Truck freight real-time prediction method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of real-time prediction of freight of trucks, and discloses a real-time prediction method and device of freight of trucks, computer equipment and a storage medium, and the method comprises the steps: obtaining the order information of a delivery order of a specified truck, collecting the real-time position and vehicle data of the specified truck in real time, generating a driving path, and transmitting the driving path to a server; acquiring real-time driving environment data; and predicting the remaining freight cost of the truck based on the real-time driving environment data, the driving path and the vehicle data, performing statistics on the cost spent by the specified truck, and performing calculation to obtain the predicted total freight for delivering the order. The beneficial effects of the invention are that through comprehensive application of real-time data and multi-dimensional information, accurate freight prediction is provided for logistics enterprises, so that the enterprises can be helped to make more scientific decisions for future transportation arrangement, and the profitability and service quality of the enterprises can be effectively improved in a market environment with intense competition.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time prediction of truck freight, and particularly to a method, device, computer equipment and storage medium for real-time prediction of truck freight. Background Art

[0002] With the rapid development of modern logistics business, the demand for truck transportation is increasing day by day. Traditional truck freight prediction often relies on manual experience or estimation of historical contract prices. The data dimension of freight prediction is often too single. Many prediction methods only consider basic mileage and cargo weight, ignoring other information, resulting in low accuracy of freight price prediction. Summary of the Invention

[0003] Based on this, in view of the existing problems of real-time prediction of truck freight, it is necessary to propose a method, device, computer equipment and storage medium for real-time prediction of truck freight.

[0004] A method for real-time prediction of truck freight, the method comprising: Obtaining order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order; Based on the destination location, real-time collecting the real-time location and vehicle data of the specified truck; Generating a driving route through the real-time location and the destination location, and obtaining real-time driving environment data based on the driving route; Predicting the remaining freight cost of the truck based on the real-time driving environment data, the driving route and the vehicle data; Counting the cost already spent by the specified truck, and calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck.

[0005] Further, the step of counting the cost already spent by the specified truck includes: Obtaining the real-time mileage from the delivery order time to the current time, and historical vehicle data; Calculating the cost already spent by the specified truck based on the real-time mileage and the historical vehicle data.

[0006] Further, the step of obtaining the real-time mileage from the delivery order time to the current time includes: Obtaining the real-time GPS trajectory from the delivery order time to the current time; Based on the real-time GPS trajectory and a preset high-precision map, using an improved hidden Markov model to eliminate positioning drift errors, and realizing continuous mileage estimation through inertial navigation data and visual odometer to obtain the real-time mileage.

[0007] Further, the step of generating a driving route based on the real-time position and the destination position includes: Generating multiple feasible routes in real time according to the destination position and the current position; Predicting the driving time, fuel consumption, and tolls for each of the feasible routes; Calculating the corresponding driving scores by weighted calculation based on the driving time, fuel consumption, and tolls of the feasible routes; Selecting the feasible route with the highest driving score as the driving route.

[0008] Further, after the step of counting the cost already spent by the specified truck and calculating the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck, the method further includes: Obtaining a market competition factor sequence and the historical order price fluctuation law based on the order information; Inputting the historical order price fluctuation law and the market competition factor sequence into a game theory optimization model to obtain the freight pricing.

[0009] Further, the step of calculating the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck includes: Obtaining the fixed cost of the specified truck based on the order information; Adding the cost already spent, the remaining freight cost of the truck, and the fixed cost to obtain the predicted total freight.

[0010] Further, after the step of counting the cost already spent by the specified truck and calculating the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck, the method further includes: Counting the actual total freight for delivering the order; According to the formula Calculating the error rate; where represents the error rate, represents the actual total freight, represents the predicted total freight; Judging whether the error rate is lower than a preset value; If the error rate is lower than the preset value, it is determined that the predicted total freight is reasonable.

[0011] A real-time prediction device for truck freight, the device includes: An acquisition module, configured to acquire order information of a specified truck for delivering an order; wherein, the order information at least includes the destination position of the order; A collection module, configured to collect the real-time position and vehicle data of the specified truck in real time based on the destination position; A generation module, configured to generate a driving route based on the real-time position and the destination position, and obtain real-time driving environment data based on the driving route; A prediction module, configured to predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and vehicle data; A calculation module, configured to count the cost already spent by the specified truck, and calculate the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck.

