Expressway service area consumption behavior simulation system based on perspective of road derivation economy

Abstract

The invention relates to the technical field of consumption behavior simulation, in particular to an expressway service area consumption behavior simulation system based on the perspective of road derivation economy. The system comprises a variable analysis unit, a simulation calculation unit, a trend time unit and a commodity recommendation unit. The simulation calculation unit simulates the consumption amount for multiple times according to the consumption behavior variable, the independent variable data, the model parameter and the error item data by using the consumption behavior model, and evaluates the mean square error of the model performance according to the actual consumption amount of the service area and the simulated consumption amount. And calculating a simulated average error according to the evaluated model performance mean square error, selecting the simulated consumption amount corresponding to the minimum simulated average error as the final consumption amount, and calculating the simulated average error in the process of multiple times of simulation and selecting the minimum simulated average error to obtain the final consumption amount. Therefore, the accumulation of the mean square error of the model performance can be reduced in the process of multi-time simulation, so that the reliability of the whole simulation of the model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of consumer behavior simulation, and in particular to a highway service area consumer behavior simulation system based on the perspective of road-derived economy. Background Art

[0002] The simulation of consumer behavior in highway service areas from the perspective of road-derived economics is a complex but extremely practical tool. Through scientific models and simulation techniques, it helps us better understand and predict consumer shopping behavior, optimize the operation and management of service areas, and ultimately promote the healthy development of the regional economy. With the continuous advancement of technology and the accumulation of data, it can be more accurate and intelligent, providing strong support for the sustainable development of highway service areas. Since consumers have different consumption behaviors in highway service areas, such as different consumption amounts, when a single consumption behavior is simulated, the simulation results will be affected by unbalanced data, which will reduce the accuracy of the simulation. Multiple simulations are used to simulate consumer behavior multiple times, but since a large number of calculation steps are involved in the process of multiple simulations, the mean square error of the model performance in the calculation process may gradually accumulate, resulting in a decrease in the reliability of the overall simulation of the model. Therefore, we provide a highway service area consumer behavior simulation system based on the perspective of road-derived economics. Summary of the invention

[0003] The purpose of the present invention is to provide a highway service area consumer behavior simulation system based on the perspective of road-derived economy to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides a highway service area consumer behavior simulation system based on the perspective of road-derived economy, including an analysis variable unit, a simulation calculation unit, a trend time unit, and a commodity recommendation unit;

[0005] The analysis variable unit obtains the traffic flow of highways, the types of goods in service areas, the completeness of service area facilities, and historical data through multiple channels, and uses logistic regression to establish a consumer behavior model based on the acquired data, and analyzes the key influencing factors of the acquired data from the perspective of road-derived economics, and then outputs the model parameters and error terms, regards the analyzed key influencing factor data as independent variables and calculates the consumer behavior variables;

[0006] The simulation calculation unit is used to receive the consumption behavior variables, independent variable data, model parameters, error term data, consumption behavior model, and historical data in the analysis variable unit, and perform multiple simulations of the consumption amount, divide the historical service area consumption amount data set into k subsets using K-fold cross validation, evaluate the mean square error of the model performance according to the simulated consumption amount, and then calculate the average error of the simulation according to the evaluated model performance mean square error, compare the average errors of the calculated simulations in turn, and select the simulated consumption amount corresponding to the minimum average error of the calculated simulation as the final consumption amount;

[0007] The trend time unit is used to receive the final consumption amount in the simulation calculation unit and the historical data and consumption behavior model in the analysis variable unit, and use the historical data and the final consumption amount to predict the consumer's consumption amount trend and the consumer's consumption frequency trend, and then predict the consumer's stay time in the highway service area;

[0008] The product recommendation unit is used to receive the historical data in the analysis variable unit and the predicted consumer stay time in the trend time unit, analyze the products that consumers like to consume in the highway service area based on the historical data, and then recommend the preferred products to consumers within the predicted consumer stay time based on the analyzed preferred products.

