CN113902197A

CN113902197A - Method and system for improving use efficiency of container based on big data application

Info

Publication number: CN113902197A
Application number: CN202111189323.6A
Authority: CN
Inventors: 唐树华; 唐建阳
Original assignee: Shenzhen Zaitogongda Technology Co ltd
Current assignee: Shenzhen Zaitogongda Technology Co ltd
Priority date: 2021-10-12
Filing date: 2021-10-12
Publication date: 2022-01-07

Abstract

The application discloses a method and a system for improving the use efficiency of a container based on big data application, which are used for acquiring the current position and first container information sent by a preset vehicle-mounted terminal; calling a preset unloading position, a loading position, a standard running speed and a standard unloading time length, and calculating a first expected time point; acquiring a second time point and second container information; if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance; if the idle load distance is larger than the distance threshold value, acquiring a plurality of sensing data sequences, and inputting the sensing data sequences into a preset first fatigue state prediction model to obtain a first fatigue value; calling a correction parameter table and a standard fatigue threshold value; acquiring a specified correction parameter, and calculating a corrected fatigue threshold value; if the first fatigue value is less than the corrected fatigue threshold value, a continuous driving instruction is generated, the utilization efficiency of the container is improved, and the safety of the container is improved.

Description

Method and system for improving use efficiency of container based on big data application

Technical Field

The application relates to the field of container transportation, in particular to a method and a system for improving the use efficiency of a container based on big data application.

Background

Freight container transportation is a special form of the freight industry, where the goods carried by vehicles are containers, and the containers may also contain goods. Containers are of uniform size and type, and there are also clear requirements for the type of goods that can be loaded. Containers are not typically co-transported with other cargo in the same vehicle. The goods loaded in the container can not be loaded and unloaded at any time in the transportation process, and the goods can only be loaded and unloaded when arriving at the unloading point or the loading point.

In order to improve the use efficiency of the container, one method is to optimize the loading and unloading mode of the container and reduce the loading and unloading time. For example, a new container loading and unloading device is designed for a container box type high-efficiency loading and unloading device (CN211110883), the loading and unloading efficiency of the container is improved, and the method can accelerate the loading and unloading speed of the container or facilitate the use of the container. Another method is to change the shape of the container to facilitate transportation. For example, a double-split container wing cabinet (CN213139920) is designed to be a double-split container wing cabinet, which is convenient for splicing two container cabinets together.

There are two typical modes of operation for container transportation. One is import business, which is characterized in that the empty container is removed again, namely import cabinet container transportation business, a vehicle transports a container from a wharf to a certain unloading point, the container returns to a storage yard after unloading, and the container is empty in the process of returning to the storage yard; another is export traffic, which is characterized by empty go-and-return, i.e. export container traffic, where vehicles pick up empty containers from the yard, arrive at the loading point and are transported to the dock after loading, where the containers are empty before loading.

Although the above-described method of optimizing the loading and unloading of the container or changing the form of the container can increase the speed of the container transportation operation, it does not contribute to the efficiency of the use of the container during the empty-going and empty-returning operations, and therefore, the efficiency of the use of the container during the container transportation process still needs to be improved.

Disclosure of Invention

The application provides a method for improving the use efficiency of a container based on big data application, which comprises the following steps:

s1, acquiring the current position and the first container information sent by the preset vehicle-mounted terminal; the vehicle corresponding to the vehicle-mounted terminal goes to a preset unloading factory from a first wharf, and the first container information at least comprises the type of a container on the vehicle;

s2, calling preset unloading positions, loading positions, the standard running speed and the standard unloading time length of the vehicle, and calculating a first predicted time point of the vehicle according to the current position, the unloading positions, the loading positions, the standard running speed and the standard unloading time length; wherein, the first predicted time point refers to the predicted arrival time of the vehicle to the loading factory after being unloaded by the unloading factory; the unloading position refers to the position of an unloading factory, and the loading position refers to the position of a loading factory;

s3, acquiring a second time point and second container information sent by a loading factory terminal, judging whether the second time point is later than the first expected time point, and judging whether the second container information is matched with the first container information;

s4, if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance between the unloading position and the loading position, and judging whether the no-load distance is larger than a preset distance threshold value;

s5, if the idle load distance is larger than a preset distance threshold, acquiring a plurality of sensing data sequences obtained by sensing by a plurality of sensors arranged in the container in advance, inputting the plurality of sensing data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value output by the first fatigue state prediction model;

s6, calling a preset correction parameter table and a standard fatigue threshold value; wherein, the correction parameter table records the corresponding relation between the transportation distance of the empty container and the correction parameter in the transportation process of the heavy container, the empty container and the heavy container;

s7, obtaining the designated correction parameter corresponding to the no-load distance according to the correction parameter table, and according to a formula: the corrected fatigue threshold value is defined as correction parameter multiplied by standard fatigue threshold value, and the corrected fatigue threshold value is calculated; and judging whether the first fatigue value is smaller than the corrected fatigue threshold value;

and S8, if the first fatigue value is less than the corrected fatigue threshold value, generating a continuous driving instruction to instruct continuous driving through the first dock, the unloading dock, the loading dock and the second dock in sequence on the premise of not replacing the driver and the vehicle.

