An intelligent analysis and visualization system and method for fuel supply chain

Abstract

The present invention provides an intelligent analysis and visualization system and method for a fuel supply chain, which relates to the field of big data technology and includes: a data acquisition and processing module: pre-processing the link data of the supply chain collected in real time to obtain target link data; a supply analysis module: analyzing the first key operating indicator obtained based on the target link data to generate a supply analysis result; a user tracking module: determining user preference content based on user behavior; a visualization module: dynamically displaying the supply analysis results and the current user preference content. The first key operating indicator is obtained by analyzing the link data of the fuel supply chain collected in real time; anomaly detection, resource forecasting, and efficiency analysis are performed based on the first key operating indicator to generate a supply analysis result; user behavior data is analyzed to determine user preference content; and the supply analysis results and user preference content of each supply link are dynamically displayed, effectively improving the overall efficiency of the fuel supply chain while enhancing the user experience.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to an intelligent analysis and visualization system and method for a fuel supply chain. Background Art

[0002] In recent years, with the continuous growth of fuel demand and the increasing complexity of supply chains, traditional management methods have become unable to meet the demand for efficient, transparent, and risk-controlled supply chains. While some existing systems can monitor and collect data from fuel supply chains, most lack intelligent analysis and visualization capabilities, making it difficult to provide comprehensive and in-depth supply chain insights and decision support. Therefore, achieving comprehensive intelligent analysis and visualization of the supply chain to improve its overall efficiency has become a key research focus.

[0003] Therefore, the present invention provides an intelligent analysis and visualization system and method for a fuel supply chain. Summary of the Invention

[0004] The present invention provides an intelligent analysis and visualization system and method for a fuel supply chain, which is used to obtain a first key operating indicator by analyzing the link data of the fuel supply chain collected in real time; perform anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator to generate supply analysis results; analyze user behavior data to determine user preference content; and dynamically display the supply analysis results and user preference content of each supply link, effectively improving the overall efficiency of the fuel supply chain while enhancing the user experience.

[0005] The present invention provides an intelligent analysis and visualization system for a fuel supply chain, comprising:

[0006] Data collection and processing module: used to collect data on the current enterprise's fuel supply chain in real time, perform data preprocessing to obtain target link data, and store the target link data;

[0007] Supply analysis module: used to analyze the target link data to obtain the first key operating indicator, perform anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator, and generate supply analysis results;

[0008] User tracking module: used to record and analyze user behavior in the system, and determine user preference content based on the obtained user behavior data;

[0009] Visualization module: used to dynamically display the supply analysis results of each supply link and the current user preference content.

[0010] Preferably, the data acquisition and processing module includes:

[0011] Use the set collection tool to collect data from each supply link in the current enterprise's corresponding fuel supply chain in real time to obtain the first link data;

[0012] After denoising, filling missing values ​​and standardizing the first link data, the target link data is obtained;

[0013] Based on the supply link type, the target link data is classified and stored in a preset data warehouse.

[0014] Preferably, the supply analysis module includes:

[0015] Indicator acquisition unit: used to define data processing logic in a set stream computing framework and connect the preset data warehouse to the set stream computing framework;

[0016] The stream computing framework is set to receive and process target link data to obtain the first key operating indicator of the current supply link;

[0017] Supply analysis unit: used to perform anomaly detection, resource forecasting and supply efficiency analysis on the current supply link based on the first key operating indicator to obtain supply analysis results.

[0018] Preferably, the supply analysis unit includes:

[0019] Abnormal analysis block: used to obtain a first abnormal coefficient based on an indicator status mark obtained by comparing the first key operating indicator of the current supply link with a set operating indicator threshold;

[0020] According to the first abnormality coefficient, based on the set link-abnormality level table, determining the abnormality level of the current supply link;

[0021] If there are abnormal indicators in the current supply link, the abnormal indicators, abnormal levels and current supply link types are combined to obtain an abnormal analysis result set, which is then output as the supply analysis result;

[0022] Prediction analysis block: used to mark the situation based on the indicator status. If there is a correct indicator in the current supply link, and the correct indicator is a link resource indicator, the current correct indicator will be marked as the first indicator;

[0023] Extracting the indicator feature of the first indicator, and combining it with the current first indicator data to input into a pre-established resource prediction model, and outputting a resource prediction result;

[0024] When the resource prediction result is greater than the set resource upper limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is that resources may be insufficient;

[0025] When the resource prediction result is less than the set resource lower limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is resource surplus;

[0026] Combining the first indicator, resource forecast results, forecast result type and current supply link type to obtain a forecast analysis result set, which is then output as a supply analysis result;

[0027] Efficiency analysis block: used to extract efficiency dimension indicators from the first key operating indicators based on the efficiency requirements of the link;

[0028] Generate real-time efficiency reports based on the indicator data of efficiency dimension indicators;

[0029] Inputting the efficiency dimension index into the corresponding operation efficiency index model of the current supply link to obtain the link efficiency result of the current supply link;

[0030] The current supply link type, link efficiency results and real-time efficiency reports are combined to obtain an efficiency analysis result set, which is then output as the supply analysis result.