[0012] A computer device, comprising a memory and a processor, where the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the following steps: Obtain the order information of the order delivered by the specified truck; wherein, the order information at least includes the destination position of the order; Based on the destination position, collect the real-time position and vehicle data of the specified truck in real time; Generate a driving route based on the real-time position and the destination position, and obtain real-time driving environment data based on the driving route; Predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and vehicle data; Count the cost already spent by the specified truck, and calculate the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck.

[0013] A computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the processor is caused to perform the following steps: Obtain the order information of the order delivered by the specified truck; wherein, the order information at least includes the destination position of the order; Based on the destination position, collect the real-time position and vehicle data of the specified truck in real time; Generate a driving route based on the real-time position and the destination position, and obtain real-time driving environment data based on the driving route; Predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and vehicle data; Count the cost already spent by the specified truck, and calculate the predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck.

[0014] Advantages of the present invention: By comprehensively using real-time data and multi-dimensional information, accurate freight prediction is provided for logistics enterprises, which can help enterprises make more scientific decisions on future transportation arrangements and effectively improve the profitability and service quality of enterprises in a highly competitive market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0016] Among them: Figure 1 It is an application environment diagram of a real-time prediction method for truck freight in an embodiment; Figure 2 It is a flowchart of a real-time prediction method for truck freight in an embodiment; Figure 3 It is a structural block diagram of a real-time prediction device for truck freight in an embodiment; Figure 4 It is a structural block diagram of a computer device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0018] Figure 1 It is an application environment diagram of a real-time prediction of truck freight in an embodiment. Refer to Figure 1 , this real-time prediction method for truck freight is applied to a real-time prediction system for truck freight. The real-time prediction system for truck freight includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected through a network. The terminal 110 may specifically be a desktop terminal or a mobile terminal, and the mobile terminal may specifically be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The server 120 may be implemented by an independent server or a server cluster composed of multiple servers. The terminal 110 is used to collect the real-time position and vehicle data of a specified truck, and the server 120 is used to generate the predicted total freight.

[0019] As Figure 2As shown, in one embodiment, a method for real-time prediction of truck freight is provided. This method can be applied to both terminals and servers. In this embodiment, it is exemplified by being applied to a server. The method for real-time prediction of truck freight specifically includes the following steps: S1: Obtain the order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order; S2: Based on the destination location, collect the real-time location and vehicle data of the specified truck in real time; S3: Generate a driving route through the real-time location and the destination location, and obtain real-time driving environment data based on the driving route; S4: Predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and the vehicle data; S5: Statistically calculate the cost already spent by the specified truck, and calculate the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck.

[0020] As described in step S1 above, obtain the order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order. The order information usually includes multiple key data, such as order number, customer information, starting location and destination location, type of goods, weight, volume, delivery time requirement, etc. Among them, the destination location is a very important parameter, which directly affects the design of the transportation route, the prediction of driving time, and the calculation of freight. Data extraction can be carried out through an automated interface to ensure the real-time and accuracy of information. In addition, if the order is modified, it can also be directly modified in the order information, that is, data is updated.

[0021] As described in step S2 above, the real-time location and vehicle data of a specified truck are collected based on the destination location. Specifically, the location and relevant vehicle data of the specified truck can be obtained in real time through an on-board OBD device. On-board OBD (On-Board Diagnostics) is an automatic detection system used to monitor and record the operating status of a vehicle and the functions of the emission control system. The OBD system can help technicians and vehicle owners monitor the vehicle performance in real time. The on-board OBD device is installed on the truck and can provide real-time information such as the current location, speed, and driving direction of the vehicle. In addition, the vehicle data also includes information such as fuel consumption, load, and engine status, which are all important variables for evaluating vehicle performance and freight costs. By combining with the destination location, the system can monitor the driving progress of the truck in real time and prepare for subsequent path analysis and cost prediction. To ensure the accuracy and timely update of the data, the system needs to be able to stably receive and process high-frequency data streams and adopt data cleaning and filtering techniques to exclude outliers, so as to provide reliable basic data for subsequent steps. In some embodiments, the real-time location may be the starting point of the delivery order.