Further, the step S2 of retrieving the preset unloading position, loading position, standard driving speed of the vehicle and standard unloading time length, and calculating the first predicted time point of the vehicle according to the current position, unloading position, loading position, standard driving speed and standard unloading time length includes:

s201, calling preset unloading positions, loading positions, the standard running speed of the vehicle and the standard unloading time length;

s202, acquiring a first driving route between the current position and the unloading position, and acquiring a second driving route between the current position and the unloading position;

s203, adding the length of the first running route and the length of the second running route to obtain a running length, and dividing the running length by a standard running speed to obtain a running duration;

and S204, adding the current time, the running time and the standard unloading time length to obtain a first expected time point.

Further, before step S5, the method for predicting the fatigue state of the container includes that the first fatigue state prediction model is trained in a supervised learning manner based on a preset neural network model, and if the empty load distance is greater than a preset distance threshold, multiple sensing data sequences sensed by multiple sensors pre-arranged in the container are obtained, and the multiple sensing data sequences are input into the preset first fatigue state prediction model for processing, so as to obtain a first fatigue value output by the first fatigue state prediction model, the method includes:

s41, calling a plurality of sample sensing data sequences from a preset database; the system comprises a sample vehicle, a sample sensor, a data sequence acquisition unit and a data sequence acquisition unit, wherein the sample sensing data sequence is obtained by sensing sample sensors arranged on sample containers on the sample vehicle, and the sample vehicle is transporting the containers filled with goods when the sample sensing data sequence is acquired;

s42, manually labeling the multiple sample sensing data sequences to label different fatigue values;

s43, dividing the marked multiple sample sensing data sequences into multiple training sensing data sequences and multiple verification sensing data sequences according to a preset proportion;

s44, inputting the multiple training sensing data sequences into a preset neural network model for training to obtain a temporary fatigue state prediction model;

s45, verifying the temporary fatigue state prediction model by adopting the plurality of sensing data sequences for verification to obtain a verification result, and judging whether the verification result is passed;

and S46, if the verification result is that the verification is passed, recording the temporary fatigue state prediction model as a first fatigue state prediction model.

Further, step S5, where, if the empty-load distance is greater than the preset distance threshold, acquiring multiple sensing data sequences sensed by multiple sensors pre-arranged in the container, and inputting the multiple sensing data sequences into a preset first fatigue state prediction model for processing, so as to obtain a first fatigue value output by the first fatigue state prediction model, includes:

s501, if the no-load distance is larger than a preset distance threshold, acquiring a plurality of sensing data sequences obtained by sensing of a plurality of sensors distributed in advance in the container; the number of the multiple sensing data sequences is n, and n is an integer greater than 1;

s502, inputting the multiple sensing data sequences into a preset first fatigue state prediction model, so that the first fatigue state prediction model is according to a formula:

sequentially calculating n influence curve functions; wherein wi (t) is an ith influence curve function, Pi (t) is a curve of a sensing value of the ith sensing data sequence along with the change of time, pi (t) is a preset standard sensing data curve corresponding to Pi (t), qi is a preset ith deviation threshold value, and qi is larger than 0;

s503, obtaining a first time window length of the influence curve function wi (t) which is equal to qi, obtaining a total time length of the influence curve function wi (t), and calculating a time ratio Bi of the first time window length divided by the total time length so as to obtain n time ratios respectively corresponding to the n influence curve functions;

s504, according to a formula:

calculating a first fatigue value H; wherein K is a preset standard fatigue value.

Further, the step S8 of generating a continuous driving instruction to instruct continuous driving through the first terminal, the unloading terminal, the loading terminal and the second terminal in sequence without replacing the driver and the vehicle, if the first fatigue value is less than the corrected fatigue threshold value, includes:

s801, if the first fatigue value is smaller than a correction fatigue threshold value, adopting a preset in-vehicle camera to collect and process the face of a vehicle driver to obtain a face image;

s802, inputting the face image into a preset second fatigue state prediction model for processing to obtain a second fatigue value output by the second fatigue state prediction model; the second fatigue state prediction model is obtained by training based on a deep convolutional neural network model;

s803, judging whether the second fatigue value is smaller than a corrected fatigue threshold value;

and S804, if the second fatigue value is smaller than the corrected fatigue threshold value, generating a continuous driving instruction to indicate that continuous driving is performed through the first wharf, the unloading factory, the loading factory and the second wharf in sequence on the premise of not replacing the driver and the vehicle.

