Bin filling space prediction method and device, storage medium and electronic equipment

Abstract

The invention discloses a bin filling space prediction method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring vehicle transportation information in a preset historical time range, and determining historical filling spaces corresponding to a plurality of transportation lines based on the vehicle transportation information in the preset historical time range; for each transportation line, based on a historical filling space corresponding to the transportation line, performing periodic dimension time sequence prediction to obtain a first filling space; performing non-sequential prediction based on a historical filling space corresponding to the transportation line to obtain a second filling space; and performing cross validation based on the first filling space and the second filling space corresponding to each transportation unit time of the transportation line in the next period, and obtaining a target predicted filling space of each transportation unit time of the transportation line in the next period, thereby improving the accuracy of filling space prediction.

Description

Technical Field

[0001] The present invention relates to the field of smart logistics technology, and in particular to a method, device, storage medium and electronic equipment for predicting warehouse filling space. Background Art

[0002] With the continuous development of e-commerce business, logistics and transportation business is also developing rapidly. Logistics and transportation is the transportation of goods from the starting point to the target point by transport vehicles, where the starting point and the destination point can be outlets in the logistics and transportation network. Transport vehicles carry goods through warehouses. If the transported goods fail to fill the warehouse, the remaining warehouse space can be used as filling space. By introducing external sources of goods, the space waste caused by the vacancy of the remaining warehouse space can be reduced. For example, see Figure 1 , Figure 1 This is a schematic diagram of the storage space for transport vehicles.

[0003] In the process of implementing the present invention, it is found that there are at least the following technical problems in the prior art: the prediction of filling space is generally based on manual experience judgment, and the manual judgment method has the problems of low efficiency, lag and poor accuracy. Summary of the invention

[0004] The present invention provides a method, device, storage medium and electronic equipment for predicting a bin-filling space, so as to improve the accuracy of bin-filling space judgment.

[0005] According to one aspect of the present invention, a method for predicting bin filling space is provided, comprising:

[0006] Acquire vehicle transportation information within a preset historical time range, and determine historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range;

[0007] For each transport route, a time series prediction in the period dimension is performed based on the historical filling space corresponding to the transport route, and a first filling space corresponding to each transport unit time of the transport route in the next period is obtained; and a second filling space is obtained by a non-time series prediction based on the historical filling space corresponding to the transport route, wherein the second filling space is the filling space of any transport unit time of the transport route in the next period;

[0008] Based on the cross-validation of the first filling space and the second filling space corresponding to each transport unit time of the transport route in the next cycle, the target predicted filling space of the transport route for each transport unit time in the next cycle is obtained.

[0009] Optionally, determining the historical filling spaces corresponding to multiple transport routes respectively based on the vehicle transportation information within the preset historical time range includes: determining multiple transport routes and the vehicle transportation information corresponding to each of the transport routes based on the vehicle transportation information within the preset historical time range; for any transport route, determining the historical filling space of the transport route within each transport unit time based on the last vehicle transportation information of the transport route within each transport unit time.

[0010] Optionally, the determining of the historical warehouse filling space of the transport route within each transport unit time based on the last vehicle transport information of the transport route within each transport unit time includes: determining the cargo prediction space based on the cargo volume and the cargo expansion coefficient in the last vehicle transport information; determining the historical warehouse filling space within the transport unit time based on the vehicle type volume in the last vehicle transport information and the cargo prediction space.

[0011] Optionally, the method also includes: for any transport route, determining a commonly used vehicle model based on the vehicle model information in the vehicle transport information corresponding to the transport route; and correcting the vehicle model volume in the last vehicle transport information based on the commonly used vehicle model volume, wherein the corrected vehicle model volume is the maximum value of the commonly used vehicle model volume and the vehicle model volume in the last vehicle transport information.

[0012] Optionally, the temporal prediction of the period dimension is performed based on the historical filling space corresponding to the transport route to obtain the first filling space corresponding to each transport unit time of the transport route in the next period, including: for any transport unit time in the period, determining the temporal position of the transport unit time in the period; within multiple periods of the preset historical time range, obtaining multiple historical filling spaces corresponding to the transport route at the transport unit time based on the temporal position; and obtaining the first filling space corresponding to the transport unit time of the transport route in the next period based on the multiple historical filling spaces corresponding to the transport route at the transport unit time and the predicted attenuation coefficient.

[0013] Optionally, the non-time-series prediction is performed based on the historical filling space corresponding to the transport route to obtain the second filling space, including: determining the average value, variance value and preset percentile value of the historical filling space based on the historical filling space corresponding to the transport route; and obtaining the second filling space based on the average value, variance value and preset percentile value of the historical filling space.

[0014] Optionally, the method for determining the target predicted filling space for each transport unit time of the transport route in the next cycle includes: if the second filling space is smaller than the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the second filling space is determined as the target predicted filling space; if the second filling space is greater than or equal to the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the minimum value in the first filling space corresponding to each transport unit time in the next cycle is determined as the target predicted filling space.

