Big data visualization of the quantification of influencing factors and key monitoring indicators in the refined oil products market based on fuzzy mathematics

Abstract

In this paper, an in-depth study on the quantification of influencing factors and big data visualization of key monitoring indicators in the refined oil products market is carried out through fuzzy mathematical methods, and a system for quantifying influencing factors and big data visualization of key monitoring indicators in the refined oil products market with the fuzzy mathematical background is designed and implemented. The system realizes the functions of flow visualization, attack visualization, target tracking visualization, etc., and optimizes the system from the perspectives of performance and visualization effect. It achieves the display and interaction of multi-dimensional data in space and time with multiple views, angles, and dimensions. Data tagging and data correlation for key aspects of the product production process are realized through fuzzy mathematics and other means, and a quality traceability system for the manufacturing industry is realized on this basis, through which the data of some key stages of the product production process can be displayed retrospectively. The study proves that the business model of refined oil logistics platform based on value network can significantly improve the user’s perceived value and benefit all parties within the value network, realizing the complementary advantages of refined oil production enterprises and logistics platform companies, improving the efficiency of enterprise’s logistics and maximizing the profit of each subject within the value network to achieve profitability for all parties.

Keywords

Fuzzy mathematics refined products market quantification of influences key monitoring indicators big data visualization

1 Introduction

The deepening globalization process has brought great influence on the operating environment of enterprises, especially after the international financial crisis, international large oil companies have been adjusting their sales business of refined oil products, and by their advanced technology and management level, they are optimizing their marketing strategies on a global scale and rapidly expanding overseas markets, making the market competition in the refined oil industry more and more intense [1]. However, with the gradual liberalization of the industrial policy, the monopoly of the industry was gradually broken, private enterprises and foreign enterprises also entered the refined oil market, further intensifying the degree of market competition. Also, the arrival of the “Internet +” era, not only to the social and economic structure and industrial pattern has brought a profound impact, but also changed people’s habits of life and ideas, to the traditional marketing concept and marketing model has brought a very big impact [2]. Along with the thin profit of refined oil sales, the maturity of new energy development and the diversification of customer demand, how to change the marketing mindset of enterprises, innovative marketing model, using advanced technology to tap the huge data resources of the refined oil sales industry, to provide customers with accurate marketing and customized marketing programs has become the refined oil sales enterprises urgently need to crack the problem [3 –8]. The oil retail business is the core business of oil sales enterprises, the development of the retail market is the source of profits in the petroleum and petrochemical industry, the key to corporate development [9]. At present, the society has entered the era of Internet and big data technology popularization, how to use the Internet and big data, change the marketing strategy, is the refined oil retail market profit, a good opportunity to increase efficiency [10]. Oil sales companies are good representatives of refined oil retail enterprises, through the combination of oil sales companies in the refined oil retail market marketing strategy and big data technology to explore and summarize the case analysis, research oil sales companies in the refined oil retail market marketing status quo, identify the characteristics and deficiencies of the current marketing policy, for gas station retail business characteristics, combined with enterprise big data application ideas, to tallish a precision marketing service system based on big data and centered on customer service is the research focus of this paper [11].