The application provides a system for improving container availability factor based on big data application includes:

the first container information acquisition unit is used for indicating and acquiring the current position and the first container information sent by a preset vehicle-mounted terminal; the vehicle corresponding to the vehicle-mounted terminal goes to a preset unloading factory from a first wharf, and the first container information at least comprises the type of a container on the vehicle;

the first estimated time point calculation unit is used for indicating and calling a preset unloading position, a loading position, the standard running speed and the standard unloading time length of the vehicle, and calculating a first estimated time point of the vehicle according to the current position, the unloading position, the loading position, the standard running speed and the standard unloading time length; wherein, the first predicted time point refers to the predicted arrival time of the vehicle to the loading factory after being unloaded by the unloading factory; the unloading position refers to the position of an unloading factory, and the loading position refers to the position of a loading factory;

a first estimated time point judgment unit, configured to instruct to acquire a second time point and second container information sent by a loading factory terminal, judge whether the second time point is later than the first estimated time point, and judge whether the second container information matches the first container information;

the empty load distance judging unit is used for indicating that if the second time point is later than the first time point and the second container information is matched with the first container information, the empty load distance between the unloading position and the loading position is calculated, and whether the empty load distance is larger than a preset distance threshold value is judged;

a first fatigue value obtaining unit, configured to indicate that, if the empty-load distance is greater than a preset distance threshold, multiple sensing data sequences obtained by sensing by multiple sensors pre-arranged in the container are obtained, and the multiple sensing data sequences are input into a preset first fatigue state prediction model for processing, so as to obtain a first fatigue value output by the first fatigue state prediction model;

the standard fatigue threshold value calling unit is used for indicating to call a preset correction parameter table and a standard fatigue threshold value; wherein, the correction parameter table records the corresponding relation between the transportation distance of the empty container and the correction parameter in the transportation process of the heavy container, the empty container and the heavy container;

a first fatigue value judging unit, configured to instruct to obtain, according to the correction parameter table, a specified correction parameter corresponding to the no-load distance, according to a formula: the corrected fatigue threshold value is defined as correction parameter multiplied by standard fatigue threshold value, and the corrected fatigue threshold value is calculated; and judging whether the first fatigue value is smaller than the corrected fatigue threshold value;

and the continuous driving instruction generating unit is used for generating a continuous driving instruction to instruct continuous driving sequentially through the first wharf, the unloading factory, the loading factory and the second wharf on the premise of not replacing the driver and the vehicle if the first fatigue value is smaller than the corrected fatigue threshold value.

The present application provides a computer device comprising a memory storing a computer program and a processor implementing the steps of any of the above methods when the processor executes the computer program.

The present application provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of any of the above.

The method, the system, the computer equipment and the storage medium for improving the use efficiency of the container based on big data application acquire the current position and the first container information sent by the preset vehicle-mounted terminal; calling a preset unloading position, a preset loading position, a standard running speed and a standard unloading time length of the vehicle, and calculating a first expected time point of the vehicle; acquiring a second time point and second container information sent by a loading factory terminal; if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance; if the no-load distance is larger than a preset distance threshold value, acquiring a plurality of sensing data sequences, and inputting the sensing data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value; calling a preset correction parameter table and a standard fatigue threshold value; acquiring a designated correction parameter corresponding to the no-load distance, and calculating a corrected fatigue threshold value; if the first fatigue value is smaller than the corrected fatigue threshold value, a continuous driving instruction is generated, the utilization efficiency of the container is improved, and the safety of the container in the transportation process is improved.

This application has realized two technological effects, has improved the utilization efficiency of container promptly to and the security in the container transportation has been improved. In contrast to the aforementioned problem of the efficiency of the container transportation process, which is still to be improved, the present application achieves two technical effects, and these two technical effects are progressive due to the particularities of the freight container transportation. Specifically, the method comprises the following steps:

the application adopts the heavy-load-removing and heavy-load-returning mode (the process of returning empty containers to a storage yard is avoided) to improve the utilization efficiency of the containers, but after the heavy-load-removing and heavy-load-returning mode is adopted, the container vehicles are special vehicles which are more difficult to drive compared with common vehicles, after the heavy-load-removing and heavy-load-returning scheme is adopted, drivers move heavy vehicles (heavy-load container vehicles) from a wharf A to a factory B, after the containers are unloaded from the factory B, empty vehicles (loaded with empty containers) move to a factory C, the containers are loaded in the factory C to become heavy vehicles, and then the heavy vehicles move to a wharf D (the wharf D and the wharf A can be the same or different), so the adopted heavy-load-removing and heavy-returning scheme is actually three stages including heavy vehicles, empty vehicles and heavy vehicles.