[0015] Optionally, the method further includes: determining the transport frequency of the transport route based on the vehicle transport information corresponding to the transport route, and screening the transport routes for fill-in based on the transport frequency of the transport route; or, determining the transport frequency corresponding to each unit time of the transport route in the periodic dimension based on the vehicle transport information corresponding to the transport route, screening the unit time of the transport route for fill-in based on the transport frequency corresponding to each unit time, and using the screened unit time as the transport unit time.

[0016] According to another aspect of the present invention, there is provided a bin-filling space prediction device, comprising:

[0017] An information acquisition module, used to acquire vehicle transportation information within a preset historical time range, and determine historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range;

[0018] A prediction module is used to perform a time series prediction in a period dimension for each transport route based on the historical filling space corresponding to the transport route, to obtain a first filling space corresponding to each transport unit time of the transport route in the next period; and to perform a non-time series prediction based on the historical filling space corresponding to the transport route to obtain a second filling space, wherein the second filling space is the filling space of the transport route at any transport unit time in the next period;

[0019] The verification module is used to cross-validate the first filling space and the second filling space corresponding to each transportation unit time of the transportation route in the next cycle to obtain the target predicted filling space of the transportation route for each transportation unit time in the next cycle.

[0020] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0021] at least one processor; and

[0022] a memory communicatively connected to the at least one processor; wherein,

[0023] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the bin-filling space prediction method described in any embodiment of the present invention.

[0024] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the bin-filling space prediction method described in any embodiment of the present invention when executed.

[0025] The technical solution of the embodiment of the present invention performs time series prediction and non-time series prediction based on the historical filling space of each transport route within a preset historical time range, obtains different prediction results through two prediction methods, and cross-validates based on the time series prediction results and the non-time series prediction results to improve the accuracy of filling space prediction.

[0026] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 This is a schematic diagram of the storage space of the transport vehicle;

[0029] Figure 2 is a flow chart of a method for predicting bin filling space provided by an embodiment of the present invention;

[0030] Figure 3 It is a structural schematic diagram of a filling space prediction device provided by an embodiment of the present invention;

[0031] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] Figure 2 is a flow chart of a bin-filling space prediction method provided by an embodiment of the present invention. This embodiment is applicable to the case of periodically performing bin-filling space prediction. The method can be executed by a bin-filling space prediction device. The bin-filling space prediction device can be implemented in the form of hardware and / or software. The bin-filling space prediction device can be configured in a mobile terminal such as a mobile phone, a tablet computer, a server, or an electronic device such as a computer. Figure 2 As shown, the method includes:

[0035] S110, obtaining vehicle transportation information within a preset historical time range, and determining historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range.

[0036] S120. For each transport route, a time series prediction in the period dimension is performed based on the historical filling space corresponding to the transport route to obtain a first filling space corresponding to each transport unit time of the transport route in the next period; and a second filling space is obtained by a non-time series prediction based on the historical filling space corresponding to the transport route, wherein the second filling space is the filling space of the transport route at any transport unit time in the next period.

[0037] S130, cross-validating the first filling space and the second filling space corresponding to each transport unit time of the transport route in the next cycle, to obtain a target predicted filling space for each transport unit time of the transport route in the next cycle.

[0038] In this embodiment, the vehicle transportation information is information recording the transportation process of the transport vehicle. Exemplarily, the vehicle transportation information includes departure time, arrival time, starting point, ending point, vehicle model information, cargo volume information and driving distance, etc.

[0039] Each transport vehicle may be equipped with a mobile terminal, which may be a device such as a mobile phone or a handheld device. The mobile terminal may be connected to the management device for communication. The mobile terminal stores the vehicle model information of the transport vehicle. The transport personnel record the information of each node of the transport process by operating the mobile terminal to form vehicle transport information. For example, before departure, the departure time and the starting point are recorded. When the destination is reached, the arrival time, travel distance and ending point are recorded. The vehicle transport information corresponding to the transport process is transmitted to the management device through the mobile terminal. The management device receives and records the vehicle transport information corresponding to each transport process of each transport vehicle. The management device stores and manages the vehicle transport information based on time information (such as departure time).

[0040] A prediction request for filling space is received, and vehicle transportation information within a preset historical time range is read from the management device; or a scheduled task is pre-set, which is a prediction task for filling space that is executed at a preset time interval. When a triggering condition of the scheduled task is met (i.e., the interval between the current time and the previous execution time of the scheduled task meets the preset time interval of the scheduled task), vehicle transportation information within a preset historical time range is read from the management device.

[0041] Based on the vehicle transportation information within the preset historical time range, multiple transportation routes are determined. Specifically, the transportation route corresponding to the vehicle transportation information can be determined based on the departure time, starting point and ending point in the vehicle transportation information. It can be understood that the preset historical time range is divided into multiple cycles, each cycle includes multiple unit times, and the departure time used to divide the transportation route is the time information within each unit time. Exemplarily, a cycle can be a week, and each unit time can be a day. The departure time of the vehicle transportation information includes year-month-day information and 24-hour system information. For example, the departure time of the vehicle transportation information can be 10:30 on xx-xx-xx. Accordingly, it is determined whether it is within the preset historical time range through the year-month-day information in the departure time of the vehicle transportation information, and the transportation route is divided according to the 24-hour system information in the departure time of the vehicle transportation information, for example, the departure time used to divide the transportation route is 10:30.

