The number of trucks in the industrial enterprise operation data set A contains the activity level data of industrial enterprises at different time points, such as shipment volume, that is, at data point a m represents the activity level A of the industrial enterprise at time point m; then the calculation formula of the correlation r is

[0061] Where r is the correlation coefficient, which is used to measure the degree of correlation between F and A. The range of the correlation coefficient r is [-1,1], where r = 1 indicates a completely positive correlation, r = -1 indicates a completely negative correlation, and r = 0 indicates no linear relationship;

[0062] F i and A i : represents the i-th observed value of the variable F and A in the truck flow and industrial enterprise activity level in the synchronized data set D, respectively. The observed value is the data value collected in practice;

[0063] and Represent the mean values ​​of variables F and A respectively, and are used to calculate the difference between the observed value and the mean value;

[0064] N: The number of observations.

[0065] In the molecule part: This part calculates the sum of the products of the differences between each observation of F and A and their respective means. This is actually calculating the covariance between two variables, which is a measure of how much they vary together.

[0066] Denominator: This part calculates the square root of the sum of the squares of the differences between each observation of F and A and their respective means. This is actually the product of the standard deviations of the two variables and is used to standardize the covariance so that the range of r is between [-1,1].

[0067] The formula for calculating the correlation coefficient r can be understood as the covariance of the two variables divided by the product of their respective standard deviations. This is done to obtain a dimensionless value that varies between -1 and 1, so that the correlations between different pairs of variables can be more easily compared. If r is close to 1, it means that there is a strong positive correlation between truck traffic and industrial enterprise activity levels, that is, when one increases, the other tends to increase. If, is close to -1, it means that there is a strong negative correlation, that is, when one increases, the other tends to decrease. If, is close to 0, it means that there is no obvious linear relationship between the two.

[0068] A model is established based on the data collated in S2, and training and optimization are performed to simulate the activity level of industrial enterprises;

[0069] In a preferred embodiment of the present invention, in S3, the model is established specifically by establishing a suitable model based on the characteristic relationship between the vehicle flow and the activity level of industrial enterprises obtained by the analysis in S2, and the preferred model is a polynomial regression model.

[0070] In a preferred embodiment of the present invention, in S3, the training optimization specifically includes returning the data sorted in S2 to the established model again, training and optimizing the model and testing the model, and finally obtaining an optimized prediction model.

[0071] It should be noted that after obtaining the correlation coefficient between F and A, the correlation coefficients are fitted to construct a polynomial regression model between F and A, while the non-correlation coefficients are excluded, thereby obtaining a polynomial regression model, that is,

[0072] A=β0+β1F+β2F 2 +……+β k F k

[0073] Among them, A is the level of industrial enterprise activity, F is the truck flow,

[0074] β0, β1, β2, ..., β k is the model parameter, i.e., the regression coefficient. The model parameter is obtained by fitting the data and represents the strength and direction of the relationship between the independent variable and the dependent variable. β0 is the intercept term, which represents the expected value of the activity level of industrial enterprises when the truck flow is 0; β1 is the linear term coefficient of the truck flow F, which represents the direct impact of the truck flow on the activity level of industrial enterprises; β2, ..., β k are the coefficients of the polynomial terms of truck flow, which can capture the nonlinear relationship between truck flow and the activity level of industrial enterprises through the model;

[0075] F, F 2 , ..., F kis the polynomial term of the truck flow. By introducing different powers of the truck flow as additional independent variables into the model, a more complex relationship is fitted.

[0076] Among them, the model parameters are obtained by the least squares method, specifically:

[0077] The objective function is constructed by the least squares method to find a set of parameter values ​​that minimizes the sum of square differences between the model prediction value and the actual value (i.e., the residual sum of squares). When the residual sum of squares is minimized, the error between the obtained model and the observed value is minimized, that is, the obtained model is more in line with the actual situation. The objective function is:

[0078]

[0079] Where n is the number of samples in the synchronized dataset D.