The container vehicle is a special vehicle, the driving between the empty load and the heavy load is greatly different, in three stages of heavy vehicle-empty vehicle-heavy vehicle, if the empty vehicle is longer (certainly not extremely long, if extremely long, the driver is easy to adapt to the change), the driver must switch the driving habit between the empty load and the heavy load (if the driver continuously drives for a plurality of times in three stages, the switching is more than twice, and for the frequent switching, the safety is lower), and the driving habit of the driver has certain inertia, so the misoperation of the driver is easily caused, and the accident rate is improved. (if the conventional scheme, i.e., the scheme of returning the empty box to the yard, is used, the empty time of the driver can be considered to be extremely long, and thus there is enough time to accommodate, so that this problem does not occur).

Therefore, although the application is apparently directed to one technical problem, two progressive technical problems are actually addressed, so that the application also has two progressive technical effects, namely, the utilization efficiency of the container is improved, and the safety during the transportation process of the container is improved.

Drawings

FIG. 1 is a schematic flow chart illustrating a method for improving container utilization efficiency based on big data application according to an embodiment of the present application;

FIG. 2 is a block diagram illustrating an exemplary embodiment of a big data based system for improving container usage efficiency;

fig. 3 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.

The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Referring to fig. 1, an embodiment of the present application provides a method for improving container usage efficiency based on big data application, including the following steps:

As described in the above steps S1-S4, the current position and the first container information sent by the preset vehicle-mounted terminal are acquired; the vehicle corresponding to the vehicle-mounted terminal goes to a preset unloading factory from a first wharf, and the first container information at least comprises the type of a container on the vehicle; calling a preset unloading position, a preset loading position, a standard running speed and a standard unloading time length of the vehicle, and calculating a first predicted time point of the vehicle according to the current position, the unloading position, the preset loading position, the standard running speed and the standard unloading time length; wherein, the first predicted time point refers to the predicted arrival time of the vehicle to the loading factory after being unloaded by the unloading factory; the unloading position refers to the position of an unloading factory, and the loading position refers to the position of a loading factory; acquiring a second time point and second container information sent by a loading factory terminal, judging whether the second time point is later than the first expected time point or not, and judging whether the second container information is matched with the first container information or not; and if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance between the unloading position and the loading position, and judging whether the no-load distance is greater than a preset distance threshold value.

The method integrates the business of the import and export into a new business, so that the vehicle does not return to a storage yard after the import freight container is unloaded, and directly goes to the loading point of the export freight container for loading, and then is transported to a wharf, thereby realizing seamless butt joint of the containers in the business of the import and export, repeating the process and returning, and improving the utilization rate of the containers.

Therefore, the current position of the vehicle, the first container information, the unloading position, the loading position, the standard driving speed of the vehicle and the standard unloading time length from the first dock to the preset unloading factory are determined, which is the basis for determining whether the vehicle can be driven continuously in three stages of heavy driving, empty driving and heavy driving (i.e. heavy back-and-forth), wherein the first heavy driving stage refers to the heading from the dock to the unloading factory, the middle empty driving stage refers to the heading from the unloading factory to the loading factory, and the second heavy driving stage refers to the heading from the loading factory to the dock.

The step S2 of retrieving preset unloading position, loading position, standard driving speed of the vehicle and standard unloading time length, and calculating a first predicted time point of the vehicle according to the current position, unloading position, loading position, standard driving speed and standard unloading time length includes:

Thus assuming that the vehicle is able to travel to the loading plant, the first expected point in time it will arrive. The first expected time point is calculated because the loading factory has requirements for the container and the loading time, and if the loading time is earlier and the vehicle cannot arrive before the loading time, the time cannot be matched, so that the strategy of performing the repeat returning is not suitable, and of course, whether the container is matched or not needs to be considered, which will be determined in the subsequent steps. The standard unloading time length refers to a standard time length required for unloading the goods of the container on the vehicle according to a standard unloading process.