[0022] As described in step S3 above, a driving path is generated based on the real-time location and the destination location, and real-time driving environment data is obtained based on the driving path. The generation of the driving path usually involves map services and route planning algorithms. By considering the real-time traffic conditions and road conditions, the system can obtain an optimal or approximately optimal path. On this basis, it is also necessary to obtain real-time environment data related to this driving path, such as road congestion, accident information, weather conditions, etc. By integrating real-time traffic data and meteorological information, the system will be able to more comprehensively evaluate the potential risks and uncertainties of the driving path. The driving environment data can include the real-time road conditions (congestion index, accident warning) of the traffic management platform, as well as the integrated meteorological data (rainfall, wind speed) and terrain data (slope, altitude). The specific acquisition method can be obtained by acquiring data through the corresponding APP.

[0023] As described in step S4 above, predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and the vehicle data. By using machine learning and data analysis techniques, comprehensively analyze the real-time driving environment data, the driving route, and the vehicle data to predict the remaining freight cost of the truck. Specifically, a prediction model can be pre-constructed, and the model can be trained based on historical data, where the historical data is the historical driving environment data, the historical driving route, and the historical vehicle data corresponding to historical orders. And use the actual freight cost as a label for labeled training. The prediction model can be any one of a convolutional neural network, a support vector machine, and a time series model. In addition, the model needs to be able to quantify the impact of various factors on the freight, such as how an increase in driving time and route complexity affects fuel consumption and the overall cost. Through these analyses, the system can provide an accurate remaining freight cost, thus helping enterprises make more scientific decisions on future transportation arrangements.

[0024] As described in step S5 above, count the cost already spent by the specified truck, and calculate the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck. It is necessary to count the cost already spent by the truck during the delivery of the current order, and integrate it with the previously predicted remaining freight cost to finally obtain the predicted total freight for delivering this order. The already spent cost is the cost that has already been incurred, that is, the cost incurred from the starting point of delivering the order to the real-time position. These data can be analyzed and calculated through the existing data of the vehicle. By adding the already spent cost and the remaining freight cost, the system can provide a comprehensive freight prediction, which not only enables logistics enterprises to estimate costs more accurately, but also allows enterprises to have more sufficient basis when conducting price negotiations.

[0025] In one embodiment, step S5 of counting the cost already spent by the specified truck includes: S501: Obtain the real-time mileage from the moment of delivering the order to the current moment, and the historical vehicle data; S502: Calculate the cost already spent by the specified truck based on the real-time mileage and the historical vehicle data.

[0026] As described in the above steps S501 - S502, obtain the real - time mileage data and historical vehicle data related to the specified truck from the database or real - time monitoring system. The real - time mileage refers to the distance traveled by the truck from the moment the delivery order is received to the current moment. This data is usually updated and recorded in real - time through an in - vehicle OBD device, so that the specific driving status of the vehicle can be accurately grasped. To obtain the real - time mileage, the system needs to regularly collect and calculate the vehicle's positioning data to ensure the real - time and accuracy of this data during the delivery process. The historical vehicle data provides background information for cost calculation. These data can help analyze the performance and economy of the vehicle under different conditions, and then improve the accuracy when calculating the cost already spent. To improve the accuracy of the data, the system needs to perform data cleaning to eliminate potential interference factors such as outliers, ensuring that the finally collected mileage data and historical vehicle data are reliable. Use the obtained real - time mileage and historical vehicle data to calculate the cost already spent by the specified truck during the delivery process. During the calculation process, estimate the fuel cost based on the real - time mileage. The historical vehicle data includes the instantaneous fuel consumption. The instantaneous fuel consumption refers to the speed at which the vehicle consumes fuel at a certain specific moment or under specific conditions, usually expressed as the amount of fuel consumed per hour (such as liters / hour). The instantaneous fuel consumption provides information on the fuel efficiency of the vehicle under the current driving conditions, reflecting the impact of factors such as driving behavior, road conditions, and vehicle load on fuel consumption. Therefore, calculate the average fuel consumption based on the instantaneous fuel consumption, and then combine it with the real - time mileage to more accurately estimate this part of the cost.

[0027] In one embodiment, step S501 of obtaining the real - time mileage from the moment the delivery order is received to the current moment includes: S5011: Obtain the real - time GPS trajectory from the delivery order time to the current time; S5012: Based on the real - time GPS trajectory and a preset high - precision map, use an improved Hidden Markov Model to eliminate the positioning drift error, and realize continuous mileage estimation through inertial navigation data and visual odometry to obtain the real - time mileage.

