A method, device and storage medium for inventory prediction based on full automobile circulation big data
Through the inventory prediction method based on full-scale automobile circulation big data, combined with vehicle production, sales and on-the-go data, the LSTM model is used to predict future inventory cycles, which solves the shortcomings of the traditional inventory management methods in real-time and accuracy, and improves inventory management efficiency and competitiveness.
Patent Information
- Application Number
- CN202411225149.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-09-03
AI Technical Summary
Traditional inventory management methods are insufficient in real-time, comprehensiveness and accuracy, and cannot effectively respond to the rapidly changing needs of the automobile market.
The inventory prediction method based on full-scale automobile circulation big data is adopted. By obtaining the production, sales and on-the-go data of vehicles, combining prediction models, the current inventory cycle is calculated, and environmental factors and macro-control factors are used to predict the future inventory cycle. Memory network (LSTM) performs deep learning and outputs future inventory cycles.
It realizes accurate calculation and prediction of inventory cycles, improves the comprehensiveness and real-timeness of data, reduces inventory costs and market risks, and improves the operational efficiency and market competitiveness of automobile manufacturers and dealers.
Smart Images

Figure CN118761711B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of vehicle inventory technology, and in particular to an inventory prediction method, device and storage medium based on full automobile circulation big data. Background Art
[0002] In the automotive industry, inventory management has a significant impact on the operational efficiency and capital turnover of manufacturers, dealers and related companies.
[0003] Traditional inventory methods mainly rely on local statistics and decentralized collection of single vehicle entry and exit identification. This method lacks in real-time, comprehensiveness and accuracy, and cannot effectively respond to the rapidly changing needs of the automotive market.
[0004] In view of this, the present invention is proposed. Summary of the invention
[0005] In response to the above problems, the present invention proposes an inventory prediction method, device and storage medium based on the full automobile circulation big data. According to the full data of vehicle sales, production, and in-transit, combined with the prediction model, accurate inventory is obtained to effectively respond to the rapidly changing needs of the automobile market.
[0006] In a first aspect, an embodiment of the present invention provides an inventory prediction method based on full automobile circulation big data, comprising:
[0007] S110, obtaining the production date and sales date of multiple target vehicles, and determining the sales status or inventory status of each target vehicle;
[0008] S120: If the target vehicle in the sale state has positioning information and recorded mileage, determine whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle;
[0009] S130: if the target vehicle is in transit, re-determine the target vehicle as being in inventory;
[0010] S140, for each target vehicle in inventory status, calculating the current inventory cycle; if the current inventory cycle exceeds a set time, removing the target vehicle;
[0011] S150, for the target vehicles in the remaining inventory status, calculating the current average inventory cycle;
[0012] S160, inputting the average inventory cycle within a specified historical period, the environmental factors and macro-control factors of the region where the target vehicle is located into the prediction model to obtain the future inventory cycle output by the prediction model; the prediction model includes an input layer and a long short-term memory network LSTM layer connected in sequence; the input layer is used to extract features of the quantized average inventory cycle within a specified historical period, the environmental factors and macro-control factors of the region where the target vehicle is located to obtain a feature vector, and the LSTM layer is used to perform deep learning and memory on the feature vector to output the future inventory cycle;
[0013] Prior to S110, it also included:
[0014] Collect training samples for the prediction model, the training samples include the average inventory cycle of multiple sample vehicles with inventory status within a specified historical period, environmental factors of the area where the sample vehicles are located within the specified historical period, and macro-control factors within the specified historical period; the labels of the training samples are the average inventory cycle of multiple sample vehicles from production to sales; and train the prediction model based on the training samples and labels.
[0015] Optionally, the environmental factors include climate factors, geographical factors and holiday factors; the macro-control factors include policy factors, economic indicators, city size, road mileage and population factors.
[0016] Optionally, the step S120 of determining whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle includes:
[0017] Determine the actual mileage traveled by the target vehicle according to the positioning information of the target vehicle;
[0018] Reading the recorded mileage from the dashboard or large screen of the target vehicle;
[0019] If the difference between the actual mileage and the recorded mileage is greater than or equal to the set value, the target vehicle is on the way;
[0020] If the difference between the actual mileage and the recorded mileage is less than a set value, the target vehicle is not on the way.
[0021] Optionally, in S140, for each target vehicle in inventory status, calculating the current inventory cycle includes:
[0022] Subtract the production date of each target vehicle in inventory status from the current date to obtain the current inventory cycle.