[0009] As a further improvement of the technical solution, the analysis variable unit includes a model analysis module and a calculation variable module;

[0010] The model analysis module obtains the traffic flow of the expressway, the types of goods in the service area, the data on the completeness of the service area facilities, the actual consumption amount in the service area, and the historical data through multiple channels, and uses logistic regression to establish a consumption behavior model based on the obtained traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities. The obtained traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities data are then input into the consumption behavior model. The consumption behavior model analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities data from the perspective of road-derived economics, analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities, and outputs the model parameters corresponding to the analyzed key influencing factor data, as well as the error terms generated between the model parameters;

[0011] The variable calculation module is used to receive the key influencing factor data, model parameters, and error term data of the highway traffic flow, service area commodity types, and service area facility perfection analyzed in the model analysis module, regard the analyzed highway traffic flow, service area commodity types, and service area facility perfection data as independent variables, and use the consumer behavior model to calculate the consumer behavior variables based on the independent variables, model parameters, and error term data.

[0012] As a further improvement of the present technical solution, the simulation calculation unit is used to receive the consumption behavior variables, independent variable data, model parameters, error term data in the calculation variable module and the consumption behavior model, the historical service area consumption amount data set, and the actual consumption amount of the service area in the model analysis module, and input the consumption behavior variables into the consumption behavior model. The consumption behavior model outputs the model parameter β4 corresponding to the consumption behavior variable, and then uses the consumption behavior model to perform multiple simulations of the consumption amount according to the consumption behavior variables, independent variable data, model parameters β0, β1, β2, β3 and the corresponding model parameter β4 and error term data, and records the number of simulations. The historical service area consumption amount data set is divided into subsets using K-fold cross validation, and the number of divided subsets is recorded. The mean square error of the model performance is evaluated according to the actual consumption amount of the service area, the simulated consumption amount and the number of divided subsets, and then the average error of the simulation is calculated according to the mean square error of the model performance evaluated for the i-th time and the number of simulations. The average errors of the calculated simulations are compared in turn, and the simulated consumption amount corresponding to the minimum average error of the calculated simulation is selected as the final consumption amount.

[0013] As a further improvement of the technical solution, the implementation principle of simulating the consumption amount multiple times using the consumption behavior model in the simulation calculation unit is:

[0014] First, collect the consumption behavior variable y, independent variable data, model parameters β0, β1, β2, β3 and the corresponding model parameter β4, error term data ε, simulate the consumption amount multiple times, and obtain the simulated consumption amount Ss. The specific algorithm formula is:

[0015] Ss=β0+β1X1+β2X2+β3X3+β4y+ε.

[0016] As a further improvement of the technical solution, the mean square error implementation principle for evaluating model performance in the simulation calculation unit is:

[0017] First, the actual consumption amount Aa in the service area, the simulated consumption amount Ss and the number of divided subsets k are collected to evaluate the mean square error of the model performance, and the mean square error Me of the evaluated model performance is obtained. The specific algorithm formula is:

[0018]

[0019] Among them, SS i Refers to the consumption amount Ss, Me of the i-th simulation i Refers to the mean square error Me of the model performance evaluated for the i-th time.

[0020] As a further improvement of the technical solution, the simulation calculation unit calculates the average error of the simulation to achieve the following steps:

[0021] Collect and evaluate the mean square error Me of the model performance for the i-th time and the number of simulations n to calculate the average error of the simulation, and obtain the calculated average error An. The specific algorithm formula is:

[0022]

[0023] Among them, An i Refers to the average error An of the i-th calculation simulation.