Hu proposed the concept of direct marketing based on traditional marketing concepts, emphasizing that modern enterprises need to use databases to drive their marketing, relying on the Internet, big data, cloud platform, and other advanced information technology to achieve the precision marketing activities of enterprises [12]. On this basis, an accurate market segmentation model is established from the marketing perspective, and the market data of the Three Guns Group is targeted and highly mined using the SPSS tool [13]. Taking the phenomenon of a massive increase in global data volume as an object of study, Mura found that big data technology can make consumer behavior more personalized and the laws of consumer demand can be captured by big data technology, which is helpful for enterprises to grasp the laws and changes of consumer demand in time and realize more accurate marketing on this basis, which at the same time puts forward new requirements for enterprises, which need to handle and analyze current events promptly [14]. Quickly respond to various market changes, accelerate the shortening of channel distance with consumers, and increase price transparency and gain consumer trust [15]. With the continuous development and application of bioengineering in economic management life, Jaffee introduced the concept of supergene in biology into the marketing management theory and formed a supergene marketing strategy model for enterprises, which proposed strategies for enterprises to carry out supergene marketing from the aspects of constructing sales field, sales strategy level, team health status and salesman ratio arrangement [16]. According to Bugert, in the fast-developing internet era, enterprises must be able to change their marketing concepts in time to attract consumers by designing more personalized products and enhancing their sense of experience and satisfaction to achieve innovation in marketing strategies [17]. Xie conducted an in-depth study on big data precision marketing, based on defining a clear definition of big data precision marketing, analyzing the importance of this innovation, and proposing specific measures for enterprises to carry out big data precision marketing [18].

This paper introduces big data theory into the marketing field of the refined oil retail market, enriches the research results of refined oil retail market sales theory, extends the research scope of big data theory, and demonstrates the theoretical feasibility of big data technology to help improve the precise marketing of refined oil retail market. This paper analyzes the current marketing situation of the refined oil products retail market and discusses the characteristics and shortcomings of the current marketing strategies. In the specific case of oil sales company, with the help of big data technology, to establish a customer-centered marketing system, according to the big data reflected in the customer’s ability to consumer behavior and customer consumption habits to achieve precision marketing, not only can help enterprises reduce promotional costs, but also help enterprises to improve revenue and profitability. This paper takes the oil sales company as an example, in-depth discussion of the necessity and feasibility of big data applications in the oil retail industry, analysis, and design of a set of complete and scientific marketing strategy, the industry has a certain degree of generality of the retail market, in the pilot, the operation can be applied after the promotion.

2 Fuzzy mathematics-based big data visualization design for the quantification of market influencing factors and key monitoring indicators of refined oil products

2.1 Fuzzy mathematical model design for the quantification of factors influencing the market for refined petroleum products

Fuzzy mathematics is a new type of mathematical analysis method used for fuzzy phenomena, which extends the binary logic (0, 1) to the continuous value logic of an infinite number of values on [0, 1], deriving the traditional mathematics from the binary logic to the continuous bit logic, avoiding the absolute “good” and “bad” in evaluation or empirical evidence. The limitations of “bad", the division of relative “good” and “bad” in the appropriate numerical limit threshold space, the expansion of the controllable range, the further clarification of ambiguity, and the selection of the result [19]. The combination of fuzzy mathematical method and entropy power Topsis method can effectively exert the advantages of objectivity and impartiality of the two types of models, improve the uncertainty and subjective judgment in the evaluation of indicators, further combine qualitative and quantitative analysis, and enhance the evaluation effect of urban comfort in the study area, which can not only show the ambiguity of evaluation factors and measurement process, but also reduce the errors formed by subjective judgment, and make the evaluation results more accurate [20].

First, construct the original matrix X: $X = (x_{ij})_{m * n} = [\begin{matrix} x_{11} & x_{21} & . . . & x_{n 1} \\ x_{12} & x_{22} & . . . & x_{n 2} \\ . . . & . . . & . . . & x_{11} \\ x_{1 m} & x_{2 m} & . . . & x_{nm} \end{matrix}]$ (1)

By normalizing the raw data, the standardized matrix Y is obtained: $Y_{ij} = \frac{y_{ij}}{\sum_{i = 1}^{n} y_{ij}}$ (2) $Y = (Y_{ij})_{n * m} = [\begin{matrix} Y_{11} & Y_{21} & . . . & Y_{n 1} \\ Y_{12} & Y_{22} & . . . & Y_{n 2} \\ . . . & . . . & . . . & Y_{11} \\ Y_{1 m} & Y_{2 m} & . . . & Y_{nm} \end{matrix}]$ (3)