[0042] Optionally, the vehicle transportation information whose starting and ending points are the same and whose interval of the 24-hour information of the departure time is less than a preset interval threshold is determined to belong to the same transportation route. For example, in the first vehicle transportation information, the starting point is A, the ending point is B, and the departure time is 10:30; in the second vehicle transportation information, the starting point is A, the ending point is B, and the departure time is 9:30. Taking the preset interval threshold of six hours as an example, it is determined that the first vehicle transportation information and the second vehicle transportation information belong to the same transportation route.

[0043] Optionally, the 24-hour system information is divided into multiple time ranges, and the vehicle transportation information with the same starting point and ending point and located in the same time range is determined to belong to the same transportation route. For example, the 24-hour system information is divided into two time ranges: 0:00-12:00 and 12:00-24:00. For example, in the first vehicle transportation information, the starting point is A, the ending point is B, and the departure time is 10:30; in the second vehicle transportation information, the starting point is A, the ending point is B, and the departure time is 9:30; in the third vehicle transportation information, the starting point is A, the ending point is B, and the departure time is 14:30; in the fourth vehicle transportation information, the starting point is B, the ending point is C, and the departure time is 10:30. It is determined that the first vehicle transportation information and the second vehicle transportation information belong to the same transportation route, the third vehicle transportation information independently belongs to a transportation route, and the fourth vehicle transportation information independently belongs to a transportation route.

[0044] For each transport route, there may be filling space during the process of transport vehicles transporting goods along the transport route. In order to better introduce external sources of goods to enrich the filling space, it is necessary to ensure the stability of the filling space. For each transport route, the stability of the transport route is determined, and the transport routes that are not stable are screened out. Optionally, the transport frequency of the transport route is determined based on the vehicle transport information corresponding to the transport route, and the transport route is screened for filling based on the transport frequency of the transport route. Specifically, the number of vehicle transport information corresponding to each transport route within a preset historical time range is counted, and the transport frequency of each transport route is obtained based on the number of vehicle transport information corresponding to each transport route divided by the number of unit time within a preset historical time range. The transport frequency of each transport route is compared with the first frequency threshold. If the transport frequency of the transport route is greater than the first frequency threshold, it is determined that the transport route is stable. If the transport frequency of the transport route is less than the first frequency threshold, it is determined that the transport route is not stable, and the prediction of the filling space of the transport route is canceled. Here, the filling space can be understood as the size of the filling space.

[0045] Optionally, the transport frequency of the transport route in each cycle is determined based on the vehicle transport information corresponding to the transport route, the transport frequency of each cycle is judged based on the second frequency threshold, and the frequency difference of the transport frequencies of adjacent cycles is determined, and the fluctuation state of the transport frequency of each cycle is characterized by the frequency difference. If the transport frequency of each cycle is greater than the second frequency threshold, and multiple frequency differences are all less than the preset fluctuation threshold, it is determined that the transport route is stable. If there are a preset number of cycles in which the transport frequency is less than the second frequency threshold, or there is a preset proportion of frequency differences that are greater than the preset fluctuation threshold, it is determined that the transport route is not stable, and the prediction of the filling space of the transport route is canceled. Among them, the first frequency threshold and the second frequency threshold can be set according to demand, and the first frequency threshold and the second frequency threshold can be the same or different, for example, it can be 80%.

[0046] In some embodiments, a transport route may be a specific unit time for transporting goods, for example, transporting from the starting point m to the ending point n every Monday, and having stability within a specific unit time. In view of the above situation, the stability of the transport route is determined in the dimension of unit time, and the transport route may be retained if the transport route has stability for a specific unit time. Among them, for multiple unit times within a cycle, for any transport route, the unit time for performing cargo transportation is recorded as the transport unit time of the transport route, and the unit time for not performing cargo transportation is recorded as the non-transportation unit time of the transport route. For transport route 1 that only departs on Saturdays and Sundays (performing cargo transportation), Saturdays and Sundays are the transport unit time of transport route 1, and Monday to Friday are the non-transportation unit time of transport route 1. For transport route 2 that departs from Monday to Sunday, every day from Monday to Sunday is the transport unit time of transport route 2.

[0047] Optionally, based on the vehicle transportation information corresponding to the transportation route, the transportation frequency corresponding to each unit time of the transportation route in the cycle dimension is determined, and the unit time of the transportation route is screened for filling based on the transportation frequency corresponding to each unit time, and the screened unit time is used as the transportation unit time. The cycle includes multiple unit times, such as the first unit time, the second unit time, etc. For any transportation route, the number of vehicle transportation information corresponding to the transportation route per unit time within the preset historical time range and the number of each unit time within the preset historical time range are determined to determine the transportation frequency corresponding to each unit time. Taking the first unit time as Monday and the preset historical time range as the past three months as an example, the number of departures of transportation route 1 on Mondays in the past three months, that is, the number of vehicle transportation information, is counted, and the transportation frequency corresponding to transportation route 1 on Mondays is obtained based on the ratio of the total number of departures on each Monday and the number of Mondays in the past three months. The transportation frequency corresponding to transportation route 1 on Monday is compared with the third frequency threshold. If the transportation frequency corresponding to transportation route 1 on Monday is greater than the third frequency threshold, it indicates that the transportation of transportation route 1 on Monday is stable, and Monday is recorded as the transportation unit time of transportation route 1.

