CN117709830B - Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology - Google Patents
Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology Download PDFInfo
- Publication number
- CN117709830B CN117709830B CN202410161877.2A CN202410161877A CN117709830B CN 117709830 B CN117709830 B CN 117709830B CN 202410161877 A CN202410161877 A CN 202410161877A CN 117709830 B CN117709830 B CN 117709830B
- Authority
- CN
- China
- Prior art keywords
- controllable
- data
- metamorphic
- variable
- proportion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000013068 supply chain management Methods 0.000 title claims abstract description 32
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 27
- 238000005516 engineering process Methods 0.000 title claims description 26
- 238000012549 training Methods 0.000 claims abstract description 37
- 230000008569 process Effects 0.000 claims abstract description 34
- 230000001105 regulatory effect Effects 0.000 claims abstract description 5
- 230000006866 deterioration Effects 0.000 claims description 29
- 238000003860 storage Methods 0.000 claims description 20
- 230000015654 memory Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 8
- 238000013480 data collection Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 5
- 230000001276 controlling effect Effects 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent supply chain management system and method realized by artificial intelligence and the internet of things, which relate to the technical field of supply chain management, and are used for collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance, training a metamorphic proportion prediction model, collecting variable characteristic data for each batch of goods to be transported, generating recommended controllable characteristic data of the goods to be transported, transporting the goods to be transported by using the recommended controllable characteristic data, collecting regulation and control condition data in real time, judging whether the recommended controllable characteristic data needs to be regulated or not based on the regulation and control condition data, and if so, regulating the recommended controllable characteristic data by using an Actor-Critic network model through the internet of things; the transportation quality of the commodity supply chain is improved by means of the pre-parameter regulation and the self-adaptive parameter regulation in the transportation process.
Description
Technical Field
The invention relates to the technical field of supply chain management, in particular to an intelligent supply chain management system and method realized by an artificial intelligence and internet of things technology.
Background
In conventional commodity supply chain management, the influence of environmental factors on commodity quality in the transportation process is a long-standing problem puzzling the industry. Particularly, in the transportation of sensitive commodities such as temperature, humidity and the like, the problems of quality reduction, spoilage and the like of the commodities in the transportation process are often caused by the fact that the change of different environmental conditions cannot be effectively dealt with. These problems not only affect the market value of the commodity, but also present significant economic loss and brand reputation risks to the supply chain manager.
With the development of the internet of things technology, at present, some sensors can be installed in a storage space to remotely monitor and remotely control environmental factors such as temperature and humidity in the storage space, however, the networked sensors can only realize the function of collecting data, but cannot provide control suggestions of the temperature and the humidity based on the collected data, so that the sensors only have control functions and have no control targets;
The Chinese patent with the authority bulletin number of CN116823068B discloses a restaurant food supply chain management system, which belongs to the technical field of data processing and comprises a food storage node matching unit, a food supply label generating unit and a food supply label analyzing unit; the food storage node matching unit is used for acquiring production information of food and matching corresponding storage nodes according to the production information of the food; the food supply label generating unit is used for acquiring storage information of food at the storage node and generating a corresponding food supply label; the food supply label analyzing unit is used for analyzing the food supply label and removing abnormal food of the food supply label, however, the method does not solve the problem of how to reduce the deterioration ratio by controlling the storage environment.
Therefore, the invention provides an intelligent supply chain management system and method realized by artificial intelligence and the Internet of things technology.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides the intelligent supply chain management system and the intelligent supply chain management method realized by the artificial intelligence and the Internet of things technology, and the transportation quality of the commodity supply chain is improved by means of the pre-parameter regulation and the self-adaptive parameter regulation in the transportation process.