Information entropy is shown in formula (4). $F_{j} = - \frac{1}{ln m} \sum_{i = 1}^{n} x_{ij} ln x_{xj}$ (4)

Measurement of indicator weights: where F_j is the weight of the corresponding indicator in the urban comfort evaluation system. $M_{j} = \frac{1 - M_{j}}{\sum_{i = 1}^{n} 1 - M_{j}}$ (5)

Precision marketing refers to the enterprise to establish a personalized customer communication service system based on modern information technology to achieve low-cost expansion of enterprises, that is, enterprises need to fully understand customer information based on personalized preferences and needs of customers for targeted marketing strategies, and at the same time, after mastering certain customer and market information, the database marketing and direct marketing of an organic combination of a kind of new marketing trends. Specifically, the concept of precision marketing can be explained on three levels: firstly, the idea of precision marketing, the goal of marketing management of a company is to achieve marketing without marketing, and the transitional stage to achieve this goal is precision marketing. Secondly, precision marketing not only needs to reduce the cost of the enterprise as much as possible but also needs to achieve the sustainable development of the enterprise. Finally, the means of precision marketing are capable of being measured. Specifically, precision marketing mainly consists of two parts: one is precision and the other is marketing, the former means that the strategy can be measured and is precise; the latter means that marketing should help customers find their real needs, provide them with the exact products and services they need, and solve their problems. Therefore, precision marketing essentially means that a company should precisely locate its customer groups and establish good brand value and corporate image by promoting and delivering its products and services, and use this to attract new customers, retain old customers, and establish a brand value orientation among its customer groups, as shown in Fig. 1.

Fig. 1

Fuzzy mathematical quantification model of oil products market influences.

To explore the level of comfort, it is still necessary to calculate a fuzzy composite assessment of urban comfort by calculating the weight set and fuzzy matrix to reflect the weight and influence of each corresponding indicator, based on the measurement of indicator h. The calculation is as follows. $h = WU$ (6)

Coefficient of variation: $C_{v} = \frac{1}{y} \sqrt{\frac{1}{m} \sum_{i = 1}^{m} (y_{i} - 1)^{2}}$ (7)

Thiel coefficient: $N = \frac{1}{n} \sum_{i = 1}^{n} x_{ij} ln x_{xj}$ (8)

Among the fuzzy mathematical models, the maximum affiliation method is the most widely used, but it also has limitations, i.e., when the difference between different levels of affiliation of the discriminating object is weak, it is likely to lead to gaps in the evaluation levels and form unreasonable evaluation matrices, therefore, this study further combines the linear weighted average principle to calculate the affiliation levels based on the research literature related to urban comfort at home and abroad. $B = \frac{\sum_{j}^{15} B_{j}^{k} j}{\sum_{j}^{15} B_{j}^{k}}$ (9)

2.2 Design of big data visualization for monitoring indicators

Big data visualization is the application of visualization methods in the context of big data, with the development of science and technology, the data of various industries show explosive growth, people continue to guide the practice by accessing massive amounts of data information, the value of big data is huge, and the process of visualization is one of the most important ways for humans to access its value. Text information is one of the most common in our life. By analyzing the text, we can obtain its semantic features and visualize the specified features. Network information is a more common data structure that often requires the visualization of points and connections to uncover the interconnectedness of the data. Patio-temporal data refers to data containing temporal and spatial information, and the visualization of this type of data needs to show the distribution of geographic locations and the changes of data over time. Data contains multiple information, and multidimensional data is displayed by choosing the appropriate visualization method. Geometry-based visualization methods are geometric elements to represent data, the common ones are scattered plot and parallel coordinate visualization; pixel-based visualization, which can represent the data obtained in different dimensions with different pixel colors; icon-based technology, which is to describe the data at different latitudes in the form of icons, such as star charts, face charts, etc. Hierarchy-based technology is to divide multi-dimensional data with hierarchy into multiple subspaces, which will also be divided in the same way in the subspace, and finally displayed in graphs; graph-based technology is to distribute the data in graphs such as points, lines, and surfaces, mainly for data with strong structural relationships, as shown in Fig. 2.