[0080] By solving the above formula to minimize the problem, we can find the solution that makes S(β0, β1, ..., β K ) is the smallest parameter value, which is the model parameter value or regression coefficient.

[0081] In this way, the model is obtained, and then the synchronized data set is divided into a training set and a test set. The polynomial model is trained using the training set to obtain the parameters in the multivariate linear regression model.

[0082] in, As parameters, is the intercept term, are the coefficients of each item. After obtaining the parameters, the prediction model is obtained by re-substituting the parameters into the polynomial regression model, that is,

[0083]

[0084] is the estimated level of industrial enterprise activity, F, F 2 ,……,F k The truck flow is used as the basis for constructing a prediction model for the activity level of industrial enterprises through truck flow.

[0085] At the same time, the prediction model is based on the analysis and learning of historical data by the model, and the results of automatic settlement are calculated. The model can use the existing truck flow data to estimate the future truck flow according to its inherent laws and trends. When the model predicts that the future truck flow will increase significantly, it means that according to the analysis of the model, the number of truck flows will be higher than the historical average or current level in a certain period of time in the future. This also means that market demand will rise and companies will face greater order volume or sales pressure. In order to cope with this potential demand growth, companies need to increase production capacity in advance, such as adding production lines and expanding production scale to ensure that they can meet market demand in a timely manner. At the same time, companies may also need to adjust their inventory strategies and increase inventory levels to cope with possible sales peaks and avoid out-of-stock or insufficient supply. In addition, the prediction model also has the impact of seasonality on truck flow that needs to be considered, so seasonal interaction terms need to be added to the polynomial regression model:

[0086] Once again, a seasonal variable is constructed. After adding the seasonal variable to the model, the four seasons (spring, summer, autumn, and winter) are identified, and a dummy variable D is created for each season. s , where s represents the season, and its interaction term with truck flow F is added to the model to construct seasonal model 1:

[0087]

[0088] Alternatively, construct a specific seasonal feature for each season, so that each season has a different polynomial effect, and construct seasonal model 2:

[0089]

[0090] in:

[0091] A is the activity level of industrial enterprises.

[0092] F is the truck flow.

[0093] β0, β1, ..., β k are the parameters of the polynomial regression model. In seasonal model 2, these parameters are different for each season, i.e., β 1s , β 2s ..., β ks . s is the regression coefficient for the interaction with season s and truck traffic (not needed in the second model). s is a dummy variable constructed for season s. When the observed value belongs to season s, then D s =1, otherwise D s =0, ε is the error term.

[0094] In the seasonal model one expression:

[0095] This model explains the relationship between the level of industrial enterprise activity A and the truck flow F, while taking into account the seasonal effect. k are the parameters of the basic polynomial regression model. These parameters describe the basic relationship between truck traffic F and industrial enterprise activity level A. β1 means that for every unit increase in truck traffic, the level of industrial enterprise activity is expected to increase by β1 units (other conditions unchanged).

[0096] D s is the seasonal dummy variable. For each season (spring, summer, fall, winter), we have a corresponding dummy variable. If the data point belongs to a particular season, the dummy variable value for that season is 1, and the dummy variable value for other seasons is 0. In this way, the model can capture different effects for each season.

[0097] Υ s is the regression coefficient for the interaction with season s and truck flow. The interaction term D s *F allows the model to take into account the interaction between seasonality and truck traffic. In other words, it allows each season to have a different effect on the relationship between truck traffic and the level of industrial firm activity.

[0098] ε is the error term.

[0099] In the seasonal model II expression: This model allows for different polynomial effects for each season. Similar to the first model, β0 is the intercept term.

[0100] For each season s, we have a specific set of parameters β 1s , β 2s ..., β ks The parameters are the same as those in the regression model. They describe the relationship between the truck traffic F and the level of industrial activity A during the season. In other words, the model can now fit a different polynomial curve for each season.