[0031] Preferably, the first abnormality coefficient is obtained based on the indicator status mark obtained by comparing the first key operating indicator of the current supply link with the set operating indicator threshold, including:

[0032] Compare the first key operating indicator of the current supply link with the set operating indicator threshold;

[0033] If the first key operating indicator is greater than the set operating indicator threshold, the current first key operating indicator is marked as an abnormal indicator;

[0034] If the first key operating indicator is not greater than the set operating indicator threshold, the current first key operating indicator is marked as a normal indicator;

[0035] Based on the number of normal / abnormal marks of the first key operating indicator in the current supply link, a first abnormality coefficient of the current supply link is obtained;

[0036] The calculation formula of the first anomaly coefficient is as follows:

[0037] ; In the formula, Y represents the first abnormal coefficient of the current supply link; Expressed as the total number of the first key operating indicators of the current supply link; It is expressed as the number of abnormal indicators in the current supply link; It is expressed as the important influence weight of the i-th abnormal indicator on the current supply link; It represents the indicator data of the i-th abnormal indicator of the current supply link; It represents the set operating index threshold of the i-th abnormal index of the current supply link, where i=1, 2, 3, , n1.

[0038] Preferably, the user tracking module includes:

[0039] Group analysis unit: used to perform data mining on the currently acquired behavior data of the target user to obtain a first behavior feature;

[0040] Inputting the first behavior feature into a pre-established user interest recognition model to obtain a first interest group;

[0041] Recommendation analysis unit: used for extracting the current user interest group of the target user from the user interest record;

[0042] If the current user interest group is consistent with the first interest group, marking the corresponding recommended content of the current user interest group as the first recommended content;

[0043] If the current user interest group is inconsistent with the first interest group, replacing the current user interest group of the target user with the first interest group;

[0044] Marking the recommended content corresponding to the first interest group as first recommended content;

[0045] Performing a priority analysis on the first recommended content to obtain a first priority coefficient;

[0046] The calculation formula of the first priority coefficient is as follows:

[0047] Where, It is represented as the first priority coefficient of the jth first recommended content; It is represented as the historical interaction frequency of the jth first recommended content; It is expressed as the influence weight of the popularity of the recommended content on the priority of the recommended content; It is represented as the time interval between the jth first recommended content and the last content interaction; Indicates the average length of time users stay on the currently recommended content; It is represented as the maximum time interval between all current first recommended contents and the last content interaction; It is represented as the average time interval between all current first recommended contents and the last content interaction; It is expressed as the weight of the impact of user interaction on the priority of recommended content; It is expressed as the content timeliness coefficient of the jth first recommended content; It is expressed as the ratio of the current remaining effective duration of the j-th first recommended content to the total effective duration; It is expressed as the weight of the influence of content timeliness on the priority of recommended content;

[0048] The first recommended contents are sorted from large to small according to the first priority coefficients to generate a first recommendation list, which is output as user preferred contents.

[0049] Preferably, the visualization module includes:

[0050] Supply visualization unit: used to create real-time dashboards and reports using visualization tools. Based on the efficiency analysis result set in the supply analysis results of each supply link, the efficiency dimension indicators and corresponding link efficiency results of the corresponding link are displayed.

[0051] When an abnormal detection result set is received, an abnormal alarm and abnormal indicators are immediately displayed in the preset display area;

[0052] Display resource forecasts in the preset display area according to the resource forecast result set;

[0053] User visualization unit: used to dynamically interact with target users based on their needs based on custom blocks; and to provide personalized recommendations to target users based on recommendation blocks;

[0054] Custom blocks: used in smart panels to provide user-defined dashboards, allowing target users to dynamically interact based on their needs.

[0055] Provides a set interactive function, allowing target users to drill down to specific data dimensions according to their own needs and view the corresponding efficiency dimension indicators that affect the efficiency of different supply links;

[0056] Recommendation block: used to receive and display personalized recommended content to target users in the smart display area based on the user's preferred content.