[0048] The transport route has at least one transport unit time, which can indicate that the transport route 1 has a stable transport unit time. By determining that the transport route has stability, the filling space of the transport route can be predicted. Optionally, the transport unit time corresponding to the transport route can be each unit time within the cycle, or it can be an interval unit time within the cycle, or it can also be a local continuous unit time within the cycle. Taking a week as a cycle as an example, the transport unit time corresponding to the transport route can be each day of the week, or it can be an alternate day, or Monday to Friday, or Saturday and Sunday, etc., which can be determined according to the actual transportation situation of the transport route.

[0049] For a stable transport route, determine the historical filling space of the transport route within a preset historical time range, that is, the historical filling space corresponding to each transport unit time of the transport route within the preset historical time range. Optionally, the historical filling space corresponding to a transport route within any transport unit time can be understood as the sum of the historical filling spaces determined by at least one vehicle transport information corresponding to the transport route within any transport unit time. For example, on xx / xx / xx, transport route 1 dispatched three times, corresponding to three vehicle transport information, and the historical filling space was determined based on each vehicle transport information, and the sum of the historical filling spaces determined by the three vehicle transport information was determined as the historical filling space of transport route 1 on xx / xx / xx.

[0050] It can be understood that when a transport route has multiple trains within the same transport unit time, the storage space of the vehicle that departed earlier can be filled according to restrictions such as scheduling requirements, that is, there may be filling space only in the last vehicle within the transport unit time. Accordingly, for any transport unit time, the historical filling space of the transport unit time is determined based on the transportation information of the last vehicle, thereby simplifying the calculation process of the historical filling space.

[0051] Correspondingly, the determining of the historical filling spaces respectively corresponding to a plurality of transport routes based on the vehicle transport information within the preset historical time range includes: determining a plurality of transport routes and the vehicle transport information corresponding to each of the transport routes based on the vehicle transport information within the preset historical time range; for any transport route, determining the historical filling space of the transport route within each transport unit time based on the last vehicle transport information of the transport route within each transport unit time.

[0052] The last vehicle transportation information is the vehicle transportation information with the latest departure time in the transportation unit time of the transportation route. For example, on xx / xx / xx, transportation route 1 departs three times at 9:30, 10:00 and 10:30 respectively. The vehicle transportation information corresponding to the departure time of 10:30 is recorded as the last vehicle transportation information, and the historical filling space of transportation route 1 on xx / xx / xx is determined based on the last vehicle transportation information.

[0053] The last vehicle transport information includes vehicle type information and cargo quantity information, and the historical filling space is determined based on the vehicle type information and cargo quantity information. Specifically, the vehicle type volume can be determined based on the vehicle type information, and the cargo quantity information can be the cargo volume or cargo weight. The volume can be determined based on the cargo weight and density. The historical filling space can be determined based on the vehicle type volume and cargo volume, for example, the difference between the vehicle type volume and the cargo volume is used as the historical filling space.

[0054] In order to avoid the situation where the actual occupied volume of the goods is greater than the theoretical calculated volume due to gaps in the cargo stacking, a cargo expansion coefficient is set to improve the accuracy of the historical filling space. Optionally, the historical filling space of the transport route in each transport unit time is determined based on the last vehicle transportation information of the transport route in each transport unit time, including: determining the cargo prediction space based on the cargo volume and the cargo expansion coefficient in the last vehicle transportation information; determining the historical filling space in the transport unit time based on the vehicle model volume in the last vehicle transportation information and the cargo prediction space.

[0055] Accordingly, the product of the cargo volume and the cargo expansion coefficient is determined as the cargo prediction space, and the difference between the vehicle model volume and the cargo prediction space is determined as the historical warehouse filling space corresponding to the transportation unit time. For example, historical warehouse filling space = vehicle model volume - cargo volume * cargo expansion coefficient.

[0056] On the basis of the above embodiment, for any transport route, the commonly used vehicle model is determined based on the vehicle model information in the vehicle transport information corresponding to the transport route; the vehicle model volume in the last vehicle transport information is corrected based on the commonly used vehicle model volume, wherein the corrected vehicle model volume is the maximum value of the commonly used vehicle model volume and the vehicle model volume in the last vehicle transport information.