In order to achieve the above purpose, an intelligent supply chain management method implemented by artificial intelligence and internet of things technology is provided, which comprises the following steps:
step one: collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance;
Step two: taking variable characteristic sample data and controllable characteristic sample data as input and taking metamorphic proportion label sample data as output, and training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process;
Step three: for each batch of goods to be transported, collecting variable characteristic data, and generating recommended controllable characteristic data of the goods to be transported based on the variable characteristic data and a deterioration proportion prediction model;
Step four: transporting the commodity to be transported by using the recommended controllable characteristic data, and collecting regulation and control condition data in real time;
Step five: judging whether the recommended controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the recommended controllable feature data through the Internet of things by using an Actor-Critic network model; otherwise, not processing;
The mode of collecting the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data in advance is as follows:
N sections of test transportation routes are selected in advance, different controllable characteristic parameter values are adopted for transportation of commodities in each section of test transportation route, the controllable characteristic parameter values in each transportation process are collected to be used as a group of controllable characteristic sets, and all the controllable characteristic sets form controllable characteristic sample data; collecting variable characteristic parameter values in each transportation process and metamorphic proportion labels of commodities after each transportation is completed; the variable characteristic parameter values in each transportation process form a group of variable characteristic sets, all the variable characteristic sets form variable characteristic sample data, and all the metamorphic proportion labels form metamorphic proportion label sample data; wherein N is the number of selected test transportation routes;
the controllable characteristic is a characteristic capable of being remotely and automatically controlled and adjusted in the process of transporting the commodity;
The variable feature is a feature that the pointer changes for different haul roads;
The spoilage proportion label of the commodity is the ratio of the spoiled commodity quantity to the total quantity of the commodity transported after the transportation is completed;
The mode of training the deterioration ratio prediction model for predicting the deterioration ratio in the commodity transportation process is as follows:
forming a group of metamorphic proportion feature vectors by a controllable feature set and a variable feature set generated by each transport commodity of each test transport route;
Taking each group of metamorphic proportion feature vectors as input of a metamorphic proportion prediction model, taking a predicted value of metamorphic proportion of the transportation corresponding to the metamorphic proportion feature vectors as output, taking a metamorphic proportion label of the transportation corresponding to the metamorphic proportion feature vectors as a prediction target, taking a difference value between the predicted value of the metamorphic proportion and the metamorphic proportion label as a prediction error, and taking the sum of minimized prediction errors as a training target; training the deterioration ratio prediction model until the sum of prediction errors reaches convergence, and stopping training; the deterioration ratio prediction model is a polynomial regression model;
the method for collecting the variable characteristic data for each batch of commodities to be transported is as follows:
Collecting expected transportation routes of goods to be transported;
Collecting parameter values of various variable features each time the variable features are transported on the expected transportation route in the past as historical parameter values of the variable features;
for each variable feature, calculating an average of all corresponding historical parameter values as an expected parameter value for the variable feature;
all expected parameter values constitute variable characteristic data;
the method for generating the recommended controllable feature data of the commodity to be transported comprises the following steps:
Marking a functional expression corresponding to the trained deterioration ratio prediction model as H (B, K), wherein B represents a constant set of variable characteristics and K represents a variable set of controllable characteristics;
Marking the variable characteristic data as B0, solving the parameter values of each controllable characteristic corresponding to the function H (B0, K) reaching the minimum value by using a derivative method or a gradient descent algorithm, and forming the recommended controllable characteristic data by the solved parameter values of each controllable characteristic;
the mode for collecting the regulation and control condition data in real time is as follows:
In the process of transporting goods to be transported, collecting actual parameter values of various variable features in a driven road section in real time through a vibration sensor, and collecting real-time concentration of metamorphic gas in a storage space in real time through a metamorphic gas sensor;
The actual parameter values of various variable characteristics and the real-time concentration of the metamorphic gas form regulation and control condition data;
The method for judging whether the recommended controllable feature data needs to be adjusted is as follows:
Presetting an abnormal regulation threshold value for each variable feature, and judging that the recommended controllable feature data needs to be adjusted if any variable feature exists and the difference value between the actual parameter value and the expected parameter value is larger than the abnormal regulation threshold value or the real-time concentration of metamorphic gas is larger than the preset metamorphic gas concentration threshold value;
The method for adjusting the recommended controllable feature data through the Internet of things by using the Actor-Critic network model comprises the following steps:
Initializing network parameters of an Actor network and a Critic network;
When each time it is judged that adjustment is required, the following steps are executed:
Step 11: taking the parameter value of the current controllable feature as the current state;
step 12: the Actor network outputs the adjusted parameter values of each controllable feature;
Taking the parameter values of the adjusted controllable features as the next state;
step 13: calculating an actual rewarding value Q; the actual rewarding value Q is rewarding obtained after the parameter values of the controllable features are adjusted each time;
the actual reward value Q is calculated by the following steps:
Marking a constant set of variable features consisting of actual parameter values of each variable feature as B1;
marking the recommended controllable feature data before adjustment as K0, and marking the recommended controllable feature data after adjustment as K1;
; wherein w1 and w2 are respectively preset proportional coefficients, and C is the power difference value of the transport vehicle in unit time which is required to be consumed after and before adjustment;
step 14: updating the value of the prize value function using an update formula of the Critic network to adjust the estimate of the actual prize value Q for the decision result;
Step 15: the parameters of the Actor network are updated using the update formula of the Actor network to increase the probability of selecting a high rewards decision result in a given state.