And then acquiring a second time point and second container information sent by a loading factory terminal, judging whether the second time point is later than the first expected time point, and judging whether the second container information is matched with the first container information. Because the containers have different types and can load different goods, it is necessary to determine whether the second container information matches the first container information, and determining whether the second container information matches the first container information may be implemented by any feasible method, for example, determining whether the container type in the second container information is the same as the container type in the first container information, and if so, determining that the second container information matches the first container information. Of course, other further information may be considered, such as determining whether the container owners of the import/export transaction request are the same, and if so, further determining a match

If the second time point is later than the first time point and the second container information is matched with the first container information, the container vehicle is indicated to be suitable for re-repeat-back transportation, so that the no-load distance between the unloading position and the loading position is calculated, and whether the no-load distance is larger than a preset distance threshold value or not is judged. And the judgment of the idle load distance is carried out, aiming at determining whether the transportation of the heavy load-removing and heavy load-returning can cause larger burden to a driver, because the three stages of heavy load-removing, empty load-removing and heavy load-removing are involved in the transportation of the heavy load-removing and heavy load-removing, an applicant finds that if the special vehicle of the container vehicle is in the driving process of the heavy load-removing and heavy load-removing, the burden of thousands of fears to the driver is lighter if the special vehicle is in the driving state of the heavy load-removing or empty load-removing, and the three stages of heavy load-removing, empty load-removing and heavy load-removing are involved, the driver needs to switch driving habits twice, which can cause extra burden, so that the driver is more fatigued and the safety is reduced. If the driving state of the empty vehicle is extremely short, for example, the unloading factory is adjacent to the loading factory, the burden can be ignored, and if the driving state of the empty vehicle is long, the influence of the driving on the driver needs to be considered. Therefore, the determination of the empty distance is performed in order to determine whether or not the driver is suitable for starting the continuous driving for the re-drive and the determination of the container information and the like in the foregoing is performed in order to determine whether or not the vehicle is suitable for the driving for the re-drive and the re-return.

Further, the container can be inspected at an unloading factory, and if defects exist, the container cannot be subjected to packing and transportation of export goods.

As described in the above steps S5-S8, if the empty-load distance is greater than the preset distance threshold, obtaining a plurality of sensing data sequences sensed by a plurality of sensors pre-arranged in the container, and inputting the plurality of sensing data sequences into a preset first fatigue state prediction model for processing, so as to obtain a first fatigue value output by the first fatigue state prediction model; calling a preset correction parameter table and a standard fatigue threshold value; wherein, the correction parameter table records the corresponding relation between the transportation distance of the empty container and the correction parameter in the transportation process of the heavy container, the empty container and the heavy container; obtaining a designated correction parameter corresponding to the no-load distance according to the correction parameter table, and according to a formula: the corrected fatigue threshold value is defined as correction parameter multiplied by standard fatigue threshold value, and the corrected fatigue threshold value is calculated; and judging whether the first fatigue value is smaller than the corrected fatigue threshold value; if the first amount of fatigue is less than the corrected threshold amount of fatigue, a continuous driving instruction is generated to instruct continuous driving through the first dock, the unloading factory, the loading factory, and the second dock in sequence without replacing the driver and the vehicle.

If the idle distance is larger than the preset distance threshold value, the current state of the driver needs to be considered, and the idle distance is digitalized, namely the first fatigue value of the driver. Since the driving state of a driver can reflect the driving process, the container is internally pre-provided with a plurality of sensors so as to sense data when the driver drives the vehicle. When the driver with sufficient energy drives the vehicle, the sensed data should be non-abrupt, such as abnormal acceleration or abnormal deceleration, and should be driven at a constant speed as much as possible. These sensory data sequences can therefore reflect the driving state of the driver, i.e. the first fatigue value. Each sensory data sequence is a time sequence. The first fatigue state prediction model can be obtained in any feasible way, for example, a suitable prediction model is trained on the basis of a neural network model; or a non-machine learning mode is adopted, and the specific analysis model of the application is adopted to predict the first fatigue value. Wherein the sensors are for example speed sensors, pressure sensors, acceleration sensors, etc., it is noted that the sensors of the present application are arranged on the container, not directly on the vehicle, since the present application relates to special vehicles, and the sensors arranged on the container are more representative of the reaction of the driver's state.

The plurality of sensing data sequences may be obtained by transmitting via a vehicle-mounted terminal, or a signal transmission module may be provided on the container in advance, and the signal transmission module may be used to transmit the plurality of sensing data sequences.

The driver's state, the tired value is special in this application promptly, and special reason lies in, in ordinary driving process, the driver's degree of easily tired out is different with the degree of easily tired out of the three stage of heavy truck, empty wagon, heavy truck in this application, and the driver is changeed tired out in this application, consequently, the first tired out numerical value of this application preliminary prediction, only be applicable to ordinary driving task, for example, the simple heavy truck driving task, but be difficult to directly be applied to in the scene of this application. To overcome this problem, the present application employs a way of correcting the fatigue threshold. Specifically, the method and the device have the advantages that the driver drives the pure heavy vehicle in advance (sequentially through the first wharf, the unloading factory, the loading factory and the second wharf), and then the fatigue value in the process is manually marked to determine the standard fatigue threshold value, so that the standard fatigue threshold value is a value determined manually, and the probability of accidents occurring below the value is high. Of course, the standard fatigue threshold may also be calculated by collecting historical accident values due to excessive driver fatigue, determining the status of the driver on similar driving routes, and then counting the standard fatigue threshold. Wherein the standard fatigue threshold corresponds to a total drive length of the first quay, the unloading factory, the loading factory, and the second quay.