[0057] Among them, the commonly used vehicle model is the vehicle model frequency of the vehicle transportation information corresponding to the transportation route that exceeds the preset threshold. Specifically, the number of departures of each vehicle model in the vehicle transportation information corresponding to any transportation route within the preset historical time range is counted, and for any vehicle model, the ratio of the number of departures of the vehicle model to the total number of departures of the transportation route is determined as the vehicle model frequency of the vehicle model. The vehicle model corresponding to the maximum vehicle model frequency is determined as the commonly used vehicle model, or the vehicle model frequency that exceeds the preset threshold is determined as the commonly used vehicle model.

[0058] In order to reduce the interference of uncommon vehicle models on the prediction of filling space, the vehicle model volume in the last vehicle transportation information is corrected based on the volume of common vehicle models. Specifically, if the vehicle model in the last vehicle transportation information is an uncommon vehicle model, the vehicle model volume in the last vehicle transportation information is corrected based on the volume of common vehicle models, and the maximum value of the common vehicle model volume and the vehicle model volume in the last vehicle transportation information is obtained as the corrected vehicle model volume, and the historical filling space within the transportation unit time is determined based on the corrected vehicle model volume. Exemplarily, historical filling space = max (common vehicle model volume, vehicle model volume) - cargo volume * cargo expansion coefficient.

[0059] For any transport route, the filling space of the transport route in the next cycle is predicted based on the historical filling spaces corresponding to multiple transport unit times of the transport route within the historical preset time range. Optionally, the historical filling spaces corresponding to multiple transport unit times of the transport route within the historical preset time range are predicted in a time series and a non-time series, to obtain a time series prediction result and a non-time series prediction result, i.e., a first filling space and a second filling space.

[0060] Time series prediction can be understood as the prediction of the filling space for the unit time in the next cycle based on the historical filling space related to the unit time in the cycle dimension. Optionally, the time series prediction of the cycle dimension is performed based on the historical filling space corresponding to the transport route to obtain the first filling space corresponding to each transport unit time of the transport route in the next cycle, including: for any transport unit time in the cycle, determine the time series position of the transport unit time in the cycle; within multiple cycles of the preset historical time range, obtain multiple historical filling spaces corresponding to the transport route at the transport unit time based on the time series position; based on the multiple historical filling spaces corresponding to the transport route at the transport unit time and the predicted attenuation coefficient, obtain the first filling space corresponding to the transport unit time of the transport route in the next cycle.

[0061] A cycle includes multiple transport unit times. For any transport unit time, the temporal position of the transport unit time in the cycle is determined. Taking Monday as an example, Monday is the first day of the week, that is, the temporal position of Monday in the cycle is 1. Correspondingly, the preset historical time range includes multiple cycles. Based on the temporal position of the transport unit time, the historical filling space corresponding to the temporal position is read in each cycle distribution to obtain multiple historical filling spaces corresponding to the transport unit time, which can form a historical filling space sequence of the transport unit time.

[0062] Through the temporal position of the transport unit time, the historical filling space related to the transport unit time in the preset historical time range is read to avoid the interference of the historical filling space corresponding to other transport unit times, and realize the temporal prediction of the transport route in the transport unit time.

[0063] In this embodiment, in order to improve the accuracy of time series prediction, a prediction attenuation coefficient is set, and the first filling space of the transportation unit time in the next cycle is obtained based on the change trend of multiple historical filling spaces of the transportation line in the transportation unit time and the prediction attenuation coefficient. Specifically, the first filling space of the transportation line in the transportation unit time in the next cycle is obtained based on the weighted average of multiple historical filling spaces of the transportation line in the transportation unit time and the prediction attenuation coefficient. Exemplarily, the calculation process of the first filling space can be: V predict- h=∑V i-h r i / ∑r i , where V i-h is the historical filling space of the hth time sequence position in the i-th period, r i To predict the attenuation coefficient, for example, it can be 0.9, V predict-his the first filling space of the transport unit time at the hth time sequence position. For each transport unit time corresponding to h, the above calculation process is performed to obtain the corresponding first filling space. Different transport unit times correspond to different historical filling spaces. Accordingly, the first filling space of each transport unit time in the next cycle can also be different. Taking a week as an example, the first filling space of each day in a week can be obtained, such as V predict-1 、V predict-2 、V predict-3 、V predict-4 、V predict-5 、V predict-6 and V predict-7 It is understandable that, when the transport unit time of the transport route is a local unit time within a cycle, the first filling space corresponding to the local unit time within the cycle can be obtained, thereby reducing the prediction processing of non-unit time and the amount of calculation.

[0064] Optionally, time series prediction of the transport routes in transport unit time may also be performed based on calculation methods such as moving average models and exponential smoothing models, which are not limited here.

[0065] Non-time series prediction can be understood as a prediction process based on the data distribution of the historical filling space of the transportation route in each transportation unit time within a preset historical time range, which has no corresponding relationship with the specific transportation unit time. In this embodiment, non-time series prediction of the filling space can be performed based on the normal distribution of the historical filling space within the preset historical time range and the preset quantile value.

[0066] Optionally, a non-time-series prediction is performed based on the historical filling space corresponding to the transport route to obtain the second filling space, including: determining the average value, variance value and preset percentile value of the historical filling space based on the historical filling space corresponding to the transport route; and obtaining the second filling space based on the average value, variance value and preset percentile value of the historical filling space.