The intelligent supply chain management system realized by the artificial intelligence and the Internet of things technology comprises a training data collection module, a model training module, a suggestion controllable feature data generation module and a suggestion controllable feature data adjustment module; the data interaction is carried out among the modules in a data exchange or interface calling mode;
The training data collection module is used for collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance and sending the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data to the model training module;
The model training module is used for taking the variable characteristic sample data and the controllable characteristic sample data as input, taking the metamorphic proportion label sample data as output, training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process, and sending the metamorphic proportion prediction model to the suggestion controllable characteristic data generation module and the suggestion controllable characteristic data adjustment module;
the recommendation controllable feature data generation module is used for collecting variable feature data for each batch of goods to be transported, generating recommendation controllable feature data of the goods to be transported based on the variable feature data and the deterioration proportion prediction model, and sending the recommendation controllable feature data to the recommendation controllable feature data adjustment module;
And the suggestion controllable feature data adjusting module is used for transporting the commodity to be transported by using the suggestion controllable feature data, collecting regulation and control condition data in real time, judging whether the suggestion controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the suggestion controllable feature data by using an Actor-Critic network model through the Internet of things.
An electronic device is proposed, comprising: a processor and a memory, wherein the memory stores a computer program for the processor to call;
And the processor executes the intelligent supply chain management method realized by the artificial intelligence+the Internet of things technology by calling the computer program stored in the memory.
A computer-readable storage medium is proposed, on which a computer program is stored that is erasable;
When the computer program runs on the computer equipment, the computer equipment is caused to execute the intelligent supply chain management method realized by the artificial intelligence and the Internet of things technology.
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the method, variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data are collected in advance, the variable characteristic sample data and the controllable characteristic sample data are taken as input, the metamorphic proportion label sample data are taken as output, a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process is trained, the variable characteristic data are collected for each batch of commodities to be transported, and recommended controllable characteristic data of the commodities to be transported are generated based on the variable characteristic data and the metamorphic proportion prediction model; the effect of improving the transportation quality of the commodity supply chain through the regulation and control of the pre-parameters aiming at the transportation roads with different variable characteristics is realized by training the deterioration proportion prediction model;
(2) The method further uses the recommended controllable feature data to transport the commodity to be transported, collects the regulation and control condition data in real time, judges whether the recommended controllable feature data needs to be adjusted based on the regulation and control condition data, and adjusts the recommended controllable feature data through the Internet of things by using an Actor-Critic network model if the recommended controllable feature data needs to be adjusted; through training on-line deep reinforcement learning, the problem that when road conditions or traffic conditions in the actual transportation process are not matched with expected parameter values, the parameter values of the variable features are adaptively adjusted according to the road conditions or traffic conditions is solved, and the effect of improving the transportation quality of the commodity supply chain more flexibly and intelligently is achieved.