Fig. 2

Indicator big data visualization model design.

Quality traceability is very important for enterprises, especially manufacturing enterprises, through quality traceability, it can diagnose where there are bottlenecks in the production process, where there are abnormalities in the production process and detect whether there are defects in the processing process of the products, etc. Through the continuous follow-up and optimization of these problems, production efficiency can be improved and the quality of the products can be improved continuously to improve customer satisfaction. The system collects the key data of the production of the equipment, so it can associate the data with the current production of the product, and when it is necessary to trace the quality of the product, it only needs to statistic, analyze and display the associated production data, so that it can complete the quality trace function intuitively and conveniently. For example, some enterprises have their systems of enterprise resource management, planning and scheduling, warehouse coordination, and so on, which can be connected to these systems to trace back the data to other supply chain data. The data of the supply chain such as orders, sales, and coordination of products are bound to realize the complete quality traceability system of products.

For the refined oil production enterprises, their main procurement costs come from upstream suppliers for the pricing of raw materials and wages of employees and factors paid by the production chain input, and their income mainly comes from the sale of products at the purchase price provided by distributors. “Zero-sum transaction", the new concept and the supplier is no longer just a high price to sell out to achieve an increase in profit margins but requires a high production capacity and efficiency, strong market flexibility, excellent product quality, and high inventory turnover rate. At the same time, the high degree of homogeneity of refined oil products also determines that the purpose of refined oil manufacturers should not be to sell more commodities to distributors but to cooperate with distributors to sell more homogeneous commodities through differentiated services from the perspective of customer concerns about oil quality and distribution efficiency, and to maximize the profits of both companies.

Finished oil producers are the foundation of the value network, focusing on increasing production efforts in segments and areas where they have advantages, discarding segments where they are not good at and consume a lot of energy and resources, and simplifying internal organization and processes with a focus on improving customer satisfaction. Integrate internal logistics with external logistics under a unified command and scheduling. Also, the company’s logistics coordination and management level are improved, in other words, it is equipped with the ability to conduct logistics management training and to use information technology to improve the efficiency of communication with customers. It also improves user stickiness and provides access to data assets that can be used to improve service levels. The members of the network focus on their advantageous links and areas, through collaboration to achieve complementarily, promote the virtuous operation of the value network. Coordination platform enterprises use information means to supervise the whole process, to empower logistics service providers and refined oil producers with information, and to improve customer service levels as a whole. To solve the problem of information asymmetry in the supply chain, the enterprises, customers, and suppliers involved in the supply chain are integrated into their networks, and by providing customers with lower cost, continuous and full service, they can improve their network value and service level, thus improving the competitiveness of the participants.

For the problem that vehicle driving safety and cargo safety cannot be supervised in the transportation process, the platform developed the refined oil coordination in-transit safety supervision system by cooperating with the driving safety intelligent mobile terminal equipment provider. The system monitors the driver’s unsafe behavior in driving through the vehicle camera, vehicle tanker liquid level monitoring device, combined with real-time artificial intelligence recognition technology, and alerts the alarm in real-time. At the same time, real-time data and images will be uploaded to the platform in time during the transportation process, and the cargo owner can monitor the driver’s behavior and cargo condition throughout the whole process. This system fundamentally solves the risk of transportation and cargo security during the transportation of refined oil products.

Through the user for stored-value card business information provided by the retail enterprises can grasp the user’s name, gender, age, unit, address, and other detailed data content: according to the transaction POS, to grasp the user’s type of oil, promotional programs, discounts, payment methods, and other content information: if the enterprise will have its massive user data, business data and other data combined to establish a shared datacenter, using big data analysis and mining and other new technologies to target the data resources to help enterprises gain an in-depth understanding and insight into the needs of users, and according to the needs of the targeted development of accurate marketing programs, tailor-made personalized and precise customer service, to maximize market value.