[0067] Calculate the mean and variance value of the historical filling space corresponding to the transport route, and determine the first candidate filling space based on the mean and variance value. The first candidate filling space can be the difference between the mean and variance value. Sort the historical filling spaces corresponding to the transport route from large to small, and determine the historical filling space corresponding to the preset quantile in the sorting, that is, the preset quantile value, as the second candidate filling space. Compare the first candidate filling space and the second candidate filling space, and determine the maximum value between the first candidate filling space and the second candidate filling space as the second filling space. For example, the second filling space is max(μ-σ, q), where μ is the mean of multiple historical filling spaces corresponding to the transport route within the preset historical time range, σ is the variance of multiple historical filling spaces corresponding to the transport route within the preset historical time range, and q is the preset quantile value. In some embodiments, the preset quantile value can be the lower decile value.

[0068] The second filling space is used as the second filling space for each transport unit time in the next cycle, that is, the second filling space for each transport unit time in the next cycle is the same.

[0069] It is understandable that non-sequential predictions can be made for different transport unit times according to the business needs of external cargo sources or the actual departure conditions of transport routes. For example, each unit time within a cycle, or an interval unit time within a cycle, or a local continuous unit time within a cycle. For each unit time within a cycle, a non-sequential prediction is made based on the historical filling spaces corresponding to each unit time within multiple cycles within a historical preset time range, and a second filling space is obtained, which can be recorded as V predict-per-day , which can reflect the second filling space of each day in the next cycle; for the interval unit time in the cycle, the historical filling space within the historical preset time range is sampled at intervals to obtain a set of two historical filling spaces, and non-time series prediction is performed based on each set of historical filling spaces to obtain two second filling spaces. The minimum value of the two second filling spaces is used as the second filling space for the interval unit time in the next cycle, which can be recorded as V predict-parity , which can reflect the second filling space in the case of interval departure in the next cycle. For local continuous unit time, for example, it can be the middle of the week (Monday to Friday) or weekend (Saturday and Sunday) within a week, accordingly, based on the historical filling space corresponding to the middle of the week or weekend in multiple cycles within the historical preset time range, non-time series prediction is performed to obtain the second filling space corresponding to the middle of the week, which can be recorded as V predict-weekday , which can reflect the second filling space in the next cycle when the train departs during the week, or the second filling space corresponding to the weekend, which can be recorded as V predict-weekend , which can reflect the second filling space in the next cycle when the train departs on weekends.

[0070] In order to improve the prediction accuracy of the bin-filling space, a cross-validation is performed based on the first bin-filling space predicted by the time series and the second bin-filling space predicted by the non-time series to obtain a target prediction bin-filling space. Specifically, in order to avoid the situation where the bin-filling space is predicted to be too large, the first bin-filling space and the second bin-filling space are compared, and the minimum value of the first bin-filling space and the second bin-filling space is used as the target prediction bin-filling space.

[0071] Optionally, for any transport unit time, the second filling space is compared with the first filling space corresponding to the transport unit time, and the first filling space and the second filling space are used as the target predicted filling space for the transport unit time. Here, the target predicted filling space for each transport unit time can be different.

[0072] In order to improve the stability of the target predicted filling space, the second filling space is compared with the first filling space corresponding to each transport unit time to obtain a target predicted filling space. Here, the target predicted filling space of each transport unit time is the same to unify the number of external sources of goods and reduce the complexity of introducing external sources of goods.

[0073] Optionally, the method for determining the target predicted filling space for each transport unit time of the transport route in the next cycle includes: if the second filling space is smaller than the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the second filling space is determined as the target predicted filling space; if the second filling space is greater than or equal to the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the minimum value in the first filling space corresponding to each transport unit time in the next cycle is determined as the target predicted filling space.

[0074] For example, taking a week as an example, each transport unit time in the cycle satisfies: V predict_per_day <

[0075] min(V predict1 ,V predict2 ,V predict3 ,V predict4 ,V predict5 ,V predict6 ,V predict7 ), and the second filling space is used as the target predicted filling space for each transportation unit time in the next cycle.

[0076] Similarly, weekends within the cycle satisfy V predict-weekend <min(V predict6 ,V predict7 ), and the second filling space is used as the target predicted filling space for each transportation unit time on weekends in the next cycle.

[0077] Each transport unit time in a week within the cycle satisfies: V predict-weekday <min(V predict1 ,V predict2 ,V predict3 ,V predict4 ,V predict5 ), and take the second filling space as the target predicted filling space for each transportation unit time during the week in the next cycle.

[0078] The transport unit time interval within the cycle satisfies: V predict-parity < The second filling space is used as the target predicted filling space for the transport unit time interval in the next cycle.

[0079] If, during the above verification process, the second filling space is greater than or equal to the minimum value in the first filling space corresponding to each transport unit time in the next cycle, the minimum value in the first filling space corresponding to each transport unit time in the next cycle is determined as the target predicted filling space for each transport unit time in the next cycle.