[0057] The present invention provides an intelligent analysis and visualization method for a fuel supply chain, comprising:

[0058] Step 1: Collect the data of the current enterprise's fuel supply chain in real time and perform data preprocessing to obtain the target link data, and store the target link data;

[0059] Step 2: Analyze the target link data to obtain a first key operating indicator, perform anomaly detection, resource forecasting, and efficiency analysis based on the first key operating indicator, and generate a supply analysis result;

[0060] Step 3: Record and analyze user behavior in the system, and determine user preference content based on the acquired user behavior data;

[0061] Step 4: Dynamically display the supply analysis results of each supply link and the current user preference content.

[0062] Compared with the prior art, the present invention has the following advantages:

[0063] The first key operating indicator is obtained by analyzing the real-time collected data of the fuel supply chain; anomaly detection, resource forecasting and efficiency analysis are performed based on the first key operating indicator to generate supply analysis results; user behavior data is analyzed to determine user preferences; the supply analysis results and user preferences of each supply link are dynamically displayed, effectively improving the overall efficiency of the fuel supply chain while enhancing user experience.

[0064] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0065] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0067] Figure 1 This is a structural diagram of an intelligent analysis and visualization system for a fuel supply chain according to an embodiment of the present invention;

[0068] Figure 2 The figure is a flow chart of an intelligent analysis and visualization method for a fuel supply chain according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0070] The embodiment of the present invention provides an intelligent analysis and visualization system for fuel supply chain, such as Figure 1 Shown, including:

[0071] Data collection and processing module: used to collect data on the current enterprise's fuel supply chain in real time, perform data preprocessing to obtain target link data, and store the target link data;

[0072] Supply analysis module: used to analyze the target link data to obtain the first key operating indicator, perform anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator, and generate supply analysis results;

[0073] User tracking module: used to record and analyze user behavior in the system, and determine user preference content based on the obtained user behavior data;

[0074] Visualization module: used to dynamically display the supply analysis results of each supply link and the current user preference content.

[0075] In this embodiment, the fuel supply chain refers to the information flow process from fuel production to end users, including fuel procurement, transportation, storage, processing, consumption and distribution; the target link data is obtained by preprocessing the link data of the current enterprise's fuel supply chain collected in real time, wherein the preprocessing includes denoising, missing value filling and standardization; the supply analysis results refer to the abnormality analysis result set, the resource prediction result set and the efficiency analysis result set; the user preference content refers to the first recommendation list, wherein the first recommendation list is obtained by sorting the first recommendation content from large to small according to the first priority coefficient.

[0076] The beneficial effects of the above technical solution are: obtaining the first key operating indicator by analyzing the link data of the fuel supply chain collected in real time; performing anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator to generate supply analysis results; analyzing user behavior data to determine user preference content; dynamically displaying the supply analysis results and user preference content of each supply link, effectively improving the overall efficiency of the fuel supply chain while enhancing user experience.

[0077] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain, wherein the data acquisition and processing module includes:

[0078] Use the set collection tool to collect data from each supply link in the current enterprise's corresponding fuel supply chain in real time to obtain the first link data;

[0079] After denoising, filling missing values ​​and standardizing the first link data, the target link data is obtained;

[0080] Based on the supply link type, the target link data is classified and stored in a preset data warehouse.

[0081] In this embodiment, the collection tool is pre-set, such as an API interface, a web crawler, etc.; the fuel supply chain refers to the information flow process from fuel production to end users, including fuel procurement, transportation, storage, processing, consumption and distribution; the first-link data refers to the operating data of each fuel link in the fuel supply chain, such as inventory, consumption, transportation time, cost, etc.; the target link data is obtained after data preprocessing of the first-link data, wherein the preprocessing includes denoising, missing value filling and standardization; the supply link types include fuel procurement, transportation, storage, processing, consumption and distribution, etc.; the preset data warehouse refers to a pre-established database for storing data of each supply link in the fuel supply chain and the preprocessed data of each supply link in the fuel supply chain.

[0082] The beneficial effect of the above technical solution is: by real-time tracking and collecting data from each supply link in the enterprise's fuel supply chain, and then processing and classifying the data for storage, it helps to understand the operating status of the fuel supply chain and lay the data support for subsequent visualization.

[0083] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain, wherein the supply analysis module includes:

[0084] Indicator acquisition unit: used to define data processing logic in a set stream computing framework and connect the preset data warehouse to the set stream computing framework;

[0085] The stream computing framework is set to receive and process target link data to obtain the first key operating indicator of the current supply link;

[0086] Supply analysis unit: used to perform anomaly detection, resource forecasting and supply efficiency analysis on the current supply link based on the first key operating indicator to obtain supply analysis results.