2.3 Evaluation system indicator design

Data integration is the primary problem that needs to be solved in big data precision marketing. Scattered fragments of data is a challenge that needs to be overcome on the way to realize Big Data precision marketing. How to integrate these scattered databases and independent data processing methods and achieve common technology and data between departments is the basic condition for maximizing data value and achieving precision marketing. Enterprises can build a big data exchange and sharing platform to integrate the data of their information systems, and at the same time through the data generated by user behavior, help enterprises to dig deep into user behavior, understanding of user behavior and other data values of user data, on the one hand, to track and analyze user behavior needs, while real-time monitoring and optimization of the operating situation and marketing results to enhance: on the one hand, it can track and analyze user behavior and demand, and at the same time monitor and optimize operations and marketing results in real-time. On the other hand, it can help enterprises analyze data at a later stage, reactivate and analyze the sedimented underlying user transaction data, enhance the value of neglected data, and provide a basic data foundation and technical analysis platform for more accurate user demand services and targeted personalized marketing strategies, as shown in Fig. 3.

Fig. 3

Evaluation System Framework.

Establish a big data visualization system platform, through data visualization ideas and tools to display complex big data correlation analysis results, support access to a variety of data sources, while supporting third-party interface data, such as text data, traditional database data, web data, and other data sources: support for data mining algorithms, predictive analytics, and other algorithms, as well as high-quality data management. The flexible use of big data visualization technology can summarize and display the data of different business lines through the implied relationship between the data, and through the overall analysis and summary of the summarized business data, it can help enterprises to mine the data to cover the business laws and business characteristics, which greatly improves the ability of big data to assist precision marketing.

Retail enterprises based on user segmentation and precision marketing are different from traditional market segmentation and analysis. They can carry out precision marketing according to users’ needs, consumption habits, behavior, and other requirements, and then carry out market segmentation according to their analysis, which requires information data collection of explicit and implicit aspects of customers, and through the analysis of big data mining tools, they can come up with a complete view of users, and then analyze the deep mining. Target market positioning, precision marketing for operators. Sales enterprises have accumulated a large amount of user information in the course of their daily operations, and with the help of big data analysis platform technology to conduct in-depth mining and analysis of their user data, they can obtain the common characteristics and demand orientation of different consumers and personalize and tailor their products. At the same time, it can also reduce the communication distance and frequency between enterprises and users to achieve one-to-one accurate service. With the rapid development of the mobile Internet, communication channels have become diverse, saving communication time and enhancing communication efficiency. To enable users to get good user experience and also to achieve the best marketing results, and to enable enterprises to track and analyze the entire marketing process, marketing methods from the traditional mass business broadcast push slowly towards one-to-one user needs-centered precision marketing transition.

The company’s main products are refined oil, lubricating oil, fuel oil, and other chemical products, with a secondary business of tobacco, alcohol and other daily necessities, auto parts and agricultural materials, housing, and mechanical equipment leasing. After nearly 10 years of development, the scale of the company has been growing. Market share has grown from small to large, competitive strength has grown from weak to strong, and the company has initially achieved the goal of having one out of three markets. The control variables shipment size is significantly negatively correlated with on-time performance and responsiveness to coordination orders, reflecting, in part, the lower level of the perceived value of coordination service provision by shipper firms with larger shipment sizes. The control variables shipment frequency is significantly positively correlated with coordination order punctuality and logistics order response speed, with a significance level of 0.1, reflecting that the higher the shipment frequency, the higher the level of perceived utility of logistics services by the shipper firms.

3 Analysis of results

3.1 System performance and effect optimization analysis

Big data visualization has high requirements for rendering performance, and the system provides good support for big data visualization by comparing WEBGL with MapV and choosing the better way to improve the rendering performance. This paper visualizes the test data and compares the rendering performance of these two methods. This paper uses the Chrome tool for testing, selects the performance option, and calculates the time from the input URL to the completion of all rendering of the data points as a measure. The results are averaged over seven refreshes for each data set, with the highest and lowest values removed, as shown in Fig. 4.