[0023] Optionally, obtaining the production dates and sales dates of multiple target vehicles in S110 includes:
[0024] Correct and supplement the production date based on the 10th digit of the vehicle identification code of the target vehicle.
[0025] Optionally, the 150 includes:
[0026] For target vehicles of the same model, calculate the current average inventory cycle of the model;
[0027] For target vehicles of the same brand, calculate the brand’s current average inventory cycle;
[0028] For the same type of target vehicles, calculate the current average inventory cycle of the type;
[0029] For target vehicles of the same manufacturer, calculate the manufacturer’s current average inventory cycle;
[0030] The future inventory cycle includes the future inventory cycle of the vehicle model, the future inventory cycle of the brand, the future inventory cycle of the type and the future inventory cycle of the manufacturer.
[0031] In a second aspect, an embodiment of the present invention provides an electronic device, including:
[0032] at least one processor, and a memory communicatively coupled to at least one of the processors;
[0033] The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute any inventory prediction method based on full automobile circulation big data.
[0034] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are used to enable the computer to execute any inventory prediction method based on full automobile circulation big data.
[0035] Compared with the prior art, the technical effects of the present invention include:
[0036] 1. The present invention accurately calculates the current inventory cycle through the production, sales and in-transit status of vehicles, and further predicts the future inventory cycle in combination with environmental factors and macro-control factors.
[0037] 2. The present invention determines whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle; if the target vehicle is on the way, the target vehicle is re-determined as an inventory state; and vehicles with a long inventory cycle are eliminated, thereby obtaining accurate vehicles with sales prospects and removing noise, thereby obtaining inventory vehicles with market significance, which are used to obtain the inventory cycle later.
[0038] 3. The present invention is based on the full amount of automobile circulation big data to achieve the comprehensiveness and real-time nature of the data, improve the accuracy and reliability of inventory cycle calculation; use big data analysis technology to deeply explore the potential information and trends in the data, and provide strong support for inventory management and market forecasting; improve the operational efficiency and market competitiveness of automobile manufacturers, dealers and related companies, and reduce inventory costs and market risks.
[0039] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above description and other purposes, features and advantages of the present invention more obvious and easy to understand, the preferred embodiments are specifically cited and described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0041] Figure 1 A flowchart of an inventory prediction method based on full automobile circulation big data provided by an embodiment of the present invention;
[0042] Figure 2 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0043] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0044] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be connected, detachably connected, or integrated; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0045] Example 1
[0046] The present invention proposes an inventory prediction method based on the full automobile circulation big data. The present invention is applicable to the situation where the future inventory cycle of unsold vehicles is predicted based on the vehicle circulation data. The traditional inventory cycle is the time from production to the present for unsold inventory vehicles, that is, the inventory time. The present invention aims to predict the future inventory cycle of unsold inventory vehicles until the time of sale, that is, the future sales date-production date. Since the future sales date is uncertain, the present invention proposes a method to predict the future inventory cycle.
[0047] The method specifically comprises the following steps:
[0048] S110: Obtain production dates and sales dates of multiple target vehicles, and determine the sales status or inventory status of each target vehicle.
[0049] Extract production data of domestic automobile manufacturers (i.e. manufacturers), sales data of dealers, logistics and transportation data, etc., and synchronize them to the development environment. Through relevant algorithm models, pre-process the data such as cleaning, sorting, deduplication, standardization and fusion to ensure the accuracy and consistency of the data. If the dealer sells the target vehicle, the target vehicle will be marked as sold and the sales date will be written; if it is not sold, the sales date will be blank.
[0050] Using big data analysis technology, the pre-processed data is deeply analyzed and mined. Specifically, the vehicle identification code is used as a unique identifier to associate production data and sales data, thereby forming a complete state change of each single vehicle from production to sales.
[0051] Optionally, if the production date of some target vehicles is missing or wrong, the production date is corrected and supplemented according to the 10th digit of the vehicle identification code of the target vehicle. The 10th digit of the vehicle identification code indicates the year of production of the vehicle and can be used as the basis for correcting and supplementing the production date.
[0052] The production data and sales data shall at least include: vehicle identification code, manufacturer, brand, type, model, recorded mileage, positioning information, sales date, and production date. If the target vehicle has a sales date, it is determined to be in sales status; if it does not have a sales date, it is determined to be in inventory status. Positioning information (collected through the GPS global positioning system) can be read from the vehicle terminal or the cloud, which is a time series consisting of multiple location points. The recorded mileage can be read from the dashboard or large screen, which is the mileage of the target vehicle. If the target vehicle generates recorded mileage and positioning information, it will be recorded accordingly, otherwise it will be empty.