[0028] As described in the above steps S5011 - S5012, the system tracks the location information of the specified truck in real time through the GPS device to obtain the GPS trajectory data from the order delivery time to the current moment. Specifically, the GPS system locates the truck through satellite signals and updates its specific location regularly. These location information are converted into GPS trajectories, usually stored in the form of latitude and longitude coordinates. According to the set time interval (for example, collected once per second or every few seconds), the system will continuously record the location of the vehicle. The key is that the obtained GPS trajectory must be continuous and accurate for subsequent mileage calculation. In practical applications, due to environmental factors (such as signal occlusion, weather changes, etc.) or hardware limitations, the GPS signal may be delayed or lost, resulting in noisy acquired data. This requires the system to have a certain anti - interference ability and be able to improve the positioning accuracy through historical data correction (such as through filtering algorithms). In addition, the real - time GPS trajectory will provide basic data for subsequent trip analysis and cost calculation, ensuring the traceability and transparency of the overall logistics process. The system first combines the real - time GPS trajectory with a pre - set high - precision map to improve the accuracy of location estimation. Since GPS positioning may deviate due to various factors (i.e., positioning drift), an improved hidden Markov model (IHMM) is used to eliminate the positioning error. The improved hidden Markov model (Improved Hidden Markov Model, IHMM) is an extension and improvement of the hidden Markov model (HMM). The state transition of HMM only depends on the current state, while IHMM may consider the influence of historical states and can better simulate complex dynamic processes. IHMM is more suitable for the analysis of complex dynamic systems.

[0029] Specifically, in IHMM, a state space is defined. The state can represent the location of the vehicle on the high - precision map (such as latitude and longitude coordinates, road segment numbers, etc.). Due to the instability of the GPS signal, there may be an error between the real location and the measured location, which needs to be represented as a hidden state. The observation sequence is the location measured by GPS (which may not match the actual location due to drift). Each observation value will correspond to a possible hidden state. The transition probability between states is defined according to the high - precision map, representing the probability that the vehicle moves from one state to another. For example, a reasonable driving path is designed based on the road network to simulate the normal driving behavior of the vehicle. The probability of observing a specific location given a hidden state can be estimated based on historical data or map features. According to the hidden state obtained through the IHMM model, the preliminary GPS location is corrected. Using the location information provided by the hidden state, the location recorded by GPS is adjusted to filter out the drift error caused by signal interference, occlusion, etc.

[0030] To achieve continuous mileage estimation, the system also applies inertial navigation data and visual odometry. The Inertial Navigation System (INS) continuously monitors the vehicle's motion state using accelerometers and gyroscopes to achieve very accurate position estimation within a short period; while visual odometry analyzes the feature points of the road ahead through image recognition and processing techniques to assist in judging the vehicle's path. First, data from the accelerometer and gyroscope are obtained from the vehicle's Inertial Measurement Unit (IMU), and image data of the surrounding environment are obtained using a camera mounted on the vehicle. The vehicle's speed and position are calculated using the accelerometer data. The speed is obtained by integrating the acceleration data, and then the displacement is obtained by integrating the speed. The relative motion of the vehicle is calculated by analyzing the feature points of consecutive frames, and the IMU data and visual odometry data are combined using a sensor fusion algorithm. A common method is the Kalman filter, which can effectively reduce the error caused by the noise of a single sensor. Based on the fused data, the system continuously updates the vehicle's current position and driving mileage. Whenever new IMU or visual data arrives, new estimations and corrections are made using the previous methods.

[0031] In one embodiment, the step S3 of generating a driving path through the real-time position and the destination position includes: S301: Generate multiple feasible paths in real time according to the destination position and the current position; S302: Predict the driving time, fuel consumption, and tolls for each of the feasible paths; S303: Calculate the corresponding driving scores by weighted calculation according to the driving time, fuel consumption, and tolls of the feasible paths; S304: Select the feasible path with the highest driving score as the driving path.