[0087] In this embodiment, the set streaming computing framework is a pre-configured streaming computing framework, such as Apache Flink, Apache Spark Streaming, etc., which is used to process the target link data to obtain key operating indicators; defining data processing logic refers to writing a processing function that executes the computing logic in the set streaming computing framework and implementing data cleaning, conversion, aggregation and other logic in the processing function; the preset data warehouse refers to a pre-established database for storing data of each supply link in the fuel supply chain and pre-processed data of each supply link in the fuel supply chain; the first key operating indicator is an operating indicator obtained by using the set streaming computing framework to process the target link data, such as time and cost.

[0088] The beneficial effects of the above technical solution are: by utilizing the real-time processing capabilities of the streaming computing framework, the corresponding first key operating indicators of each supply link are obtained; based on the first key operating indicators, anomaly detection, resource forecasting and supply efficiency analysis are performed on each supply link, which is conducive to the realization of intelligent management of the fuel supply chain.

[0089] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain, wherein the supply analysis unit includes:

[0090] Abnormal analysis block: used to obtain a first abnormal coefficient based on an indicator status mark obtained by comparing the first key operating indicator of the current supply link with a set operating indicator threshold;

[0091] According to the first abnormality coefficient, based on the set link-abnormality level table, determining the abnormality level of the current supply link;

[0092] If there are abnormal indicators in the current supply link, the abnormal indicators, abnormal levels and current supply link types are combined to obtain an abnormal analysis result set, which is then output as the supply analysis result;

[0093] Prediction analysis block: used to mark the situation based on the indicator status. If there is a correct indicator in the current supply link, and the correct indicator is a link resource indicator, the current correct indicator will be marked as the first indicator;

[0094] Extracting the indicator feature of the first indicator, and combining it with the current first indicator data to input into a pre-established resource prediction model, and outputting a resource prediction result;

[0095] When the resource prediction result is greater than the set resource upper limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is that resources may be insufficient;

[0096] When the resource prediction result is less than the set resource lower limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is resource surplus;

[0097] Combining the first indicator, resource forecast results, forecast result type and current supply link type to obtain a forecast analysis result set, which is then output as a supply analysis result;

[0098] Efficiency analysis block: used to extract efficiency dimension indicators from the first key operating indicators based on the efficiency requirements of the link;

[0099] Generate real-time efficiency reports based on the indicator data of efficiency dimension indicators;

[0100] Inputting the efficiency dimension index into the corresponding operation efficiency index model of the current supply link to obtain the link efficiency result of the current supply link;

[0101] The current supply link type, link efficiency results and real-time efficiency reports are combined to obtain an efficiency analysis result set, which is then output as the supply analysis result.

[0102] In this embodiment, the setting of the operating indicator threshold is pre-set, and is obtained by selecting the maximum value of the indicator from all historical indicator values ​​obtained after normalizing the historical data of the first key operating indicator within a preset time period and calculating the average value of the indicator, and then calculating the average value of the maximum value of the indicator and the average value of the indicator; the first abnormality coefficient is used to express the degree of abnormality of the current supply link; the setting link-abnormality level table is composed of the supply link type, the abnormality coefficient range and the corresponding abnormality level; the abnormality analysis result set is composed of the abnormal indicator, the abnormality level and the current supply link type; the first indicator refers to the link resource indicator marked as normal, wherein the link resource indicator refers to a predetermined resource allocation indicator related to the supply link, such as inventory and order quantity; the first indicator data refers to the data value of the current first indicator.

[0103] In this embodiment, the resource prediction model is a model obtained by training a neural network using data generated after preprocessing and feature extraction of historical data of link resource indicators within a preset time period as training data, and is used to predict the future resource status of the current indicator; the prediction result types include two types: possible resource shortage and resource surplus; setting the resource upper limit refers to 90% of the maximum safe resource amount of the current link resource indicator; setting the resource lower limit refers to 15% of the maximum safe resource amount of the current link resource indicator, where the maximum safe resource amount is determined based on the current enterprise's existing inventory strategy, such as safety stock and economic order quantity; the prediction analysis result set is composed of the first indicator, resource prediction result, prediction result type and current supply link type.