[0053] S120. If the target vehicle in the sales state has positioning information and recorded mileage, determine whether the target vehicle is on the way based on the positioning information and recorded mileage of the target vehicle; if it is on the way, execute S130; if it is not on the way, execute S140.
[0054] The present invention takes into account that if the target vehicle is sold from one dealer to another, the vehicle will be marked as sold and the sale date will be indicated, but the vehicle has not yet been sold to the consumer. This situation needs to be eliminated. After research, the inventor found that the target vehicle will be transported between different dealers by trailer transportation, which will cause a large gap between the mileage / recorded mileage of the target vehicle and the actual mileage obtained by positioning. Therefore, this can be used to determine the status of the target vehicle.
[0055] Specifically, first determine the actual mileage of the target vehicle based on the positioning information of the target vehicle; specifically, determine the actual mileage through the path length between the first positioning position and the last positioning position. The actual mileage here is not necessarily the driving mileage / recorded mileage. Read the recorded mileage from the dashboard or large screen of the target vehicle; if the difference between the actual mileage and the recorded mileage is greater than or equal to the set value (for example, 5km), the target vehicle is en route; if the difference between the actual mileage and the recorded mileage is less than the set value, the target vehicle is not en route, and the status of the target vehicle is still for sale.
[0056] S130: re-determine the target vehicle as being in inventory, and continue to execute S140.
[0057] The target vehicles in transit have not actually been sold to consumers and are still determined to be in inventory status. S121 aims to correct some target vehicles that were originally in sales status to inventory status.
[0058] S140: For each target vehicle in inventory status, calculate the current inventory cycle.
[0059] The current inventory cycle is obtained by subtracting the production date of each target vehicle in the inventory status from the current date. In this way, the current inventory cycle of each target vehicle can be obtained. It can be seen that as time goes by, when the target vehicle has not been sold, the current inventory cycle will become longer and longer.
[0060] S141. Determine whether the current inventory cycle exceeds the set duration. If so, execute S142; if not, execute S150.
[0061] S142: Eliminate target vehicles whose current inventory cycle exceeds a set time. Continue to S150.
[0062] The set duration can be determined based on market research, for example, 2 years. When the current inventory cycle exceeds 2 years, the target vehicle is likely to never be sold again and will remain in inventory. The manufacturer may recycle these vehicles. Therefore, the present invention needs to remove the target vehicles whose current inventory cycle exceeds 2 years to improve the accuracy of inventory cycle prediction.
[0063] S150 . For the target vehicles in the remaining inventory status, calculate the current average inventory cycle.
[0064] After the aforementioned inventory status correction and elimination of some target vehicles, multiple target vehicles in the remaining inventory status are obtained. Based on the aforementioned production data and sales data, the target vehicles in the inventory status in S150 are classified according to vehicle model, brand, type and manufacturer. The types include passenger cars, commercial vehicles, new energy vehicles and oil vehicles. The vehicle models can be divided according to the definitions of vehicle models by each manufacturer.
[0065] Then, for target vehicles of the same model, the current inventory cycle of each target vehicle is summed and averaged to calculate the current average inventory cycle of the model; similarly, for target vehicles of the same brand, the current average inventory cycle of the brand is calculated; for target vehicles of the same type, the current average inventory cycle of the type is calculated; for target vehicles of the same manufacturer, the current average inventory cycle of the manufacturer is calculated. In this way, the inventory cycle situation is obtained from multiple perspectives such as model, brand, type and manufacturer.
[0066] S160: Input the average inventory cycle within a specified historical period, the environmental factors of the region where the target vehicle is located, and the macro-control factors into the prediction model to obtain the future inventory cycle output by the prediction model.
[0067] The average inventory cycle within the specified historical time period is the current average inventory cycle at each time step within the specified historical time period.
[0068] The prediction model in this embodiment is a model suitable for time series prediction. Preferably, the prediction model of the present invention includes an input layer and a long short-term memory network LSTM layer connected in sequence; wherein the input layer is used to extract features of the quantified average inventory cycle within a specified historical period, environmental factors and macro-control factors in the area where the target vehicle is located to obtain a feature vector, and the LSTM layer is used to perform deep learning and memory on the feature vector to output the future inventory cycle.