Fig. 4

Comparison of rendering performance between the two implementations.

From the line graph, we can see the rendering performance of WEBGL and MapV for a different amount of data points. The rendering completion time of WEBGL and MapBox at 1 million data points is 7.238 s, which is a good performance. The rendering performance of MapV is poor, and when the amount of data increases to 1 million data points, the rendering completion time is already more than 25 s, which is unbearable for users.

When the data volume reaches 1 million, the visualization of WEBGL and MapV is shown in Fig. 5. WebGL draws a map with MapBox technology and uses the height of the column to represent the distribution of the data, so users can see the data distribution clearly, while MapV’s visualization has a lot of overlapping and overlapping points, which cannot be displayed.

Fig. 5

1 million point visualization rendering.

Mapa uses canvas drawing, while WEBGL uses GPU rendering to achieve hardware acceleration, and combined with MapBox, it provides good support for big data rendering. The visualization of big data on the page has high requirements for the rendering ability of large amounts of data. In this paper, through the above tests, we compare the two perspectives of rendering completion time and visualization effect and find that WEBGL and MapBox have good support for the visualization of big data.

The operation of the DOM will take a lot of time and is a bottleneck of the browser’s performance, for this reason, the system chooses to use the React framework. There is no direct operation of the real DOM using the ref method, this also follows the idea of React. Reduce HTTP requests in the process of sending a request to receive a response takes more time, and this is related to the network speed at the time, so try to reduce HTTP requests can also improve browser performance. The system will not send the request again for duplicate data, or data that can be calculated from the acquired data, such as the domestic attack analysis module of this system, which only acquires data through a single HTTP request, and then analyzes and processes it to form a data attack graph, bar graph, pie chart, parallel coordinate graph, and so on.

Multidimensional data is of great research value as common data in life and production. The visualization of multidimensional data, as an important means of display and analysis, can help people intuitively obtain information and analyze data patterns. In this paper, based on the background of anomalous traffic attacks to realize the multidimensional data display, the geometry-based parallel coordinate visualization method is selected, and it is compared with the radar map visualization. The parallel coordinate visualization method breaks the spatial limitation of the Cartesian coordinate system, using a set of parallel axes to represent the dimension of the data, mapping the data points to a polyline in the view. The parallel coordinate visualization method can not only show the multiple dimensions of the data but also show the relationship between adjacent attributes, this paper combines the “brush” technology to achieve the parallel coordinate visualization of multi-dimensional attack data. Radar diagrams are often used to visualize multi-dimensional data, starting from the center of the axis of the data dimension values, data points can be mapped to multiple polylines enclosed in a graph. This paper uses test data to achieve visualization with different amounts of data and different dimensions. As can be seen in Fig. 6, when there are more data dimensions and the amount of data is relatively large, the readability of the radar graph deteriorates and the different polygons are repeatedly superimposed on each other, making it impossible to see the values under a specific dimension.

Fig. 6

300 30-dimensional of data visualization.

The visualization effect of parallel coordinate visualization technology is shown in the figure below, you can see that when the test data increases to 300, data dimension is 60, you can still see the details of each dimension on the parallel coordinate. In this paper, by using the “brush” technology, the user can display the dimension he wants to observe, and compress the other dimensions to display, to reduce the screen space occupation, and comprehensively compare the visualization effect of multi-dimensional data, the system chooses to parallel coordinate visualization technology, as shown in Fig. 7.

Fig. 7

100 30-dimensional of data visualization.

Data visualization requires the processing of the acquired data, which is part of the visualization process. This system is based on the background of anomalous traffic attacks to achieve visualization, which needs to show the location of the attack, and the location of the attacked, which involves converting the IP address of the data into the corresponding latitude and longitude, after research, this can be implemented in the front-end or the server-side implementation.