[0104] In this embodiment, the efficiency dimension indicator is extracted from the first key operating indicator based on the link efficiency demand of the current supply link, combined with the set demand-dimensional indicator list. For example, the business demand of the procurement link is procurement efficiency, and the corresponding efficiency dimension indicators refer to procurement cost and procurement speed; the business demand of the transportation link is transportation efficiency, and the corresponding efficiency dimension indicators refer to transportation cost and transportation time; the set demand-dimensional indicator list is composed of the supply link type, efficiency demand and corresponding dimension indicators; the link efficiency result refers to the efficiency index value calculated by inputting the current efficiency dimension indicator into the corresponding operating efficiency index model of the current supply link, and the operating efficiency index model is represented by a pre-set formula for calculating the efficiency index value of the corresponding supply link based on the efficiency dimension indicator; the real-time efficiency report is composed of efficiency dimension indicators, indicator data, supply link type and report generation time; the efficiency analysis result set is composed of the current supply link type, link efficiency result and real-time efficiency report.

[0105] The beneficial effect of the above technical solution is that by performing anomaly detection, resource forecasting and supply efficiency analysis on each supply link based on the first key operating indicator, it is conducive to realizing intelligent management of the fuel supply chain.

[0106] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain. The system obtains a first abnormality coefficient based on an indicator status mark obtained by comparing a first key operating indicator of a current supply link with a set operating indicator threshold, including:

[0107] Compare the first key operating indicator of the current supply link with the set operating indicator threshold;

[0108] If the first key operating indicator is greater than the set operating indicator threshold, the current first key operating indicator is marked as an abnormal indicator;

[0109] If the first key operating indicator is not greater than the set operating indicator threshold, the current first key operating indicator is marked as a normal indicator;

[0110] Based on the number of normal / abnormal marks of the first key operating indicator in the current supply link, a first abnormality coefficient of the current supply link is obtained;

[0111] The calculation formula of the first anomaly coefficient is as follows:

[0112] ; In the formula, Y represents the first abnormal coefficient of the current supply link; Expressed as the total number of the first key operating indicators of the current supply link; It is expressed as the number of abnormal indicators in the current supply link; It is expressed as the important influence weight of the i-th abnormal indicator on the current supply link; It represents the indicator data of the i-th abnormal indicator of the current supply link; It represents the set operating index threshold of the i-th abnormal index of the current supply link, where i=1, 2, 3, , n1.

[0113] In this embodiment, the abnormal indicator refers to the first key operating indicator that is greater than the set operating indicator threshold; the correct indicator refers to the first key operating indicator that is not greater than the set operating indicator threshold; the status marking situation refers to the number of normal / abnormal marks of all first key operating indicators in the current supply link.

[0114] In this embodiment, the weight of the important influence of the abnormal indicators on the current supply link is obtained by solving a matrix constructed by using the hierarchical analysis method to compare the abnormal indicators with each other and score their relative importance.

[0115] The beneficial effect of the above technical solution is: by analyzing the abnormality index and calculating the first abnormality coefficient, the abnormality location and abnormality degree estimation of the current supply link can be realized, which helps to realize the risk assessment of the supply link and thus improve the stability of the supply chain.

[0116] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain, wherein the user tracking module includes:

[0117] Group analysis unit: used to perform data mining on the currently acquired behavior data of the target user to obtain a first behavior feature;

[0118] Inputting the first behavior feature into a pre-established user interest recognition model to obtain a first interest group;

[0119] Recommendation analysis unit: used for extracting the current user interest group of the target user from the user interest record;

[0120] If the current user interest group is consistent with the first interest group, marking the corresponding recommended content of the current user interest group as the first recommended content;

[0121] If the current user interest group is inconsistent with the first interest group, replacing the current user interest group of the target user with the first interest group;

[0122] Marking the recommended content corresponding to the first interest group as first recommended content;

[0123] Performing a priority analysis on the first recommended content to obtain a first priority coefficient;

[0124] The calculation formula of the first priority coefficient is as follows:

[0125] Where, It is represented as the first priority coefficient of the jth first recommended content; It is represented as the historical interaction frequency of the jth first recommended content; It is expressed as the influence weight of the popularity of the recommended content on the priority of the recommended content; It is represented as the time interval between the jth first recommended content and the last content interaction; Indicates the average length of time users stay on the currently recommended content; It is represented as the maximum time interval between all current first recommended contents and the last content interaction; It is represented as the average time interval between all current first recommended contents and the last content interaction; It is expressed as the weight of the impact of user interaction on the priority of recommended content; It is expressed as the content timeliness coefficient of the jth first recommended content; It is expressed as the ratio of the current remaining effective duration of the j-th first recommended content to the total effective duration; It is expressed as the weight of the influence of content timeliness on the priority of recommended content;

[0126] The first recommended contents are sorted from large to small according to the first priority coefficients to generate a first recommendation list, which is output as user preferred contents.