[0024] As a further improvement of the technical solution, the trend time unit includes a prediction trend module and a residence time module

[0025] The prediction trend module is used to receive the final consumption amount in the simulation calculation unit and the actual consumption amount in the service area, historical data, and consumption behavior model in the model analysis module, take the historical service area consumption amount data set and the historical service area consumption frequency from the historical data, use the consumption behavior model to predict the consumer consumption amount trend according to the actual consumption amount in the service area, the historical service area consumption amount data set and the final consumption amount, and then use the consumption behavior model to predict the consumer consumption frequency trend according to the historical service area consumption frequency;

[0026] The residence time module is used to receive the predicted consumption amount trend, the predicted consumption frequency trend in the prediction trend module and the consumption behavior model in the model analysis module, and use the consumption behavior model to make predictions based on the predicted consumption amount trend and the predicted consumption frequency trend to predict the consumer's residence time in the highway service area.

[0027] As a further improvement of the present technical solution, the product recommendation unit is used to receive the historical service area consumption products in the model analysis module and the predicted consumer stay time in the stay time module, and use association rule mining to analyze the products that consumers like to consume in the highway service area based on the historical service area consumption products, and then use recommendation logic to recommend the consumer's favorite products to the consumer within the predicted consumer stay time based on the analyzed favorite consumption products.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] In the highway service area consumption behavior simulation system based on the road-derived economic perspective, the simulation calculation unit inputs the consumption behavior variables into the consumption behavior model, and the consumption behavior model outputs the model parameters corresponding to the consumption behavior variables. The consumption behavior model is then used to simulate the consumption amount multiple times according to the consumption behavior variables, independent variable data, model parameters, and error term data, and the number of simulations is recorded. The historical service area consumption amount data set is divided into k subsets, and the number of divided subsets is recorded. The mean square error of the model performance is evaluated according to the actual consumption amount of the service area, the simulated consumption amount, and the number of divided subsets. The average error of the simulation is then calculated according to the evaluated model performance mean square error and the number of simulations. The average errors of the calculated simulations are compared in turn, and the simulated consumption amount corresponding to the minimum average error of the calculated simulation is selected as the final consumption amount. By simulating the consumption amount multiple times, it is possible to avoid the single simulation result being affected by unbalanced data, thereby improving the accuracy of multiple simulations. At the same time, the average error of the simulation is calculated and the minimum average error of the simulation is selected during the multiple simulations, which can avoid reducing the accumulation of the mean square error of the model performance during the multiple simulations, thereby improving the reliability of the overall simulation of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A unit block diagram of the present invention;

[0031] Figure 2 It is a block diagram of a module unit of the present invention;

[0032] Figure 3 It is a block diagram of the analysis variable unit of the present invention;

[0033] Figure 4 It is a block diagram of the trend time unit of the present invention.

[0034] The meaning of each number in the figure is:

[0035] 1. Analysis variable unit; 11. Model analysis module; 12. Calculation variable module; 2. Simulation calculation unit;

[0036] 3. Trend time unit; 31. Prediction trend module; 32. Dwell time module; 4. Product recommendation unit. DETAILED DESCRIPTION

[0037] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] The present invention provides a highway service area consumer behavior simulation system based on the perspective of road-derived economy. Figure 1-Figure 4 , including analysis variable unit 1, simulation calculation unit 2, trend time unit 3, commodity recommendation unit 4;

[0039] The analysis variable unit 1 obtains the data of highway traffic flow, types of goods in service areas, and the completeness of service area facilities through multiple channels, and obtains historical data, uses logistic regression to establish a consumer behavior model based on the obtained data, and analyzes the key influencing factors of the obtained data from the perspective of road-derived economics, and then outputs the model parameters and error terms, regards the analyzed key influencing factor data as independent variables and calculates the consumer behavior variables;

[0040] The simulation calculation unit 2 is used to receive the consumption behavior variables, independent variable data, model parameters, error term data, consumption behavior model, and historical data in the analysis variable unit 1, and perform multiple simulations of the consumption amount, divide the historical service area consumption amount data set into k subsets using K-fold cross validation, evaluate the mean square error of the model performance according to the simulated consumption amount, and then calculate the average error of the simulation according to the evaluated model performance mean square error, compare the average errors of the calculated simulations in turn, and select the simulated consumption amount corresponding to the minimum average error of the calculated simulation as the final consumption amount;