[0069] Given that the current average inventory cycle includes the inventory cycle of vehicle models, brands, types, and manufacturers, the future inventory cycle also includes the future inventory cycle of vehicle models, the future inventory cycle of brands, the future inventory cycle of types, and the future inventory cycle of manufacturers. In other words, this embodiment can simultaneously predict the future inventory cycles of four statistical dimensions.
[0070] This embodiment requires inputting three types of data into the prediction model, including: the current average inventory cycle of four statistical dimensions, environmental factors and macro-control factors in the region where the target vehicle is located. The prediction model learns the relationship between the current average inventory cycle and environmental factors and macro-control factors, thereby predicting the future inventory cycle. Among them, environmental factors and macro-control factors are factors that can affect vehicle sales, thereby affecting the inventory cycle.
[0071] Specifically, environmental factors include climate factors, geographical factors and holiday factors. Climate factors include, for example, monsoon climate, plum rain climate, arid climate, etc. in the region where the target vehicle is located. Different climates will affect people's choice of different models and types of vehicles. Geographical factors include mountains, plains, rural roads and urban roads. Different geographical factors will affect the performance of the vehicle, thereby affecting the sales of the vehicle. The holiday factor is the total duration of holidays within a set time in the future, which can be 6 months or 1 year. The longer the total duration of holidays, the longer people have time to choose vehicles, and the higher the sales volume. Macroeconomic control factors include policy factors, economic indicators, city size, road mileage and population factors. Policy factors are specifically incentive factors for a certain brand, type, manufacturer and model, including discounts, cash back, prizes, etc. A quantitative value is given to the policy factor according to the degree of incentive. Economic indicators are economic data of the region where the target vehicle is located, such as GDP data. City size can be represented by city area. Road mileage is the total mileage of all roads in the region where the target vehicle is located, and all roads can include national and provincial roads. The population factor includes the total population of the area where the target vehicle is located.
[0072] It should be noted that since environmental factors and macro-control factors are determined based on the region where the target vehicle is located, the current average inventory cycle should also be calculated in the region. The region can be a province or a certain area, and the division method is not limited in the present invention. Therefore, this embodiment can use the region as a restriction condition to predict the future inventory cycle of a certain model, brand, type and manufacturer of vehicles in the region.
[0073] The prediction model in this embodiment needs to predict the future inventory cycle based on environmental factors and macro-control factors. The environmental factors and macro-control factors need to be quantified before they can be input into the prediction model for numerical calculations. After the above analysis, the climate factors and geographical factors in the environmental factors have not yet been quantified. This embodiment first uses the one-hot vector method to convert the climate factors and geographical factors into vectors, and then forms an input vector with the converted vectors and the quantified holiday factors, macro-control factors and the current average inventory cycle. Exemplarily, the specified historical time length can be 30 days back from the current moment. Then the current average inventory cycle, environmental factors and macro-control factors of the region for each day in these 30 days are obtained, that is, the input vector of a time step (1 day) is M= (T1 i ,T2 i ,T3 i ,T4 i ,H i ,Z i ), where T1 i ,T2 i ,T3 i ,T4 i are the inventory cycle of the model, brand, type and manufacturer at the i-th time step. i is the set of quantitative values of environmental factors at the i-th time step. iis the set of quantized values of the macro-control factors at the i-th time step. Thus, assuming that the dimension of M is 1×n, the dimension of the input vector formed by 30 vectors M is 30×n. Input the 30×n input vector to the input layer. The input layer includes multiple convolutional layers, which can integrate the current average inventory cycle, environmental factors in the region, and macro-control factors at the same time to propose their mutual influence relationship and obtain a feature vector, assuming that the dimension is 30×m. The LSTM layer receives a three-dimensional tensor from the input layer (the feature vector output by the input layer needs to be converted into a three-dimensional tensor), with a shape of (batch_size, time_step, input_size), where batch_size represents the number of samples for each training or prediction (determined according to actual business needs), time_step represents the time step contained in each sample (i.e. 30), and input_size represents the number of features contained at each time point (i.e. m). The LSTM layer performs deep learning and memory on the three-dimensional tensor through the input gate, forget gate, output gate, and memory cells. The output layer of the LSTM network outputs a two-dimensional tensor with a shape of (batch_size, output_size), where output_size represents the number of target values outputted by each sample. In this embodiment, output_size is set to 4, which represent the future inventory cycle of the model, the future inventory cycle of the brand, the future inventory cycle of the type, and the future inventory cycle of the manufacturer.