[0032] As described in the above steps S301 - S304, multiple feasible routes are generated in real time according to the current position of the freight truck and the destination position. Map service APIs (such as Google Maps, OpenStreetMap, etc.) or an integrated dynamic map database can be used to calculate possible driving routes. The key to generating feasible routes lies in accuracy and real - time. The system relies on high - precision map data and combines real - time traffic information to ensure that the generated routes are feasible at the current time. In addition, the generation of multiple routes can be based on different strategies, such as selecting the route with the shortest distance level, the route with the fastest driving time, or the route with the lowest cost, etc. On this basis, the system considers the traffic capacity of different routes to avoid delays caused by traffic congestion, accidents, or construction. In this way, the system can provide the driver with the most choices and ensure corresponding adjustments can be made according to various situations in actual operation. In step S302, the system comprehensively evaluates each generated feasible route, including the prediction of driving time, fuel consumption, and tolls. This process is achieved by a detailed analysis of the characteristics of the route. First, the prediction of driving time involves multiple factors, including the total distance of the route, road types (such as highways, urban roads, etc.), and the current traffic conditions. The system needs to use real - time traffic data, refer to historical traffic flows, and driving habits to make a scientific time prediction. Second, the prediction of fuel consumption needs to consider the specific performance parameters of the freight truck, such as the fuel consumption rate, load condition, and road gradient. During driving, the fuel consumption of the vehicle may vary due to factors such as speed changes, traffic signals, acceleration, and deceleration. In this application, the instantaneous fuel consumption is recorded, so the fuel consumption of the vehicle can be obtained by calculating the average fuel consumption. As for the calculation of tolls, it depends on the distribution of various toll stations on the route, including highway tolls, bridge fees, etc. The driving time, fuel consumption, and tolls of each feasible route are weighted and calculated to obtain the corresponding driving score. This process comprehensively evaluates information from multiple dimensions to facilitate quantitative comparison of routes. The key to weighted calculation is to determine the weights of each factor, and this weight can be set according to actual business requirements. According to the preset weights of driving time, fuel consumption, and tolls, the driving score is calculated by weighted calculation. For example, in some cases, transportation time may be more prioritized than cost, and vice versa, fuel consumption and tolls may account for a larger proportion in route selection. After standardizing the driving time, fuel consumption, and tolls, the system can use the weighted average method to calculate the comprehensive score of each route. Specifically, the calculation formula can be expressed as: driving score = w1×driving time + w2×fuel consumption + w3×tolls, where w1, w2, w3 represent the weights of each factor, all of which are preset values.Select the feasible path with the highest driving score as the driving path. Based on the calculated driving scores, select the feasible path with the highest score as the final driving path to ensure the best balance between driving efficiency and economy. In addition, to ensure the flexibility of path selection, the path can also be re-evaluated at any time according to subsequent real-time data (such as changes in traffic conditions). The system can periodically re-evaluate the path and dynamically update the driving scores. This ability to dynamically adjust can significantly improve the efficiency and responsiveness of transportation, ensuring the smooth progress of the logistics process. By selecting the path with the highest score, the system will increase the on-time delivery rate of goods, reduce operating costs, and improve customer satisfaction. It can also improve the prediction accuracy of freight rates.

[0033] In one embodiment, after step S5 of statistically calculating the cost already spent by the specified truck and calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck, the following steps are further included: S601: Obtain the market competition factor sequence and the historical order price fluctuation pattern based on the order information; S602: Input the historical order price fluctuation pattern and the market competition factor sequence into the game theory optimization model to obtain the freight pricing.

[0034] As described in steps S601 - S602 above, collect and analyze the market competition factor sequence and the historical order price fluctuation pattern related to the current order. To obtain the market competition factors, data is collected from various data sources. The market competition factor sequence specifically includes one or more of data such as market research, competitors' price strategies, industry reports, economic indicators, supply and demand relationships of surrounding vehicles, and platform bidding data. These factors can not only help the system understand the market competition situation but also be used as the basis for subsequent pricing strategies. At the same time, it is also necessary to analyze the historical order price fluctuation pattern. This step usually involves a large amount of historical data analysis, and statistical methods are used to explore the price fluctuation patterns. For example, the price changes under different seasons, different cargo types, and different order scales can be analyzed to identify price trends and potential periodic properties. By establishing a price fluctuation model, the system can better understand the historical pricing behavior and thus provide a basis for future pricing decisions. In a specific embodiment, a time series neural network (LSTM) is used to learn the price fluctuation pattern in historical waybill data to obtain the historical order price fluctuation pattern.