[0041] The trend time unit 3 is used to receive the final consumption amount in the simulation calculation unit 2 and the historical data and consumption behavior model in the analysis variable unit 1, and use the historical data and the final consumption amount to predict the consumer's consumption amount trend and the consumer's consumption frequency trend, and then predict the consumer's stay time in the highway service area;

[0042] The product recommendation unit 4 is used to receive the historical data in the analysis variable unit 1 and the predicted consumer residence time in the trend time unit 3, analyze the products that consumers like to consume in the highway service area based on the historical data, and then recommend the preferred products to consumers within the predicted consumer residence time based on the analyzed preferred consumption products.

[0043] The analysis variable unit 1 includes a model analysis module 11 and a calculation variable module 12;

[0044] The model analysis module 11 obtains the traffic flow of the expressway, the types of goods in the service area, the data on the completeness of the service area facilities, the actual consumption amount Aa of the service area, and the historical data through multiple channels, and uses logistic regression to establish a consumption behavior model based on the obtained data on the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities, and then inputs the obtained data on the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities into the consumption behavior model. The consumption behavior model analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities from the perspective of road-derived economics, analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities, and outputs the model parameters corresponding to the analyzed key influencing factor data, as well as the error terms generated between the model parameters;

[0045] Historical data includes historical service area consumption amount dataset, historical service area consumption frequency, and historical service area consumption commodities;

[0046] The variable calculation module 12 is used to receive the key influencing factor data, model parameters, and error term data of the highway traffic flow, service area commodity types, and service area facility perfection analyzed in the model analysis module 11, and regard the analyzed highway traffic flow, service area commodity types, and service area facility perfection data as independent variables, and use the consumer behavior model to calculate the consumer behavior variables based on the independent variables, model parameters, and error term data.

[0047] Steps to calculate consumption behavior variables:

[0048] Step ①, first collect independent variables, model parameters, and error term data ε. The independent variables include the key influencing factor data X1 of the analyzed highway traffic flow, the key influencing factor data X2 of the analyzed service area commodity types, and the key influencing factor data X3 of the analyzed service area facility perfection. The model parameters include β0, β1, β2, and β3.

[0049] Step ②, calculate the consumer behavior variable y using the analyzed highway traffic flow key influencing factor data X1, the analyzed service area commodity type key influencing factor data X2 and the service area facility completeness key influencing factor data X3, model parameters and error term data ε. The specific algorithm formula is:

[0050] y=β0+β1X1+β2X2+β3X3+ε;

[0051] Among them, this formula is used to calculate consumer behavior variables. As the independent variables and external environment change, the model can dynamically adjust the prediction results and provide real-time consumer behavior prediction;

[0052] The simulation calculation unit 2 is used to receive the consumption behavior variables, independent variable data, model parameters, error term data in the calculation variable module 12 and the consumption behavior model, the historical service area consumption amount data set, and the actual consumption amount Aa of the service area in the model analysis module 11, input the consumption behavior variables into the consumption behavior model, and the consumption behavior model outputs the model parameter β4 corresponding to the consumption behavior variable, and then use the consumption behavior model to simulate the consumption amount multiple times according to the consumption behavior variable y, the independent variable data, the model parameters β0, β1, β2, β3 and the corresponding model parameter β4, and the error term data ε, record the number of simulations n, use K-fold cross validation to divide the historical service area consumption amount data set into k subsets, record the number of divided subsets k, and then use the K-fold cross validation to divide the historical service area consumption amount data set into k subsets. The mean square error of the model performance is evaluated by the actual consumption amount Aa in the service area, the simulated consumption amount Ss and the number of divided subsets k. The average error of the simulation is calculated based on the mean square error Me of the model performance of the ith evaluation and the number of simulations n. The calculated average errors of the simulations are compared in turn, and the simulated consumption amount corresponding to the minimum calculated average error of the simulation is selected as the final consumption amount. By simulating the consumption amount multiple times, it is possible to avoid the impact of unbalanced data on a single simulation result, thereby improving the accuracy of multiple simulations. At the same time, calculating the average error of the simulation and selecting the minimum average error of the simulation during multiple simulations can avoid reducing the accumulation of the mean square error of the model performance during multiple simulations, thereby improving the reliability of the overall simulation of the model.