[0080] The technical solution of this embodiment performs time series prediction and non-time series prediction based on the historical filling space of each transport route within a preset historical time range, obtains different prediction results through two prediction methods, and cross-validates based on the time series prediction results and the non-time series prediction results to improve the accuracy of filling space prediction.

[0081] Figure 3 Schematic diagram of a filling space prediction device provided by an embodiment of the present invention. Figure 3 As shown, the device comprises:

[0082] The information acquisition module 210 is used to acquire vehicle transportation information within a preset historical time range, and determine the historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range;

[0083] The prediction module 220 is used to perform a time series prediction in a period dimension for each transport route based on the historical filling space corresponding to the transport route, and obtain a first filling space corresponding to each transport unit time of the transport route in the next period; and perform a non-time series prediction based on the historical filling space corresponding to the transport route to obtain a second filling space, wherein the second filling space is the filling space of the transport route at any transport unit time in the next period;

[0084] The verification module 230 is used to perform cross-validation based on the first filling space and the second filling space corresponding to each transportation unit time of the transportation route in the next cycle, so as to obtain the target predicted filling space of the transportation route for each transportation unit time in the next cycle.

[0085] The technical solution of this embodiment performs time series prediction and non-time series prediction based on the historical filling space of each transport route within a preset historical time range, obtains different prediction results through two prediction methods, and cross-validates based on the time series prediction results and the non-time series prediction results to improve the accuracy of filling space prediction.

[0086] Based on the above embodiment, optionally, the information acquisition module 210 is used to: determine multiple transportation routes and the vehicle transportation information corresponding to each of the transportation routes based on the vehicle transportation information within the preset historical time range; for any transportation route, determine the historical filling space of the transportation route within each transportation unit time based on the last vehicle transportation information of the transportation route within each transportation unit time.

[0087] Optionally, the information acquisition module 210 is also used to: determine the cargo prediction space based on the cargo volume and cargo expansion coefficient in the last vehicle transportation information; determine the historical filling space within the transportation unit time based on the vehicle model volume in the last vehicle transportation information and the cargo prediction space.

[0088] Optionally, the information acquisition module 210 is also used to: for any transport route, determine the commonly used vehicle model based on the vehicle model information in the vehicle transport information corresponding to the transport route; and correct the vehicle model volume in the last vehicle transport information based on the commonly used vehicle model volume, wherein the corrected vehicle model volume is the maximum value of the commonly used vehicle model volume and the vehicle model volume in the last vehicle transport information.

[0089] On the basis of the above embodiment, optionally, the prediction module 220 is used to: for any transport unit time within the cycle, determine the time sequence position of the transport unit time within the cycle; within multiple cycles of the preset historical time range, obtain multiple historical filling spaces corresponding to the transport line at the transport unit time based on the time sequence position; obtain the first filling space corresponding to the transport unit time of the transport line in the next cycle based on the multiple historical filling spaces corresponding to the transport line at the transport unit time and the predicted attenuation coefficient.

[0090] Optionally, the prediction module 220 is also used to: determine the average value, variance value and preset quantile value of the historical filling space based on the historical filling space corresponding to the transportation route; and obtain the second filling space based on the average value, variance value and preset quantile value of the historical filling space.

[0091] Optionally, the verification module 230 is used to: if the second filling space is smaller than the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the second filling space is determined as the target predicted filling space; if the second filling space is greater than or equal to the minimum value in the first filling space corresponding to each transport unit time in the next cycle, then the minimum value in the first filling space corresponding to each transport unit time in the next cycle is determined as the target predicted filling space.

[0092] Optionally, the information acquisition module 210 is further used to: determine the transportation frequency of the transportation route based on the vehicle transportation information corresponding to the transportation route, and perform fillable warehouse screening on the transportation route based on the transportation frequency of the transportation route; or,

[0093] Based on the vehicle transportation information corresponding to the transportation route, the transportation frequency corresponding to each unit time of the transportation route in the cycle dimension is determined, and the unit time of the transportation route is screened based on the transportation frequency corresponding to each unit time, and the screened unit time is used as the transportation unit time.

[0094] The bin-filling space prediction device provided in the embodiment of the present invention can execute the bin-filling space prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0095] Figure 4 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0096] like Figure 4As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0098] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the bin-filling space prediction method.

[0099] In some embodiments, the bin-filling spatial prediction method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the bin-filling spatial prediction method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the bin-filling spatial prediction method in any other appropriate manner (e.g., by means of firmware).

[0100] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] The computer program for implementing the bin-filling space prediction method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0102] An embodiment of the present invention further provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a processor to execute a filling space prediction method, the method comprising: obtaining vehicle transportation information within a preset historical time range, and determining historical filling spaces corresponding to multiple transportation routes based on the vehicle transportation information within the preset historical time range; for each transportation route, performing a time series prediction in a periodic dimension based on the historical filling space corresponding to the transportation route, and obtaining a first filling space corresponding to each transportation unit time of the transportation route in the next period; and performing a non-time series prediction based on the historical filling space corresponding to the transportation route to obtain a second filling space, wherein the second filling space is the filling space of any of the transportation unit time of the transportation route in the next period; cross-validating the first filling space and the second filling space corresponding to each transportation unit time of the transportation route in the next period, and obtaining a target predicted filling space for each transportation unit time of the transportation route in the next period.