[0127] In this embodiment, the behavioral data includes the number of clicks, browsing paths (such as page jump order), dwell time, search keywords, etc.; the first behavioral feature is obtained by extracting features from the user's current behavioral data; the user interest recognition model refers to a model obtained by training a neural network using training data established after preprocessing and feature extraction of the user's historical behavioral data, and is used to identify the interest group to which the current user should be divided based on the input user behavior data. The interest group refers to a set of users with similar preferences; the first interest group is obtained by inputting the first behavioral feature into a pre-established user interest recognition model; recommended content refers to content recommended to the current user, such as a data change chart of an efficiency dimension indicator; the first priority coefficient is used to express the priority of the current first recommended content; the first recommendation list is obtained by sorting the first recommended content from large to small according to the first priority coefficient.

[0128] In this embodiment, the historical interaction frequency refers to the user's past interaction behavior data on the recommended content, such as browsing history, click history, etc.; the content timeliness coefficient is a pre-configured coefficient based on the preset timeliness level of the current recommended content, where the timeliness level includes short-term timeliness, ordinary timeliness and long-term timeliness; the current remaining effective time refers to the length of time from the current moment that the recommended content remains effective for the user; the total effective time refers to the total length of time from the creation of the recommended content to the loss of its effectiveness; the influence weights assigned to the popularity of the content, user interaction and content timeliness are obtained by solving a matrix constructed after pairwise comparison and relative importance scoring using the hierarchical analysis method.

[0129] The beneficial effect of the above technical solution is: by analyzing the user behavior data obtained based on user tracking, personalized recommendations can be made to users, which can not only effectively improve the user experience, but also enhance the market competitiveness of the enterprise.

[0130] An embodiment of the present invention provides an intelligent analysis and visualization system for a fuel supply chain, wherein the visualization module includes:

[0131] Supply visualization unit: used to create real-time dashboards and reports using visualization tools. Based on the efficiency analysis result set in the supply analysis results of each supply link, the efficiency dimension indicators and corresponding link efficiency results of the corresponding link are displayed.

[0132] When an abnormal detection result set is received, an abnormal alarm and abnormal indicators are immediately displayed in the preset display area;

[0133] Display resource forecasts in the preset display area according to the resource forecast result set;

[0134] User visualization unit: used to dynamically interact with target users based on their needs based on custom blocks; and to provide personalized recommendations to target users based on recommendation blocks;

[0135] Custom blocks: used in smart panels to provide user-defined dashboards, allowing target users to dynamically interact based on their needs.

[0136] Provides a set interactive function, allowing target users to drill down to specific data dimensions according to their own needs and view the corresponding efficiency dimension indicators that affect the efficiency of different supply links;

[0137] Recommendation block: used to receive and display personalized recommended content to target users in the smart display area based on the user's preferred content.

[0138] In this embodiment, visualization tools include ableau, Power BI, Grafana, etc.; set interactive functions include filters, slicers, drill-down, etc., to ensure that users explore multi-dimensional data; the efficiency analysis result set consists of supply link type, link efficiency results and real-time efficiency reports.

[0139] In this embodiment, the smart panel refers to a platform that integrates multiple visualization tools and functions, allowing users to create and manage dashboards according to their own needs; dashboard functions include drag-and-drop layout, display of multiple chart types, real-time data updates, data interaction, and data export functions; content preference content refers to the first recommended content list; the first recommended list is obtained by sorting the first recommended content from large to small according to the first priority coefficient.

[0140] The beneficial effects of the above technical solution are: by using visualization tools to create real-time dashboards and reports, displaying key indicators and indexes, personalizing the display based on user behavior and allowing users to drill down to specific data to view factors affecting efficiency, the overall efficiency of the supply chain and user experience can be effectively improved.

[0141] The embodiment of the present invention provides an intelligent analysis and visualization method for a fuel supply chain, such as Figure 2 Shown, including:

[0142] Step 1: Collect the data of the current enterprise's fuel supply chain in real time and perform data preprocessing to obtain the target link data, and store the target link data;

[0143] Step 2: Analyze the target link data to obtain a first key operating indicator, perform anomaly detection, resource forecasting, and efficiency analysis based on the first key operating indicator, and generate a supply analysis result;

[0144] Step 3: Record and analyze user behavior in the system, and determine user preference content based on the acquired user behavior data;

[0145] Step 4: Dynamically display the supply analysis results of each supply link and the current user preference content.

[0146] The beneficial effects of the above technical solution are: obtaining the first key operating indicator by analyzing the link data of the fuel supply chain collected in real time; performing anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator to generate supply analysis results; analyzing user behavior data to determine user preference content; dynamically displaying the supply analysis results and user preference content of each supply link, effectively improving the overall efficiency of the fuel supply chain while enhancing user experience.