[0035] Input the collected market competition factor sequence and the historical order price fluctuation pattern into a game theory optimization model to obtain a reasonable freight price. The game theory optimization model can be a game theory optimization model based on Nash equilibrium. Game theory is a mathematical framework used to analyze the interactions among multiple participants in strategic decision-making. In this application, the participants include logistics enterprises, potential customers, and competitors. Specifically, the historical order price fluctuation pattern can provide basic information about price sensitivity and customer demand for the game theory model, while the market competition factors help the model judge the pricing strategies of suppliers in the current market environment. Specifically, determine the participants in the game theory optimization model, usually the competitors and oneself, determine the strategy set for each participant, and set the payoff function for each player. Input the market competition factor sequence and the historical order price fluctuation pattern, and through calculation, find the Nash equilibrium, Pareto optimal solution, or other equilibrium solutions to optimize the payoffs of each participant. For example, in a highly competitive market, a supplier may need to lower prices to attract customers, and can reasonably raise prices when demand is strong. The model can use this data to judge the optimal pricing strategy, and by simulating different pricing scenarios for truck transportation services, evaluate the payoffs and risks of each plan. Through multiple iterations and games, the system will continuously adjust the price strategy to maximize the company's profit. During the solution process, the model may consider multiple scenarios and use algorithms (such as dynamic programming, genetic algorithms, etc.) to find the optimal solution. Finally, the output pricing result will help the truck transportation company gain an advantage in the market competition, while meeting the reasonable expectations of consumers for prices, thus achieving a win-win situation for both sides.

[0036] In one embodiment, the step S5 of calculating the predicted total freight for delivering the order based on the incurred cost and the remaining freight cost of the truck includes: S511: Obtain the fixed cost of the specified truck based on the order information; S512: Add the incurred cost, the remaining freight cost of the truck, and the fixed cost to obtain the predicted total freight.

[0037] As described in the above steps S511 - S512, the fixed costs include vehicle depreciation, insurance, driver's salary, etc. Specifically, these data can be extracted from the financial management system or the accounting system. Usually, enterprises will keep detailed records of the fixed costs of each truck. Therefore, it is necessary to obtain the records matching the specific vehicle from the database. On this basis, it is also necessary to combine the order - related information, such as the transportation time period, destination, etc., to ensure that the obtained fixed - cost data is up - to - date and relevant to the current transportation task. Adding these three parts of costs together can generate a comprehensive and fact - based predicted total freight. This predicted value is not only crucial for internal cost control and resource allocation but also provides a clear basis when negotiating prices and signing contracts with customers. Through accurate cost prediction, logistics companies can improve their operational efficiency, ensure competitiveness in the market, and better meet customer needs, thereby enhancing customer satisfaction.

[0038] In one embodiment, after step S5 of statistically calculating the costs already incurred by the specified truck and calculating the predicted total freight for delivering the order based on the incurred costs and the remaining freight costs of the truck, the following steps are further included: S611: Statistically calculate the actual total freight for delivering the order; S612: Calculate the error rate according to the formula ; where represents the error rate, represents the actual total freight, represents the predicted total freight; S613: Determine whether the error rate is lower than the preset value; S614: If the error rate is lower than the preset value, then determine that the predicted total freight is reasonable.

[0039] As described in the above steps S611 - S614, accurately calculate the actual total freight related to the specified delivery order. Here, the total freight is the total cost freight, and the predicted total freight is the cost estimated in advance. Calculate the error rate based on the gap between the actually incurred freight and the predicted freight. It calculates the absolute difference between the actual freight and the predicted freight, and then calculates the ratio with the actual cost to obtain a relative error value. The calculated error rate can intuitively reflect the deviation degree of the prediction result. Judge whether the calculated error rate is within the acceptable range according to the preset value set in advance. The preset value is usually set based on full consideration of factors such as industry standards, company business goals, and market environment, aiming to provide a clear performance benchmark for the company's freight prediction. If the error rate is lower than the preset value, it indicates that the prediction model works well and can relatively accurately reflect the actual transportation cost; if the error rate is higher than the preset value, it may mean that there are deficiencies in the model, and it is necessary to re - evaluate the prediction logic, input parameters, or the algorithm itself. The result of this judgment directly affects the subsequent decision - making process, including whether to adjust the freight pricing strategy, improve data input and analysis methods, retrain the prediction model, etc. At the same time, the judgment of the error rate is also an important part of the enterprise's risk management, helping the enterprise identify potential prediction risks so as to take necessary measures for correction. If the judged error rate is lower than the preset value, it can be determined that the current predicted total freight is reasonable, giving recognition to the good prediction effect, which is helpful for subsequent financial decisions and pricing strategies. At this time, the enterprise can continue to use the current pricing strategy, formulate the transportation price based on the reasonable predicted cost, and negotiate with customers without worrying about affecting market competitiveness due to over - pricing or under - pricing. At the same time, a reasonable prediction result can also improve customer satisfaction and enhance customers' trust in the company, thus bringing more business opportunities.