In the present application, it is not appropriate to simply apply the standard fatigue threshold value, because in the transportation process of the heavy container-empty container-heavy container in the present application, the transportation distance of the empty container is related to the energy to be consumed by the driver, which is determined by the natural attribute of human, when the transportation distance of the empty container is long enough, the driver switches the driving habit to have enough buffering time, so the correction parameter is small, in the limit case, the driving distance of the heavy container is about 0, and when the transportation distance of the empty container is about equal to all routes, the correction parameter should be about 1; in the other limit case, the distance traveled by a heavy container is approximately equal to the full distance, and the distance traveled by an empty container is approximately equal to 0, the correction parameter should also be approximately equal to 1, but between these two limit cases the correction parameter should be raised. Therefore, if the correction parameter table is plotted with the horizontal axis representing the empty container transport distance and the vertical axis representing the value of the correction parameter, the curve image should have a parabolic shape. Of course, all values of the correction parameters should be 1 or more.

Further, if the idle distance is greater than a preset distance threshold, it may be further determined whether the idle distance is less than a preset second distance threshold; if the distance is smaller than the preset second distance threshold, generating a sensing data sequence acquisition instruction to indicate

And acquiring a plurality of sensing data sequences sensed by a plurality of sensors pre-arranged in the container. This is also the conclusion that can be drawn from the correction parameter table.

Therefore, a preset correction parameter table and a standard fatigue threshold value are called; and then according to the correction parameter table, acquiring a designated correction parameter corresponding to the no-load distance, and according to a formula: the corrected fatigue threshold value is defined as correction parameter multiplied by standard fatigue threshold value, and the corrected fatigue threshold value is calculated; and determining whether the first fatigue value is less than a corrected fatigue threshold. If the first amount of fatigue is less than the corrected threshold amount of fatigue, a continuous driving instruction is generated to instruct continuous driving through the first dock, the unloading factory, the loading factory, and the second dock in sequence without replacing the driver and the vehicle.

If the first amount of fatigue is less than the corrected fatigue threshold, it indicates that the driver is more energetic and thus continuous driving is possible.

Further, if the first amount of fatigue is not less than the corrected fatigue threshold, then the replacement driver may be considered to continue driving through the first dock, the unloading facility, the loading facility, and the second dock in sequence (but in a manner that would allow the new driver to drive one unfamiliar vehicle and therefore not be the optimal choice), or to replace another container and resume the above-described process of the present application.

The first wharf and the second wharf may be the same or different. The second quay is an exit quay.

Thus, a model capable of being qualified for fatigue value prediction is trained in a machine learning manner. Further, the present application may adopt other neural network models, and as the present application is applied to timing prediction, it may also adopt a long-term and short-term memory network model, or adopt a confrontation network model, a residual error network model, etc. The plurality of sample sensing data sequences should have the same one-to-one correspondence with the types of the plurality of sensing data sequences described earlier in this application. The method adopts a supervised learning mode, so that the multiple sample sensing data sequences are manually labeled to label different fatigue values. Theoretically, if mutation occurs frequently in the sample sensing data sequence, the fatigue value is high, which is a marking basis, or the difference of the states of the driver before and after driving can be observed as a basis for manual marking. But in any case the sample sensory data sequence has a correspondence to a person's fatigue value, which can be determined. And then training by adopting homologous training data, and then verifying by adopting homologous verification data so as to ensure the reliability of the finally obtained first fatigue state prediction model in actual use.

s504, according to a formula:

The present application makes a prediction of the first fatigue value in a particular manner. Because the sensing data sequence reflects the state of a driver, the method and the device collect a standard sensing data curve in advance, and then determine the difference between the standard sensing data curve and a real sensing data curve, so that the standard sensing data curve can be used as a basis for analyzing fatigue values. The standard sensing data curve is obtained by sensing with a sensor having the same position during the driving process of a driver with abundant rest on the same way (since the destination is fixed, the pre-measurement is easy to realize). Of course, in order to adapt to different road conditions, the method can be used for driving in advance for multiple times to determine multiple standard sensing data curves, and during actual analysis, one standard sensing data curve most similar to the actual curve is selected as a comparison curve. The standard fatigue value is obtained by manually analyzing the state of a driver with abundant rest. This application uses the formula:

and n influence curve functions are sequentially calculated, and the difference between two curves (a real curve and a standard curve) can be determined by integrating the numerical value and the numerical value change trend. The value of the influence curve function wi (t) is equal to the length of the first time window of qi, which indicates the length of time that the actual curve is not similar to the standard curve, and the larger the value of the influence curve function wi (t), the worse the driving state of the driver is, the larger the fatigue value is. Then according to the formula:

a first fatigue value H is calculated.