[0053] The implementation principle of using the consumption behavior model to simulate the consumption amount multiple times in simulation calculation unit 2 is:

[0054] First, collect the consumption behavior variable y, the independent variable data including the key influencing factor data X1 of the analyzed highway traffic flow, the key influencing factor data X2 of the analyzed service area commodity types and the key influencing factor data X3 of the service area facility perfection, the model parameters β0, β1, β2, β3 and the corresponding model parameter β4, the error term data ε, and simulate the consumption amount many times to obtain the simulated consumption amount Ss. The specific algorithm formula is:

[0055] Ss=β0+β1X1+β2X2+β3X3+β4y+ε;

[0056] The formula is used to simulate the consumption amount multiple times. Multiple simulations can enhance the model's resistance to input changes and data noise. No matter how the external environment or data conditions change, the model can maintain stable prediction capabilities.

[0057] The implementation principle of mean square error for evaluating model performance in simulation calculation unit 2:

[0058] First, the actual consumption amount Aa in the service area, the simulated consumption amount Ss and the number of divided subsets k are collected to evaluate the mean square error of the model performance, and the mean square error Me of the evaluated model performance is obtained. The specific algorithm formula is:

[0059]

[0060] Among them, SS i Refers to the consumption amount Ss, Me of the i-th simulation i It refers to the mean square error Me of the model performance evaluated for the i-th time. This formula is used to evaluate the mean square error of model performance. By evaluating the mean square error of model performance, we can identify possible systematic deviations in the model. When the mean square error of a specific subset is significantly higher than that of other subsets, it means that the model has weaker predictive ability in this specific situation, which helps to optimize the model and adapt it to specific consumption patterns.

[0061] The steps for calculating the average error of the simulation in simulation calculation unit 2 are as follows:

[0062] Collect and evaluate the mean square error Me of the model performance for the i-th time and the number of simulations n to calculate the average error of the simulation, and obtain the calculated average error An. The specific algorithm formula is:

[0063]

[0064] Among them, An i Refers to the average error An of the i-th calculation simulation. This formula is used to calculate the average error of the simulation. By calculating the average error of the simulation and selecting the smallest average error of the simulation, the accumulation of the mean square error of the model performance can be avoided during multiple simulations, thereby improving the reliability of the overall simulation of the model.

[0065] The trend time unit 3 includes a trend prediction module 31 and a residence time module 32

[0066] The prediction trend module 31 is used to receive the final consumption amount in the simulation calculation unit 2 and the actual consumption amount Aa of the service area, historical data, and consumption behavior model in the model analysis module 11, and take the historical service area consumption amount data set and the historical service area consumption frequency from the historical data, and use the consumption behavior model to predict the consumer consumption amount trend according to the actual consumption amount Aa of the service area, the historical service area consumption amount data set and the final consumption amount. The consumer consumption amount trend data can help the service area to more accurately predict inventory demand, prepare best-selling products in advance, and avoid sales losses caused by out-of-stock. The consumption behavior model is then used to predict the consumer consumption frequency trend according to the historical service area consumption frequency. By predicting the consumption frequency trend, the service area can more accurately predict the demand for goods, reasonably arrange inventory and procurement plans, and avoid excess or shortage of goods. For example, it is expected that the consumption frequency of certain goods will increase during the upcoming holidays, and inventory can be increased in advance;

[0067] The residence time module 32 is used to receive the predicted consumption amount trend, the predicted consumption frequency trend and the consumption behavior model in the model analysis module 11 in the prediction trend module 31, and use the consumption behavior model to make predictions based on the predicted consumption amount trend and the predicted consumption frequency trend to predict the consumer's residence time in the highway service area. Through the predicted residence time, personalized services and recommendations are provided to different consumers to enhance the consumption experience.