[0040] Referring to Figure 3 , the present invention also provides a real - time prediction device for truck freight, and the device includes: An acquisition module 902, configured to acquire order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order; An acquisition module 904, configured to collect the real - time location and vehicle data of the specified truck in real - time based on the destination location; A generation module 906, configured to generate a driving route through the real - time location and the destination location, and acquire real - time driving environment data based on the driving route; A prediction module 908, configured to predict the remaining freight cost of the truck based on the real - time driving environment data, the driving route, and the vehicle data; A calculation module 910, configured to calculate the cost already spent by the specified truck, and calculate a predicted total freight for delivering the order based on the cost already spent and the remaining freight cost of the truck.

[0041] In one embodiment, the calculation module 910 includes: A real-time mileage acquisition sub-module, configured to acquire the real-time mileage from the moment of delivering the order to the current moment, as well as historical vehicle data; A cost already spent calculation sub-module, configured to calculate the cost already spent by the specified truck based on the real-time mileage and the historical vehicle data.

[0042] In one embodiment, the real-time mileage acquisition sub-module includes: An acquisition unit, configured to acquire the real-time GPS trajectory from the time of delivering the order to the current time; An error elimination unit, configured to eliminate the positioning drift error based on the real-time GPS trajectory and a preset high-precision map by using an improved hidden Markov model, and implement continuous mileage estimation through inertial navigation data and visual odometry to obtain the real-time mileage.

[0043] In one embodiment, the generation module 906 includes: A plurality of feasible path generation sub-modules, configured to generate a plurality of feasible paths in real time according to the destination location and the current location; A travel time prediction sub-module, configured to predict the travel time, fuel consumption, and tolls of each of the feasible paths; A weighted calculation sub-module, configured to perform weighted calculation on the corresponding travel scores according to the travel time, fuel consumption, and tolls of the feasible paths; A travel path selection sub-module, configured to select the feasible path with the highest travel score as the travel path.

[0044] In one embodiment, the real-time prediction device for truck freight further includes: A market competition factor sequence acquisition module, configured to acquire a market competition factor sequence based on the order information, as well as the historical order price fluctuation rule; A market competition factor sequence input module, configured to input the historical order price fluctuation rule and the market competition factor sequence into a game theory optimization model to obtain the freight pricing.

[0045] In one embodiment, the calculation module 910 includes: A fixed cost acquisition sub-module, configured to acquire the fixed cost of the specified truck based on the order information; A predicted total freight calculation sub-module, configured to add the cost already spent, the remaining freight cost of the truck, and the fixed cost to obtain the predicted total freight.

[0046] In one embodiment, the real-time prediction device for truck freight also includes: An actual total freight statistics module for statistically calculating the actual total freight of delivering the order; An error rate calculation module for calculating the error rate according to the formula wherein, represents the error rate, represents the actual total freight, represents the predicted total freight; An error rate judgment module for judging whether the error rate is lower than a preset value; A predicted total freight reasonable determination module for determining that the predicted total freight is reasonable if the error rate is lower than the preset value.

[0047] Figure 4 FIG. shows the internal structure diagram of a computer device in one embodiment. The computer device may specifically be a terminal or a server. As Figure 4 shown, the computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the real-time prediction method for truck freight. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the real-time prediction method for truck freight. Those skilled in the art can understand that Figure 4 the structure shown in

[0048] merely shows the block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements. In one embodiment, a computer device is proposed, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the following steps: Obtain the order information of the specified truck delivery order; wherein, the order information at least includes the destination location of the order; Based on the destination location, real-time collect the real-time location and vehicle data of the specified truck; Generate a driving route through the real-time location and the destination location, and obtain real-time driving environment data based on the driving route; Statistically calculate the cost already incurred by the specified truck, and calculate the predicted total freight for delivering the order based on the incurred cost and the remaining freight cost of the truck.

[0049] By comprehensively using real-time data and multi-dimensional information, it provides accurate freight prediction for logistics enterprises, which can help enterprises make more scientific decisions on future transportation arrangements and effectively improve the profitability and service quality of enterprises in a highly competitive market environment.