[0147] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An intelligent analysis and visualization system for fuel supply chain, characterized by: include: Data acquisition and processing module: used to collect data on the current enterprise's fuel supply chain in real time, perform data preprocessing to obtain target link data, and store the target link data; Supply analysis module: used to analyze the target link data to obtain the first key operating indicator, perform anomaly detection, resource forecasting and efficiency analysis based on the first key operating indicator, and generate supply analysis results; User tracking module: used to record and analyze user behavior in the system, and determine user preference content based on the obtained user behavior data; Visualization module: used to dynamically display the supply analysis results of each supply link and the current user preferences; The user tracking module includes: Group analysis unit: used to perform data mining on the currently acquired behavior data of the target user to obtain a first behavior feature; Inputting the first behavior feature into a pre-established user interest recognition model to obtain a first interest group; Recommendation analysis unit: used for extracting the current user interest group of the target user from the user interest record; If the current user interest group is consistent with the first interest group, marking the corresponding recommended content of the current user interest group as the first recommended content; If the current user interest group is inconsistent with the first interest group, replacing the current user interest group of the target user with the first interest group; Marking the recommended content corresponding to the first interest group as first recommended content; Performing a priority analysis on the first recommended content to obtain a first priority coefficient; The calculation formula of the first priority coefficient is as follows: Where, It is represented as the first priority coefficient of the jth first recommended content; It is represented as the historical interaction frequency of the jth first recommended content; It is expressed as the influence weight of the popularity of the recommended content on the priority of the recommended content; It is represented as the time interval between the jth first recommended content and the last content interaction; Indicates the average length of time users stay on the currently recommended content; It is represented as the maximum time interval between all current first recommended contents and the last content interaction; It is represented as the average time interval between all current first recommended contents and the last content interaction; It is expressed as the weight of the impact of user interaction on the priority of recommended content; It is expressed as the content timeliness coefficient of the jth first recommended content; It is expressed as the ratio of the current remaining effective duration of the j-th first recommended content to the total effective duration; It is expressed as the weight of the influence of content timeliness on the priority of recommended content; The first recommended contents are sorted from large to small according to the first priority coefficients to generate a first recommendation list, which is output as user preferred contents.

2. The intelligent analysis and visualization system for fuel supply chain according to claim 1, characterized in that: The data acquisition and processing module includes: Use the set collection tool to collect data from each supply link in the current enterprise's corresponding fuel supply chain in real time to obtain the first link data; After denoising, filling missing values ​​and standardizing the first link data, the target link data is obtained; Based on the supply link type, the target link data is classified and stored in a preset data warehouse.

3. The intelligent analysis and visualization system for fuel supply chain according to claim 1, characterized in that: The supply analysis module includes: Indicator acquisition unit: used to define data processing logic in a set stream computing framework and connect the preset data warehouse to the set stream computing framework; The stream computing framework is set to receive and process target link data to obtain the first key operating indicator of the current supply link; Supply analysis unit: used to perform anomaly detection, resource forecasting and supply efficiency analysis on the current supply link based on the first key operating indicator to obtain supply analysis results.

4. The intelligent analysis and visualization system for fuel supply chain according to claim 3, characterized in that: The supply analysis unit includes: Abnormal analysis block: used to obtain a first abnormal coefficient based on an indicator status mark obtained by comparing the first key operating indicator of the current supply link with a set operating indicator threshold; According to the first abnormality coefficient, based on the set link-abnormality level table, determining the abnormality level of the current supply link; If there are abnormal indicators in the current supply link, the abnormal indicators, abnormal levels and current supply link types are combined to obtain an abnormal analysis result set, which is then output as the supply analysis result; Prediction analysis block: used to mark the situation based on the indicator status. If there is a correct indicator in the current supply link, and the correct indicator is a link resource indicator, the current correct indicator will be marked as the first indicator; Extracting the indicator feature of the first indicator, and combining it with the current first indicator data to input into a pre-established resource prediction model, and outputting a resource prediction result; When the resource prediction result is greater than the set resource upper limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is that resources may be insufficient; When the resource prediction result is less than the set resource lower limit corresponding to the first indicator, determining that the current prediction result type of the first indicator is resource surplus; Combining the first indicator, resource forecast results, forecast result type and current supply link type to obtain a forecast analysis result set, which is then output as a supply analysis result; Efficiency analysis block: used to extract efficiency dimension indicators from the first key operating indicators based on the efficiency requirements of the link; Generate real-time efficiency reports based on the indicator data of efficiency dimension indicators; Inputting the efficiency dimension index into the corresponding operation efficiency index model of the current supply link to obtain the link efficiency result of the current supply link; The current supply link type, link efficiency results and real-time efficiency reports are combined to obtain an efficiency analysis result set, which is then output as the supply analysis result.