Therefore, a double judgment mode is adopted to improve the prediction accuracy of the fatigue value. The first fatigue value is analyzed based on a plurality of sensing data sequences sensed by the sensor on the container, and the fatigue state is reflected on the facial expression of the driver, so that the application further collects facial images and inputs the facial images into the second fatigue state prediction model for processing to obtain a second fatigue value output by the second fatigue state prediction model. The second fatigue state prediction model is trained based on the deep convolutional neural network model, since the deep convolutional neural network model is suitable for analyzing the picture data. The deep convolutional neural network model comprises an input layer, a convolutional layer, a pooling layer, a full-link layer and an output layer. If the second amount of fatigue is less than the corrected fatigue threshold, it can be further determined that the driver is still more energetic, and thus a continuous driving instruction is generated to instruct continuous driving sequentially through the first dock, the unloading factory, the loading factory, and the second dock without replacing the driver and the vehicle.

According to the method for improving the use efficiency of the container based on big data application, the current position and the first container information sent by the preset vehicle-mounted terminal are obtained; calling a preset unloading position, a preset loading position, a standard running speed and a standard unloading time length of the vehicle, and calculating a first expected time point of the vehicle; acquiring a second time point and second container information sent by a loading factory terminal; if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance; if the no-load distance is larger than a preset distance threshold value, acquiring a plurality of sensing data sequences, and inputting the sensing data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value; calling a preset correction parameter table and a standard fatigue threshold value; acquiring a designated correction parameter corresponding to the no-load distance, and calculating a corrected fatigue threshold value; if the first fatigue value is smaller than the corrected fatigue threshold value, a continuous driving instruction is generated, the utilization efficiency of the container is improved, and the safety of the container in the transportation process is improved.

Referring to fig. 2, an embodiment of the present application provides a system for improving container usage efficiency based on big data application, including:

a first container information acquiring unit 10, configured to instruct to acquire a current location and first container information sent by a preset vehicle-mounted terminal; the vehicle corresponding to the vehicle-mounted terminal goes to a preset unloading factory from a first wharf, and the first container information at least comprises the type of a container on the vehicle;

a first estimated time point calculation unit 20, configured to instruct to call a preset unloading position, a loading position, a standard driving speed of the vehicle, and a standard unloading time length, and calculate a first estimated time point of the vehicle according to the current position, the unloading position, the loading position, the standard driving speed, and the standard unloading time length; wherein, the first predicted time point refers to the predicted arrival time of the vehicle to the loading factory after being unloaded by the unloading factory; the unloading position refers to the position of an unloading factory, and the loading position refers to the position of a loading factory;

a first estimated time point judgment unit 30 configured to instruct acquisition of a second time point and second container information transmitted by a loading factory terminal, judge whether the second time point is later than the first estimated time point, and judge whether the second container information matches the first container information;

an empty distance determining unit 40, configured to instruct, if the second time point is later than the first time point and the second container information matches the first container information, to calculate an empty distance between the unloading position and the loading position, and determine whether the empty distance is greater than a preset distance threshold;

a first fatigue value obtaining unit 50, configured to indicate that, if the empty-load distance is greater than a preset distance threshold, multiple sensing data sequences obtained by sensing by multiple sensors pre-arranged in the container are obtained, and the multiple sensing data sequences are input into a preset first fatigue state prediction model for processing, so as to obtain a first fatigue value output by the first fatigue state prediction model;

a standard fatigue threshold value retrieval unit 60 for instructing retrieval of a preset correction parameter table and a standard fatigue threshold value; wherein, the correction parameter table records the corresponding relation between the transportation distance of the empty container and the correction parameter in the transportation process of the heavy container, the empty container and the heavy container;

a first fatigue value judging unit 70, configured to instruct to obtain, according to the correction parameter table, a specified correction parameter corresponding to the empty load distance, according to a formula: the corrected fatigue threshold value is defined as correction parameter multiplied by standard fatigue threshold value, and the corrected fatigue threshold value is calculated; and judging whether the first fatigue value is smaller than the corrected fatigue threshold value;

a continuous driving instruction generating unit 80, configured to instruct, if the first fatigue value is smaller than the corrected fatigue threshold value, to generate a continuous driving instruction to instruct continuous driving sequentially via the first dock, the unloading factory, the loading factory, and the second dock without replacing the driver and the vehicle.