[0050] In one embodiment, a computer-readable storage medium is proposed, storing a computer program, which when executed by a processor causes the processor to perform the following steps: Obtain the order information of the specified truck for delivering the order; wherein, the order information at least includes the destination location of the order; Based on the destination location, real-time collect the real-time location and vehicle data of the specified truck; Generate a driving route through the real-time location and the destination location, and obtain real-time driving environment data based on the driving route; Predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and the vehicle data; Statistically calculate the cost already incurred by the specified truck, and calculate the predicted total freight for delivering the order based on the incurred cost and the remaining freight cost of the truck.

[0051] By comprehensively using real-time data and multi-dimensional information, it provides accurate freight prediction for logistics enterprises, which can help enterprises make more scientific decisions on future transportation arrangements and effectively improve the profitability and service quality of enterprises in a highly competitive market environment.

[0052] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0053] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0054] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A real-time prediction method for truck freight, characterized in that The method includes: Obtaining the order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order; Collecting the real-time location and vehicle data of the specified truck in real time based on the destination location; Generating a driving route through the real-time location and the destination location, and obtaining real-time driving environment data based on the driving route; Predicting the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and the vehicle data; Counting the cost already spent by the specified truck, and calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck; 2. The real-time prediction method of truck freight according to claim 1, characterized in that The step of counting the cost already spent by the specified truck includes: Obtaining the real-time mileage from the time of delivering the order to the current time, and historical vehicle data; Calculating the cost already spent by the specified truck based on the real-time mileage and the historical vehicle data; 3. The real-time prediction method of truck freight according to claim 2, characterized in that The step of obtaining the real-time mileage from the time of delivering the order to the current time includes: Obtaining the real-time GPS trajectory from the time of delivering the order to the current time; Based on the real-time GPS trajectory and a preset high-precision map, using an improved hidden Markov model to eliminate positioning drift errors, and realizing continuous mileage estimation through inertial navigation data and visual odometry to obtain the real-time mileage; 4. The real-time prediction method for truck freight according to claim 1, characterized in that The step of generating a driving route through the real-time location and the destination location includes: Generating multiple feasible routes in real time according to the destination location and the current location; Predicting the driving time, fuel consumption, and tolls of each of the feasible routes; Calculating the corresponding driving score by weighted calculation according to the driving time, fuel consumption, and tolls of the feasible routes; Selecting the feasible route with the highest driving score as the driving route; 5. The real-time prediction method of truck freight according to claim 1, characterized in that, After the step of counting the cost already spent by the specified truck and calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck, it further includes: Obtaining the market competition factor sequence and the historical order price fluctuation rule based on the order information; Inputting the historical order price fluctuation rule and the market competition factor sequence into a game theory optimization model to obtain the freight pricing; 6. The real-time prediction method of truck freight according to claim 1, wherein The step of calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck includes: Obtaining the fixed cost of the specified truck based on the order information; Adding the already spent cost, the remaining freight cost of the truck, and the fixed cost to obtain the predicted total freight; 7. The real-time prediction method for truck freight according to claim 1, characterized in that, After the step of counting the cost already spent by the specified truck and calculating the predicted total freight for delivering the order based on the already spent cost and the remaining freight cost of the truck, it further includes: Counting the actual total freight for delivering the order; Calculate the error rate according to the formula ; where represents the error rate represents the actual total freight represents the predicted total freight; Judging whether the error rate is lower than a preset value; If the error rate is lower than the preset value, it is determined that the predicted total freight is reasonable; 8. A real-time prediction device for truck freight, characterized in that, The device includes: An obtaining module, configured to obtain the order information of a specified truck delivery order; wherein, the order information at least includes the destination location of the order; A collection module, configured to collect the real-time location and vehicle data of a specified truck in real time based on the destination location; A generation module, configured to generate a driving route through the real-time location and the destination location, and obtain real-time driving environment data based on the driving route; A prediction module, configured to predict the remaining freight cost of the truck based on the real-time driving environment data, the driving route, and the vehicle data; A calculation module, configured to count the cost already incurred by the specified truck, and calculate the predicted total freight for delivering the order based on the incurred cost and the remaining freight cost of the truck.

9. A computer-readable storage medium, characterized in that, There is a computer program stored, and when the computer program is executed by a processor, the processor is caused to execute the steps of the real-time prediction method for truck freight as described in any one of claims 1 to 7.

10. A computer device, characterized in that, The device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute the steps of the real-time prediction method for truck freight as described in any one of claims 1 to 7.

Images