[0103] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 of the foregoing.

[0104] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0105] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0106] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0107] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0108] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting bin filling space, characterized in that: include: Acquire vehicle transportation information within a preset historical time range, and determine historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range; For each transport route, a time series prediction in the period dimension is performed based on the historical filling space corresponding to the transport route, and a first filling space corresponding to each transport unit time of the transport route in the next period is obtained; and a second filling space is obtained by a non-time series prediction based on the historical filling space corresponding to the transport route, wherein the second filling space is the filling space of any transport unit time of the transport route in the next period; Based on the cross-validation of the first filling space and the second filling space corresponding to each transport unit time of the transport route in the next cycle, the target predicted filling space of the transport route for each transport unit time in the next cycle is obtained.

2. The method according to claim 1, characterized in that The determining of the historical filling spaces respectively corresponding to the plurality of transport routes based on the vehicle transport information within the preset historical time range includes: Determine a plurality of transport routes and the vehicle transport information corresponding to each of the transport routes based on the vehicle transport information within the preset historical time range; For any transport route, based on the last vehicle transport information of the transport route in each transport unit time, the historical filling space of the transport route in each transport unit time is determined.

3. The method according to claim 2, characterized in that The determining of the historical filling space of the transportation route in each transportation unit time based on the last vehicle transportation information of the transportation route in each transportation unit time includes: Determining cargo prediction space based on cargo volume and cargo expansion coefficient in the last vehicle transportation information; The historical filling space within the transportation unit time is determined based on the vehicle model volume in the last vehicle transportation information and the cargo prediction space.

4. The method according to claim 3, characterized in that The method further comprises: For any transport route, determining a commonly used vehicle model based on vehicle model information in the vehicle transport information corresponding to the transport route; The vehicle model volume in the last vehicle transportation information is corrected based on the common vehicle model volume, wherein the corrected vehicle model volume is the maximum value of the common vehicle model volume and the vehicle model volume in the last vehicle transportation information.

5. The method according to claim 1, characterized in that The time series prediction of the period dimension is performed based on the historical filling space corresponding to the transportation route to obtain the first filling space corresponding to each transportation unit time of the transportation route in the next period, including: For any transport unit time within the cycle, determine the time sequence position of the transport unit time within the cycle; within multiple cycles of the preset historical time range, obtain multiple historical filling spaces of the transport route corresponding to the transport unit time based on the time sequence position; Based on a plurality of historical filling spaces corresponding to the transport unit time of the transport route and a predicted attenuation coefficient, a first filling space corresponding to the transport unit time of the transport route in the next cycle is obtained.

6. The method according to claim 1, characterized in that The performing of non-temporal prediction based on the historical filling space corresponding to the transportation route to obtain the second filling space includes: Determine the average value, variance value and preset quantile value of the historical filling space based on the historical filling space corresponding to the transportation route; The second filling space is obtained based on the average value, variance value and preset quantile value of the historical filling space.

7. The method according to claim 1, characterized in that The method for determining the target predicted filling space of each transport unit time of the transport route in the next cycle includes: If the second filling space is smaller than the minimum value of the first filling spaces respectively corresponding to each transport unit time in the next cycle, the second filling space is determined as the target predicted filling space; If the second filling space is greater than or equal to the minimum value in the first filling space corresponding to each transport unit time in the next cycle, the minimum value in the first filling space corresponding to each transport unit time in the next cycle is determined as the target predicted filling space.

8. The method according to claim 1, characterized in that The method further comprises: Determine the transport frequency of the transport route based on the vehicle transport information corresponding to the transport route, and perform fillable warehouse screening on the transport route based on the transport frequency of the transport route; or, Based on the vehicle transportation information corresponding to the transportation route, the transportation frequency corresponding to each unit time of the transportation route in the cycle dimension is determined, and the unit time of the transportation route is screened based on the transportation frequency corresponding to each unit time, and the screened unit time is used as the transportation unit time.

9. A filling space prediction device, characterized in that: include: An information acquisition module, used to acquire vehicle transportation information within a preset historical time range, and determine historical filling spaces corresponding to a plurality of transportation routes based on the vehicle transportation information within the preset historical time range; A prediction module is used to perform a time series prediction in a period dimension for each transport route based on the historical filling space corresponding to the transport route, to obtain a first filling space corresponding to each transport unit time of the transport route in the next period; and to perform a non-time series prediction based on the historical filling space corresponding to the transport route to obtain a second filling space, wherein the second filling space is the filling space of the transport route at any transport unit time in the next period; The verification module is used to cross-validate the first filling space and the second filling space corresponding to each transportation unit time of the transportation route in the next cycle to obtain the target predicted filling space of the transportation route for each transportation unit time in the next cycle.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the bin-filling space prediction method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the bin-filling space prediction method according to any one of claims 1 to 8 when executed.

Images