The operations performed by the units are corresponding to the steps of the method for improving the use efficiency of the container based on big data application in the foregoing embodiment one by one, and are not described herein again.

The system for improving the use efficiency of the container based on big data application acquires the current position and the first container information sent by the preset vehicle-mounted terminal; calling a preset unloading position, a preset loading position, a standard running speed and a standard unloading time length of the vehicle, and calculating a first expected time point of the vehicle; acquiring a second time point and second container information sent by a loading factory terminal; if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance; if the no-load distance is larger than a preset distance threshold value, acquiring a plurality of sensing data sequences, and inputting the sensing data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value; calling a preset correction parameter table and a standard fatigue threshold value; acquiring a designated correction parameter corresponding to the no-load distance, and calculating a corrected fatigue threshold value; if the first fatigue value is smaller than the corrected fatigue threshold value, a continuous driving instruction is generated, the utilization efficiency of the container is improved, and the safety of the container in the transportation process is improved.

Referring to fig. 3, an embodiment of the present invention further provides a computer device, where the computer device may be a server, and an internal structure of the computer device may be as shown in the figure. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used for storing data used by a method for improving the use efficiency of the container based on big data application. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method for improving container usage efficiency based on big data usage. The computer device further comprises a display screen and an input device for displaying the human interactive interface and for receiving input data, respectively.

The processor executes the method for improving the use efficiency of the container based on big data application, wherein the steps included in the method correspond to the steps of executing the method for improving the use efficiency of the container based on big data application in the foregoing embodiment one to one, and are not described herein again.

It will be understood by those skilled in the art that the structures shown in the drawings are only block diagrams of some of the structures associated with the embodiments of the present application and do not constitute a limitation on the computer apparatus to which the embodiments of the present application may be applied.

The computer equipment acquires the current position and the first container information sent by a preset vehicle-mounted terminal; calling a preset unloading position, a preset loading position, a standard running speed and a standard unloading time length of the vehicle, and calculating a first expected time point of the vehicle; acquiring a second time point and second container information sent by a loading factory terminal; if the second time point is later than the first time point and the second container information is matched with the first container information, calculating the no-load distance; if the no-load distance is larger than a preset distance threshold value, acquiring a plurality of sensing data sequences, and inputting the sensing data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value; calling a preset correction parameter table and a standard fatigue threshold value; acquiring a designated correction parameter corresponding to the no-load distance, and calculating a corrected fatigue threshold value; if the first fatigue value is smaller than the corrected fatigue threshold value, a continuous driving instruction is generated, the utilization efficiency of the container is improved, and the safety of the container in the transportation process is improved.

An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored thereon, and when the computer program is executed by a processor, the method for improving the use efficiency of the container based on big data application is implemented, where steps included in the method are respectively in one-to-one correspondence with steps of the method for improving the use efficiency of the container based on big data application, which is implemented in the foregoing embodiments, and are not described herein again.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, system, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, system, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, system, article, or method that includes the element.

The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.

Claims

1. A method for improving the use efficiency of a container based on big data application is characterized by comprising the following steps:

2. The big data exercise based container use efficiency improving method according to claim 1, wherein the step S2 of retrieving the preset unloading position, loading position, standard driving speed of the vehicle and standard unloading time length, and calculating the first predicted time point of the vehicle according to the current position, unloading position, loading position, standard driving speed and standard unloading time length comprises:

3. The big-data-application-based method for improving container usage efficiency according to claim 1, wherein the first fatigue state prediction model is based on a preset neural network model and is trained in a manner of supervised learning, and before step S5, the method further comprises, if the empty-load distance is greater than a preset distance threshold, acquiring a plurality of sensing data sequences sensed by a plurality of sensors arranged in the container in advance, and inputting the plurality of sensing data sequences into the preset first fatigue state prediction model for processing to obtain a first fatigue value output by the first fatigue state prediction model:

4. The big-data-based method for improving container usage efficiency according to claim 1, wherein the step S5 of obtaining a plurality of sensor data sequences sensed by a plurality of sensors disposed in advance in the container and inputting the plurality of sensor data sequences into a preset first fatigue state prediction model for processing to obtain a first fatigue value output by the first fatigue state prediction model if the empty-load distance is greater than the preset distance threshold comprises:

s504, according to a formula:

calculate the firstFatigue value H; wherein K is a preset standard fatigue value.

5. The big-data-usage-based method for improving container usage efficiency according to claim 1, wherein if the first fatigue value is less than the corrected fatigue threshold, generating a continuous-driving instruction to instruct continuous driving through the first terminal, the unloading factory, the loading factory, and the second terminal in sequence without replacing the driver and the vehicle at step S8 comprises:

6. A system for improving container usage efficiency based on big data applications, comprising:

7. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 5.