3.2 Analysis of the results of testing indicators

Precision marketing not only refers to the customer’s promotional policy for marketing, while the analysis of customer consumption habits can also bring benefits and value to the enterprise. The gas station for scientific business hours and sales of the change of the relationship between the measurement, targeted implementation of gas station intermittent business and personnel scheduling, can also be achieved to enhance the effectiveness of enterprises. Through big data platform analysis, the business hours, business volume, the number of times to mention the gun, and other factors associated with the analysis of remote provincial roads, township stations at night to mention the number of times to mention the gun 25% of the whole day or the number of times to mention the gun at night to 90 times In the history of the Evergreen gas station, a fixed three-shift, the two-shift system was adopted, with four additional mobile fuel attendants added to the long day shift, making a total of six people refueling on site. Combined with weekly consumption and daily consumption frequency, it is found that human resources are evenly distributed and labor is unevenly shared, with insufficient workforce during the peak period and abundant workforce during the trough period, affecting work efficiency. According to the business volume during the business hours, the gas station broke the fixed three-shift two-shift system and set up a flexible team to increase the average daily sales from 26.4 tons to 31.95 tons, while keeping the total number of employees unchanged, as shown in Fig. 8.

Fig. 8

Finished oil purchase results.

As the price of gasoline was close to the trend of international oil prices, the overall V-shaped oscillation adjustment.

And thus, the selling company’s gasoline prices were also in a period of oscillation, as shown in Fig. 9, the actual selling price of gasoline at the end of the year compared to the beginning of the year fell by 526.

Fig. 9

Price Trend of Gasoline Direct Sales.

With the continuous application of new technologies, some new Internet companies have crossed the border into the already competitive retail market for refined petroleum products under technological innovation and wedge innovation, bringing in the fresh air and intensifying market competition at the same time, as shown in Fig. 10.

Fig. 10

Small passenger vehicles and private car ownership.

The use of artificial intelligence technology to further optimize supply chain management, asset and equipment management, and gas station safety management, to build intelligent gas stations, to achieve intelligent management of equipment and refined oil. With the centralized operation control system, we can monitor and control the equipment, oil incoming, and outgoing inventory and other data of many gas stations, realize early warning of equipment failure, analyze the cause of failure, remind maintenance, remind and control the temperature and pressure of refined oil, improve management efficiency and ensure the safety of gas stations. To cope with the internal and external competition and challenges, each major refined oil sales enterprises will take active or passive measures to transform. Automation and intelligence will become the inevitable development trend, the innovation of new models will continue, big data application scenarios will be more abundant, and traditional gas stations will incorporate more innovative technologies and new retail genes.

4 Conclusion

This paper is based on fuzzy mathematics for the quantification of influential factors in the refined oil market and the visualization of big data of key monitoring indicators, this paper establishes an equipment selection model for the quantification of influential factors based on fuzzy mathematical theory, analyzes the internal and external factors in the process of quantification of influential factors. The system collects and visualizes the production data of the enterprise without any change to the refined oil market, realizes the production transparency of the enterprise to the maximum extent, and helps the enterprise to realize the quality traceability system of the refined oil market. The system achieves the storage of key monitoring indicators of the refined oil market through the use of a time-series database, the forwarding of key monitoring indicators of the refined oil market through the use of message middleware, and the visualization of key monitoring indicators of refined oil market through some modern graphical frameworks and components, and at the same time, as well as product quality traceability through electronic tagging and other technologies. The visualization system can quickly perceive the workshop site data, order task progress, machine equipment alarm information, and other data, through these information data will establish a virtual factory, the enterprise’s production status becomes transparent, help improve efficiency.

References

Chehbi-Gamoura

, Derrouiche

, Damand

, et al., Insights from big Data Analytics in supply chain management: an all-inclusive literature review using the SCOR model, Production Planning & Control 31(5) (2020), 355–382.