[0068] The product recommendation unit 4 is used to receive the historical service area consumption products in the model analysis module 11 and the predicted consumer stay time in the stay time module 32, and use association rule mining to analyze the products that consumers like to consume in the highway service area based on the historical service area consumption products. By analyzing the consumers' favorite products, the service area can more accurately predict product demand, reduce inventory backlogs and unnecessary waste, and then use recommendation logic to recommend consumers' favorite products to consumers within the predicted consumer stay time based on the analyzed favorite consumption products. Consumers have limited stay time in the service area. By predicting the stay time and recommending their favorite products, the immediate needs of consumers can be quickly met and the purchase conversion rate can be improved.

[0069] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A highway service area consumer behavior simulation system based on the perspective of road-derived economy, characterized by: It includes an analysis variable unit (1), a simulation calculation unit (2), a trend time unit (3), and a product recommendation unit (4); The analysis variable unit (1) obtains the traffic flow, types of goods in the service area, and the completeness of service area facilities of the expressway through multiple channels, and obtains historical data, uses logistic regression to establish a consumer behavior model based on the obtained data, and analyzes the key influencing factors of the obtained data from the perspective of road-derived economics, then outputs the model parameters and error terms, regards the analyzed key influencing factor data as independent variables, and calculates the consumer behavior variables; The simulation calculation unit (2) is used to receive the consumption behavior variables, independent variable data, model parameters, error term data, consumption behavior model, and historical data in the analysis variable unit (1), and perform multiple simulations of consumption amounts, divide the historical service area consumption amount data set into k subsets using K-fold cross validation, evaluate the mean square error of the model performance based on the simulated consumption amounts, and then calculate the average error of the simulation based on the evaluated mean square error of the model performance, compare the average errors of the calculated simulations in turn, and select the simulated consumption amount corresponding to the minimum average error of the calculated simulations as the final consumption amount; The trend time unit (3) is used to receive the final consumption amount in the simulation calculation unit (2) and the historical data and consumption behavior model in the analysis variable unit (1), and use the historical data and the final consumption amount to predict the consumer's consumption amount trend and the consumer's consumption frequency trend, and then predict the consumer's stay time in the highway service area; The product recommendation unit (4) is used to receive the historical data in the analysis variable unit (1) and the predicted consumer stay time in the trend time unit (3), analyze the products that consumers like to consume in the highway service area based on the historical data, and then recommend the preferred products to the consumers within the predicted consumer stay time based on the analyzed preferred products.

2. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 1 is characterized by: The analysis variable unit (1) comprises a model analysis module (11) and a calculation variable module (12); The model analysis module (11) obtains the traffic flow of the expressway, the types of goods in the service area, the data on the completeness of the service area facilities, the actual consumption amount of the service area, and the historical data through multiple channels, and uses logistic regression to establish a consumption behavior model based on the obtained data on the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities, and then inputs the obtained data on the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities into the consumption behavior model. The consumption behavior model analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities from the perspective of road-derived economics, analyzes the key influencing factors of the traffic flow of the expressway, the types of goods in the service area, and the completeness of the service area facilities, and outputs the model parameters corresponding to the analyzed key influencing factor data, as well as the error terms generated between the model parameters; The calculation variable module (12) is used to receive the key influencing factor data of the traffic flow of the expressway, the types of goods in the service area and the completeness of the service area facilities analyzed in the model analysis module (11), the model parameters, and the error term data, and regard the analyzed key influencing factor data of the traffic flow of the expressway, the types of goods in the service area and the completeness of the service area facilities as independent variables, and use the consumer behavior model to calculate the consumer behavior variables according to the independent variables, the model parameters, and the error term data.

3. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 2 is characterized by: The simulation calculation unit (2) is used to receive the consumption behavior variables, independent variable data, model parameters, error term data in the calculation variable module (12) and the consumption behavior model, the historical service area consumption amount data set, and the actual service area consumption amount in the model analysis module (11), input the consumption behavior variables into the consumption behavior model, and the consumption behavior model outputs the model parameter β4 corresponding to the consumption behavior variables, and then use the consumption behavior model to simulate the consumption amount multiple times according to the consumption behavior variables, independent variable data, model parameters β0, β1, β2, β3 and the corresponding model parameter β4 and error term data, record the number of simulations, divide the historical service area consumption amount data set into subsets by using K-fold cross validation, record the number of divided subsets, evaluate the mean square error of the model performance according to the actual service area consumption amount, the simulated consumption amount and the number of divided subsets, and then calculate the average error of the simulation according to the mean square error of the model performance evaluated for the i-th time and the number of simulations, compare the average errors of the calculated simulations in turn, and select the simulated consumption amount corresponding to the minimum average error of the calculated simulations as the final consumption amount.

4. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 3 is characterized by: The implementation principle of the simulation calculation unit (2) using the consumption behavior model to simulate the consumption amount multiple times is as follows: First, collect consumption behavior variables y, independent variable data, model parameters β0, β1, β2, β3 and corresponding model parameters β4, error term data ε, simulate consumption amounts multiple times, and obtain the simulated consumption amount Ss. The specific algorithm formula is: Ss=β0+β1X1+β2X2+β3X3+β4y+ε.

5. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 3 is characterized by: The implementation principle of the mean square error for evaluating the model performance in the simulation calculation unit (2) is: First, the actual consumption amount Aa in the service area, the simulated consumption amount Ss and the number of divided subsets k are collected to evaluate the mean square error of the model performance, and the mean square error Me of the evaluated model performance is obtained. The specific algorithm formula is: Among them, SS i Refers to the consumption amount Ss, Me of the i-th simulation i Refers to the mean square error Me of the model performance evaluated for the i-th time.

6. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 3 is characterized by: The simulation calculation unit (2) calculates the average error of the simulation to achieve the following steps: Collect and evaluate the mean square error Me of the model performance for the i-th time and the number of simulations n to calculate the average error of the simulation, and obtain the calculated average error An. The specific algorithm formula is: Among them, An i Refers to the average error An of the i-th calculation simulation.

7. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 3 is characterized by: The trend time unit (3) includes a trend prediction module (31) and a residence time module (32) The prediction trend module (31) is used to receive the final consumption amount in the simulation calculation unit (2) and the actual consumption amount in the service area, historical data, and consumption behavior model in the model analysis module (11), take the historical service area consumption amount data set and the historical service area consumption frequency from the historical data, use the consumption behavior model to predict the consumer consumption amount trend based on the actual consumption amount in the service area, the historical service area consumption amount data set and the final consumption amount, and then use the consumption behavior model to predict the consumer consumption frequency trend based on the historical service area consumption frequency; The residence time module (32) is used to receive the consumption amount trend predicted by the prediction trend module (31), the predicted consumption frequency trend and the consumption behavior model in the model analysis module (11), and use the consumption behavior model to predict the consumer's residence time in the highway service area based on the predicted consumption amount trend and the predicted consumption frequency trend.

8. The highway service area consumer behavior simulation system based on the road-derived economic perspective according to claim 7 is characterized by: The product recommendation unit (4) is used to receive the historical service area consumption products in the model analysis module (11) and the predicted consumer stay time in the stay time module (32), use association rule mining to analyze the products that consumers like to consume in the highway service area based on the historical service area consumption products, and then use recommendation logic to recommend the consumer's favorite products to the consumer within the predicted consumer stay time based on the analyzed favorite consumption products.

Images