[0101] D s It is still a seasonal dummy variable to indicate which season the data point belongs to.

[0102] In general, seasonal model 1 takes into account the effect of seasonality by adding an interaction term between seasonality and truck traffic, while seasonal model 2 allows different polynomial effects for each season, thus being able to capture the relationship between seasonality and truck traffic more flexibly. In practical applications, the choice of which model depends on the characteristics of the data and the modeling objectives. When the seasonal effect varies greatly between seasons and has different relationships with truck traffic, seasonal model 2 is more appropriate. When the seasonal effect is relatively consistent but differs in intensity between seasons, seasonal model 1 may be simpler and more suitable.

[0103] Adjust the activity status of industrial enterprises in the future by optimizing the completed model.

[0104] In a preferred embodiment of the present invention, in S4, adjusting the future activity state of the industrial enterprise specifically involves intervening and adjusting the activities of the industrial enterprise according to the predicted value output by the model.

[0105] It should be noted that the above-mentioned forecasting model is used to forecast the activity level of industrial enterprises in the future. The forecasting model is based on the analysis and learning of historical data to automatically calculate the results. The model uses the existing truck flow data to estimate the details of future truck flow and provides decision support for the production plan and logistics arrangement of industrial enterprises through the forecast value. When the forecast shows that the future truck flow will increase significantly, the enterprise needs to increase production capacity or adjust the inventory strategy in advance to cope with the potential demand growth.

[0106] Example systems:

[0107] An enterprise intelligence management system based on big data, comprising:

[0108] Area division module: used to divide streets into blocks to ensure that cameras can cover key areas and activities, including traffic nodes, public places and industrial enterprises;

[0109] Data collection unit: used to collect data monitored by cameras in each area and data on industrial enterprise activities, including data on logistics activities such as raw material procurement and product sales;

[0110] Data sorting and cleaning unit: used to sort the camera data and enterprise activity level data in chronological order and exclude irrelevant data such as passing vehicles and vehicles with special license plates;

[0111] Correlation analysis engine: used to analyze the correlation between truck traffic obtained from camera data and industrial enterprise activity level data, and calculate the correlation coefficient to measure the degree of correlation between the two;

[0112] Model building and training optimization unit: used to build a regression model, such as a polynomial regression model, based on the analysis results, and perform training optimization to capture the nonlinear relationship between truck flow and the activity level of industrial enterprises;

[0113] Seasonal Factor Analysis Module: used to identify seasonal variables and fit them into the established model to take into account the impact of seasonality on truck traffic and industrial enterprise activity levels;

[0114] Prediction model application unit: used to apply the optimized prediction model and adjust the activity status of the industrial enterprise in the future according to the prediction value output by the model;

[0115] Decision management system: used to provide decision support for production planning and logistics arrangements based on model predictions, helping enterprises to adjust production plans and logistics strategies in advance and optimize inventory management.

[0116] Those skilled in the art can understand that other details of the enterprise intelligent management system based on big data according to the embodiment of the present application are the same as the corresponding details previously described in the enterprise intelligent management method based on big data according to the embodiment of the present application, and will not be repeated here to avoid repetition.

[0117] The above is based on the ideal embodiment of the present invention. Through the above description, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention. The technical scope of the present invention is not limited to the content in the specification, and the technical scope must be determined according to the scope of the claims.

Claims

1. A method for enterprise intelligent management based on big data, characterized in that: The following steps are involved: S1: Divide each street into regions and collect data monitored by cameras and industrial enterprise activities in each region; S2: The camera data obtained in S1 is collecting and collating the data on the activity level of industrial enterprises and the traffic volume of trucks, and performing correlation analysis to study the relationship between the two; S3: Based on the data conclusions obtained from the analysis in S2, a regression model is established and trained and optimized to capture the nonlinear relationship between truck flow and the activity level of industrial enterprises; S4: The optimized model is used to predict the activity status of industrial enterprises in the future, providing decision support for their production planning and logistics arrangements.