[0074] The mechanism of the prediction model is to use environmental factors and macro-control factors to make corrections based on the current average inventory cycle in four statistical dimensions, so as to obtain the future inventory cycle of vehicle models, brands, types and manufacturers.
[0075] Before using the prediction model for prediction, the prediction model needs to be trained first so that it has the function of predicting future inventory cycles. Specifically, before S110, it also includes: collecting training samples for the prediction model, the training samples include the average inventory cycle of multiple sample vehicles with inventory status within a specified historical period, the environmental factors of the area where the sample vehicles are located within the specified historical period, and the macro-control factors within the specified historical period; the labels of the training samples are the average inventory cycle of multiple sample vehicles from production to sales (this requires that the sample vehicles are vehicles that have been sold at the current moment in order to calculate the average inventory cycle from production to sales); training the prediction model based on the training samples and labels.
[0076] For example, sample vehicle A was produced on January 1, 2018 and sold on June 1, 2019. Sample vehicle B of the same model was produced on February 1, 2018 and sold on April 31, 2018. Then, if the specified historical duration is April 2018 (30 days), the current inventory cycle of sample vehicle A is calculated as 3 months, 3 months and 1 day, 3 months and 2 days, etc. for each day in April 2018, and the environmental factors and macro-control factors of the region where sample vehicle A is located are calculated for each day in April 2018. The future inventory cycle of sample vehicle A is 17 months. Similarly, the current inventory cycle of sample vehicle B is calculated as 2 months, 2 months and 1 day, 2 months and 2 days, etc. for each day in April 2018, as well as the environmental factors and macro-control factors of the region where sample vehicle B is located. The future inventory cycle of sample vehicle B is 13 months. The label is the average of the future inventory cycles of the two sample vehicles: 15 months. It should be noted that in actual operation, it is necessary to count the average inventory cycle, environmental factors, macro-control factors, and labels of a large number of sample vehicles within a specified historical period. The inventory cycles of sample vehicles of the same model, brand, type, or manufacturer are averaged to obtain 4-dimensional labels: the average inventory cycle of the same model, the average inventory cycle of the same type, the average inventory cycle of the same brand, and the average inventory cycle of the same manufacturer. In the training prediction model, these 4-dimensional labels are subtracted from the 4-dimensional target values output by LSTM to construct a loss function, and the parameters in the prediction model are optimized by minimizing the loss function.
[0077] Compared with the prior art, the present invention has the following technical effects:
[0078] 1. The present invention accurately calculates the current inventory cycle through the production, sales and in-transit status of vehicles, and further predicts the future inventory cycle in combination with environmental factors and macro-control factors.
[0079] 2. The present invention determines whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle; if the target vehicle is on the way, the target vehicle is re-determined as an inventory state; and vehicles with a long inventory cycle are eliminated, thereby obtaining accurate vehicles with sales prospects and removing noise, thereby obtaining inventory vehicles with market significance, which are used to obtain the inventory cycle later.
[0080] 3. The present invention is based on the full amount of automobile circulation big data to achieve the comprehensiveness and real-time nature of the data, improve the accuracy and reliability of inventory cycle calculation; use big data analysis technology to deeply explore the potential information and trends in the data, and provide strong support for inventory management and market forecasting; improve the operational efficiency and market competitiveness of automobile manufacturers, dealers and related companies, and reduce inventory costs and market risks.
[0081] Example 2
[0082] like Figure 2 As shown, this embodiment provides an electronic device, including:
[0083] at least one processor; and
[0084] a memory communicatively connected to at least one of the processors; wherein,
[0085] The memory stores instructions executable by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to perform the above method. At least one processor in the electronic device can perform the above method, and thus has at least the same advantages as the above method.
[0086] Optionally, the electronic device also includes interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on a memory to display graphical information of a GUI (Graphical User Interface) on an external input / output device (such as a display device coupled to an interface). In other embodiments, if necessary, multiple processors can be used together with multiple memories, and / or multiple buses can be used together with multiple memories. Similarly, multiple electronic devices can be connected (for example, as a server array, a group of blade servers, or a multi-processor system), and each device provides some necessary operations. Figure 2 A processor 301 is taken as an example.
[0087] The memory 302 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the inventory prediction method based on full automobile circulation big data in the embodiment of the present invention. The processor 301 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 302, that is, realizing the above-mentioned inventory prediction method based on full automobile circulation big data.