Litvinenko

V.S.

, Digital economy as a factor in the technological development of the mineral sector, Natural Resources Research 29(3) (2020), 1521–1541.

Chatterjee

, Hossain

S.A.

and Kar

, Prioritization of project proposals in portfolio management using fuzzy AHP, Opsearch 55(2) (2018), 478–501.

Thöni

, Taudes

and Tjoa

A.M.

, An information system for assessing the likelihood of child labor in supplier locations leveraging Bayesian networks and text mining, Information Systems and e-Business Management 16(2) (2018), 443–476.

Saidani

, Kendall

, Yannou

, et al., Closing the loop on platinum from catalytic converters: Contributions from material flow analysis and circularity indicators, Journal of Industrial Ecology 23(5) (2019), 1143–1158.

Noman

M.A.

, Nasr

E.S.A.

, Al-Shayea

, et al., Overview of predictive condition based maintenance research using bibliometric indicators, Journal of King Saud University-Engineering Sciences 31(4) (2019), 355–367.

Sharma

S.K.

, Sengupta

and Panja

S.C.

, Mapping corruption risks in public procurement: Uncovering improvement opportunities and strengthening controls, Public Performance & Management Review 42(4) (2019), 947–975.

Yang

, Feng

, Liu

, et al., Research advances on process systems integration and process safety in China, Reviews in Chemical Engineering 36(1) (2019), 147–185.

Reuter

M.A.

, Digitalizing the circular economy, Metallurgical and Materials transactions B 47(6) (2016), 3194–3220.

10.

Vogl

G.W.

, Weiss

B.A.

and Helu

, A review of diagnostic and prognostic capabilities and best practices for manufacturing, Journal of Intelligent Manufacturing 30(1) (2019), 79–95.

11.

Singh

R.K.

, Modgil

and Acharya

, Assessment of supply chain flexibility using system dynamics modeling, Global Journal of Flexible Systems Management 20(1) (2019), 39–63.

12.

K.H.

, Chen

F.H.

and Tzeng

G.H.

, CPA firm’s cloud auditing provider for performance evaluation and improvement: an empirical case of China, Technological and Economic Development of Economy 24(6) (2018), 2338–2373.

13.

Mura

, Longo

, Micheli

, et al., The evolution of sustainability measurement research, International Journal of Management Reviews 20(3) (2018), 661–695.

14.

Okoro

and Kolios

, Multicriteria risk assessment framework for components’ risk ranking: Case study of a complex oil and gas support structure, Journal of Multi-Criteria Decision Analysis 25(5-6) (2018), 113–129.

15.

Attri

, Interpretive structural modelling: a comprehensive literature review on applications, International Journal of Six Sigma and Competitive Advantage 10(3-4) (2017), 258–331.

16.

Jaffee

E.M.

, Van Dang

, Agus

D.B.

, et al., Future cancer research priorities in the USA: a Lancet Oncology Commission, The Lancet Oncology 18(11) (2017), e653–e706.

17.

Bugert

and Lasch

, Supply chain disruption models: A critical review, Logistics Research 11(5) (2018), 1–35.

18.

Xie

, Tang

, Huang

, et al., A survey of blockchain technology applied to smart cities: Research issues and challenges, IEEE Communications Surveys & Tutorials 21(3) (2019), 2794–2830.

19.

Wang

, Wang

, Zhang

, et al., Crowd-powered sensing and actuation in smart cities: Current issues and future directions, IEEE Wireless Communications 26(2) (2019), 86–92.

20.

Hayashi

, Fujimoto

, Ishii

, et al., Versatile modeling platform for cooperative energy management systems in smart cities, Proceedings of the IEEE 106(4) (2018), 594–612.

21.

Gharaibeh

, Salahuddin

M.A.

, Hussini

S.J.

, et al., Smart cities: A survey on data management, security, and enabling technologies, IEEE Communications Surveys & Tutorials 19(4) (2017), 2456–2501.