5. The intelligent analysis and visualization system for fuel supply chain according to claim 4, characterized in that: Based on the indicator status mark obtained by comparing the first key operating indicator of the current supply link with the set operating indicator threshold, a first abnormality coefficient is obtained, including: Compare the first key operating indicator of the current supply link with the set operating indicator threshold; If the first key operating indicator is greater than the set operating indicator threshold, the current first key operating indicator is marked as an abnormal indicator; If the first key operating indicator is not greater than the set operating indicator threshold, the current first key operating indicator is marked as a normal indicator; Based on the number of normal / abnormal marks of the first key operating indicator in the current supply link, a first abnormality coefficient of the current supply link is obtained; The calculation formula of the first anomaly coefficient is as follows: ; In the formula, Y represents the first abnormal coefficient of the current supply link; Expressed as the total number of the first key operating indicators of the current supply link; It is expressed as the number of abnormal indicators in the current supply link; It is expressed as the important influence weight of the i-th abnormal indicator on the current supply link; It represents the indicator data of the i-th abnormal indicator of the current supply link; It represents the set operating index threshold of the i-th abnormal index of the current supply link, where i=1, 2, 3, , n1.

6. The intelligent analysis and visualization system for fuel supply chain according to claim 1, characterized in that: The visualization module includes: Supply visualization unit: used to create real-time dashboards and reports using visualization tools. Based on the efficiency analysis result set in the supply analysis results of each supply link, the efficiency dimension indicators and corresponding link efficiency results of the corresponding link are displayed. When an abnormal detection result set is received, an abnormal alarm and abnormal indicators are immediately displayed in the preset display area; Display resource forecasts in the preset display area according to the resource forecast result set; User visualization unit: used to dynamically interact with target users based on their needs based on custom blocks; and to provide personalized recommendations to target users based on recommendation blocks; Custom blocks: used in smart panels to provide user-defined dashboards, allowing target users to dynamically interact based on their needs. Provides a set interactive function, allowing target users to drill down to specific data dimensions according to their own needs and view the corresponding efficiency dimension indicators that affect the efficiency of different supply links; Recommendation block: used to receive and display personalized recommended content to target users in the smart display area based on the user's preferred content.

7. An intelligent analysis and visualization method for fuel supply chain, characterized in that: include: Step 1: Collect the data of the current enterprise's fuel supply chain in real time and perform data preprocessing to obtain the target link data, and store the target link data; Step 2: Analyze the target link data to obtain a first key operating indicator, perform anomaly detection, resource forecasting, and efficiency analysis based on the first key operating indicator, and generate a supply analysis result; Step 3: Record and analyze user behavior in the system, and determine user preference content based on the acquired user behavior data; Step 4: Dynamically display the supply analysis results of each supply link and the current user preference content; The step 4 comprises: Performing data mining on the currently acquired target user's behavior data to obtain a first behavior feature; Inputting the first behavior feature into a pre-established user interest recognition model to obtain a first interest group; extracting the current user interest group of the target user from the user interest record; If the current user interest group is consistent with the first interest group, marking the corresponding recommended content of the current user interest group as the first recommended content; If the current user interest group is inconsistent with the first interest group, replacing the current user interest group of the target user with the first interest group; Marking the recommended content corresponding to the first interest group as first recommended content; Performing a priority analysis on the first recommended content to obtain a first priority coefficient; The calculation formula of the first priority coefficient is as follows: Where, It is represented as the first priority coefficient of the jth first recommended content; It is represented as the historical interaction frequency of the jth first recommended content; It is expressed as the influence weight of the popularity of the recommended content on the priority of the recommended content; It is represented as the time interval between the jth first recommended content and the last content interaction; Indicates the average length of time users stay on the currently recommended content; It is represented as the maximum time interval between all current first recommended contents and the last content interaction; It is represented as the average time interval between all current first recommended contents and the last content interaction; It is expressed as the weight of the impact of user interaction on the priority of recommended content; It is expressed as the content timeliness coefficient of the jth first recommended content; It is expressed as the ratio of the current remaining effective duration of the j-th first recommended content to the total effective duration; It is expressed as the weight of the influence of content timeliness on the priority of recommended content; The first recommended contents are sorted from large to small according to the first priority coefficients to generate a first recommendation list, which is output as user preferred contents.

Images