[0088] The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 302 may further include a memory remotely arranged relative to the processor 301, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0089] The electronic device may further include: an input device 303 and an output device 304. The processor 301, the memory 302, the input device 303 and the output device 304 may be connected via a bus or other means, and the figure takes the connection via a bus as an example.
[0090] The input device 303 can receive input digital or character information, and the output device 304 can include a display device, an auxiliary lighting device (e.g., LED), and a tactile feedback device (e.g., a vibration motor), etc. The display device can include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device can be a touch screen.
[0091] Example 3
[0092] This embodiment provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable the computer to execute the above method. The computer instructions on the computer-readable storage medium are used to enable the computer to execute the above method, and thus have at least the same advantages as the above method.
[0093] The medium in the present invention may adopt any combination of one or more computer-readable media. The medium may be a computer-readable signal medium or a computer-readable storage medium. The medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of the medium (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, the medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device, or device.
[0094] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0095] The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.
[0096] Computer program code for performing the operations of the present invention may be written in one or more programming languages or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).
[0097] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.
Claims
1. A method for inventory prediction based on full automobile circulation big data, characterized in that: The method specifically comprises the following steps: S110, obtaining the production date and sales date of multiple target vehicles, and determining the sales status or inventory status of each target vehicle; S120: If the target vehicle in the sale state has positioning information and recorded mileage, determine whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle; S130: if the target vehicle is in transit, re-determine the target vehicle as being in inventory; S140, for each target vehicle in inventory status, calculating the current inventory cycle; If the current inventory cycle exceeds the set time, the target vehicle is removed; S150, for the target vehicles in the remaining inventory status, calculating the current average inventory cycle; S160, inputting the average inventory cycle within a specified historical period, the environmental factors and macro-control factors of the region where the target vehicle is located into the prediction model to obtain the future inventory cycle output by the prediction model; the prediction model includes an input layer and a long short-term memory network LSTM layer connected in sequence; the input layer is used to extract features of the quantized average inventory cycle within a specified historical period, the environmental factors and macro-control factors of the region where the target vehicle is located to obtain a feature vector, and the LSTM layer is used to perform deep learning and memory on the feature vector to output the future inventory cycle; Prior to S110, it also included: Collect training samples for the prediction model, the training samples include the average inventory cycle of multiple sample vehicles with inventory status within a specified historical period, environmental factors of the area where the sample vehicles are located within the specified historical period, and macro-control factors within the specified historical period; the labels of the training samples are the average inventory cycle of multiple sample vehicles from production to sales; and train the prediction model based on the training samples and labels.
2. The method according to claim 1, characterized in that The environmental factors include climate factors, geographical factors and holiday factors; the macro-control factors include policy factors, economic indicators, city size, road mileage and population factors.
3. The method according to claim 1, characterized in that The step S120 determines whether the target vehicle is on the way according to the positioning information and recorded mileage of the target vehicle, including: Determine the actual mileage traveled by the target vehicle according to the positioning information of the target vehicle; Reading the recorded mileage from the dashboard or large screen of the target vehicle; If the difference between the actual mileage and the recorded mileage is greater than or equal to the set value, the target vehicle is on the way; If the difference between the actual mileage and the recorded mileage is less than a set value, the target vehicle is not on the way.
4. The method according to claim 1, characterized in that: In S140, for each target vehicle in the inventory state, the current inventory cycle is calculated, including: Subtract the production date of each target vehicle in inventory status from the current date to obtain the current inventory cycle.
5. The method according to claim 1, characterized in that The production date and sales date of multiple target vehicles are obtained in S110, including: Correct and supplement the production date based on the 10th digit of the vehicle identification code of the target vehicle.
6. The method according to claim 1, characterized in that The 150 include: For target vehicles of the same model, calculate the current average inventory cycle of the model; For target vehicles of the same brand, calculate the brand’s current average inventory cycle; For the same type of target vehicles, calculate the current average inventory cycle of the type; For target vehicles of the same manufacturer, calculate the manufacturer’s current average inventory cycle; The future inventory cycle includes the future inventory cycle of the vehicle model, the future inventory cycle of the brand, the future inventory cycle of the type and the future inventory cycle of the manufacturer.
7. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to at least one of the processors; The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute the inventory prediction method based on full automobile circulation big data as described in any one of claims 1-6.
8. A computer-readable storage medium, characterized in that: The medium stores computer instructions, and the computer instructions are used to enable the computer to execute the inventory prediction method based on full automobile circulation big data as described in any one of claims 1-6.