2. According to claim 1, a method for enterprise intelligent management based on big data is characterized by: In the S1, the division is specifically to divide the street into blocks, so as to ensure that the camera can capture key areas and key activities.

3. The enterprise intelligence management method based on big data according to claim 2 is characterized by: The key areas include traffic nodes, public places, industrial enterprises and other areas where truck operation can be collected; the key activities include production activities, material procurement activities, material supply activities and other activities involving truck operation.

4. The enterprise intelligence management method based on big data according to claim 1 is characterized by: In S1, the industrial enterprise activity data is specifically data that can reflect the status of the industrial enterprise's logistics activities such as raw material procurement and product sales.

5. The enterprise intelligence management method based on big data according to claim 1 is characterized by: In S2, the collection and organization specifically includes arranging the camera data and the enterprise activity level data in chronological order into a collection, and excluding irrelevant data, and then making a one-to-one correspondence. The irrelevant data specifically includes judging and excluding passing vehicles and excluding vehicles with special license plates.

6. The enterprise intelligence management method based on big data according to claim 5 is characterized by: The basis for determining and excluding the passing vehicles is to set a time threshold. When the time a vehicle is in the camera is less than the threshold, it is determined to be a passing vehicle. The special license plate vehicles include training cars and police cars.

7. The enterprise intelligent management method based on big data according to claim 1 is characterized by: In S2, the analysis is specifically to perform a correlation analysis on the truck flow obtained from the camera data and the industrial enterprise activity level data, so as to analyze the intrinsic relationship between the two, by setting the correlation coefficient r, Among them, r is the correlation coefficient, which is used to measure the degree of correlation between F and A. The value range of the correlation coefficient r is [-1,1], where r=1 indicates a complete positive correlation, r=-1 indicates a complete negative correlation, and r=0 indicates no linear relationship.

8. The enterprise intelligence management method based on big data according to claim 1 is characterized by: In S3, model building specifically involves building a suitable model based on the characteristic relationship between vehicle flow and industrial enterprise activity levels obtained through analysis in S2, and training optimization specifically involves returning the data organized in S2 to the established model, training, optimizing, and testing the model, and ultimately obtaining an optimized prediction model.

9. The enterprise intelligence management method based on big data according to claim 1 is characterized by: In S3, seasonal variables are constructed through training optimization, and the seasonal variables are fitted into the established model to obtain a seasonal model.

10. An enterprise intelligence management system based on big data, characterized in that: include: Area division module: used to divide streets into blocks to ensure that cameras can cover key areas and activities, including traffic nodes, public places and industrial enterprises; Data collection unit: used to collect data monitored by cameras in each area and data on industrial enterprise activities, including data on logistics activities such as raw material procurement and product sales; Data sorting and cleaning unit: used to sort the camera data and enterprise activity level data in chronological order and exclude irrelevant data such as passing vehicles and vehicles with special license plates; Correlation analysis engine: used to analyze the correlation between truck traffic obtained from camera data and industrial enterprise activity level data, and calculate the correlation coefficient to measure the degree of correlation between the two; Model building and training optimization unit: used to build a regression model, such as a polynomial regression model, based on the analysis results, and perform training optimization to capture the nonlinear relationship between truck flow and the activity level of industrial enterprises; Seasonal Factor Analysis Module: used to identify seasonal variables and fit them into the established model to take into account the impact of seasonality on truck traffic and industrial enterprise activity levels; Prediction model application unit: used to apply the optimized prediction model and adjust the activity status of the industrial enterprise in the future according to the prediction value output by the model; Decision management system: used to provide decision support for production planning and logistics arrangements based on model predictions, helping enterprises to adjust production plans and logistics strategies in advance and optimize inventory management.

Images