Drawings
FIG. 1 is a flowchart of an intelligent supply chain management method implemented by the technology of artificial intelligence+Internet of things in embodiment 1 of the present invention;
FIG. 2 is a diagram of the connection relationship between modules of an intelligent supply chain management system implemented by the technology of artificial intelligence+Internet of things in embodiment 2 of the present invention;
fig. 3 is a schematic structural diagram of an electronic device in embodiment 3 of the present invention;
fig. 4 is a schematic diagram of a computer-readable storage medium according to embodiment 4 of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
As shown in fig. 1, the intelligent supply chain management method implemented by the artificial intelligence+internet of things technology comprises the following steps:
step one: collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance;
Step two: taking variable characteristic sample data and controllable characteristic sample data as input and taking metamorphic proportion label sample data as output, and training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process;
Step three: for each batch of goods to be transported, collecting variable characteristic data, and generating recommended controllable characteristic data of the goods to be transported based on the variable characteristic data and a deterioration proportion prediction model;
Step four: transporting the commodity to be transported by using the recommended controllable characteristic data, and collecting regulation and control condition data in real time;
Step five: judging whether the recommended controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the recommended controllable feature data through the Internet of things by using an Actor-Critic network model; otherwise, not processing;
Step six: repeatedly executing the fourth step to the fifth step until the goods to be transported reach the destination;
the method for collecting the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data in advance comprises the following steps:
N sections of test transportation routes are selected in advance, different controllable characteristic parameter values are adopted for transportation of commodities in each section of test transportation route, the controllable characteristic parameter values in each transportation process are collected to be used as a group of controllable characteristic sets, and all the controllable characteristic sets form controllable characteristic sample data; collecting variable characteristic parameter values in each transportation process and metamorphic proportion labels of commodities after each transportation is completed; the variable characteristic parameter values in each transportation process form a group of variable characteristic sets, all the variable characteristic sets form variable characteristic sample data, and all the metamorphic proportion labels form metamorphic proportion label sample data; wherein N is the number of selected test transportation routes;
the controllable characteristic is a characteristic that can be remotely and automatically controlled and regulated in the process of transporting the commodity, for example, the temperature, humidity and other environmental parameters stored in the transportation vehicle of the commodity, and the value of the controllable characteristic parameter is a specific value of the controllable characteristic during each transportation;
Further, the variable characteristic is a characteristic that a pointer changes for different transportation roads, and the variable characteristic relates to a transportation route, that is, the variable characteristic is an uncontrollable parameter, such as a road jolting degree average value, a road jolting degree variance value, a transportation time length and the like; it will be appreciated that the degree of jolting may be collected in real time by mounting an amplitude sensor on the transport vehicle;
The spoilage proportion label of the commodity is the ratio of the spoiled commodity quantity to the total quantity of the commodity transported after the transportation is completed; it can be understood that the distinction of deteriorated commodities can be performed by manual marking, and intelligent classification can also be performed by a computer vision technology;
Furthermore, the mode of training the spoilage ratio prediction model for predicting the spoilage ratio in the commodity transportation process by taking the variable characteristic sample data and the controllable characteristic sample data as input and the spoilage ratio label sample data as output is as follows:
forming a group of metamorphic proportion feature vectors by a controllable feature set and a variable feature set generated by each transport commodity of each test transport route;
Taking each group of metamorphic proportion feature vectors as input of a metamorphic proportion prediction model, taking a predicted value of metamorphic proportion of the transportation corresponding to the metamorphic proportion feature vectors as output, taking a metamorphic proportion label of the transportation corresponding to the metamorphic proportion feature vectors as a prediction target, taking a difference value between the predicted value of the metamorphic proportion and the metamorphic proportion label as a prediction error, and taking the sum of minimized prediction errors as a training target; training the deterioration ratio prediction model until the sum of prediction errors reaches convergence, and training the deterioration ratio prediction model according to the controllable feature set and the variable feature set, and outputting a predicted value of the commodity deterioration ratio when transportation is completed; the deterioration ratio prediction model is a polynomial regression model; the sum of the prediction errors is a mean square error;
Further, for each commodity to be transported, the method for collecting the variable characteristic data is as follows:
collecting expected transportation routes of goods to be transported; it can be understood that in the general commodity transportation process, under the condition that the upstream and downstream addresses are determined, the transportation route can be determined generally, and the expected transportation route can be obtained through the electronic map technology, which belongs to the conventional technical means in the field, and the invention is not repeated here;
Collecting parameter values of various variable features each time the variable features are transported on the expected transportation route in the past as historical parameter values of the variable features;
for each variable feature, calculating an average of all corresponding historical parameter values as an expected parameter value for the variable feature;
all expected parameter values constitute variable characteristic data;
It can be understood that the variable characteristic data expresses the jolt state and the expected transportation time period that the commodity is expected to bear when transported on the expected transportation route, and because different road jolt conditions and transportation time periods affect the deterioration speed of the commodity, the parameter values of the suitable controllable characteristics need to be designed for the specific combination of the variable characteristic data so as to reduce the deterioration proportion of the commodity;
Further, the method for generating the recommended controllable feature data of the commodity to be transported based on the variable feature data and the deterioration ratio prediction model is as follows:
marking a functional expression corresponding to the trained deterioration ratio prediction model as H (B, K), wherein B represents a constant set of variable characteristics and K represents a variable set of controllable characteristics; that is, in the functional expression H (B, K), B is a constant parameter therein, K is a variable parameter therein, a specific example is +b, where a and B are constant parameters, and x is a variable parameter;
marking the variable characteristic data as B0, solving the parameter values of each controllable characteristic corresponding to the function H (B0, K) reaching the minimum value by using a derivative method or a gradient descent algorithm, and forming the recommended controllable characteristic data by the solved parameter values of each controllable characteristic; it can be understood that by carrying out minimum solution on the function H (B0, K), the optimal variable parameter K is obtained, so that the deterioration ratio is minimized, and the effect of improving the transportation quality of the commodity supply chain through the regulation and control of the parameters in advance is realized;
It can be understood that in the actual transportation process of the commodity to be transported, the actual parameter values of various variable characteristics and the expected parameter values are different due to the change of road conditions and the possibility of traffic jam, so that dynamic adjustment is required according to the actual transportation condition;
Further, the manner of collecting the regulation and control condition data in real time is as follows:
In the process of transporting goods to be transported, collecting actual parameter values of various variable features in a driven road section in real time through a vibration sensor, and collecting real-time concentration of metamorphic gas in a storage space in real time through a metamorphic gas sensor; the deterioration gas is determined according to the specific gas type generated when the specific type of the commodity is deteriorated; it should be noted that, since the expected transportation time period is the expected transportation time period of the entire expected transportation route, the actual parameter value of the transportation time period is the sum of the currently transported time period plus the expected time period from the current location to the transportation destination;
The actual parameter values of various variable characteristics and the real-time concentration of the metamorphic gas form regulation and control condition data;
further, the method for judging whether the recommended controllable feature data needs to be adjusted based on the regulation and control condition data is as follows:
Presetting an abnormal regulation threshold value for each variable feature, and judging that the recommended controllable feature data needs to be adjusted if any variable feature exists and the difference value between the actual parameter value and the expected parameter value is larger than the abnormal regulation threshold value or the real-time concentration of metamorphic gas is larger than the preset metamorphic gas concentration threshold value;
If the difference value between the actual parameter value and the expected parameter value of any variable feature is smaller than or equal to an abnormal regulation threshold value or the real-time concentration of the metamorphic gas is smaller than or equal to a preset metamorphic gas concentration threshold value, judging that the recommended controllable feature data does not need to be regulated;
Further, the method for adjusting the recommended controllable feature data by using the Actor-Critic network model through the internet of things comprises the following steps:
Initializing network parameters of an Actor network and a Critic network; parameters include, but are not limited to, the dimensions of the state input layers of the Actor network, the number and size of hidden layers, the dimensions of the action output layers, the dimensions of the state input layers of the Critic network, the number and size of hidden layers and the dimensions of the prize value function output layers, the learning rate, discount factors, and network optimization algorithms (gradient descent method or Adam optimization algorithm, etc.);
When each time it is judged that adjustment is required, the following steps are executed:
Step 11: taking the parameter value of the current controllable feature as the current state;
step 12: the Actor network outputs the adjusted parameter values of each controllable feature;
Taking the parameter values of the adjusted controllable features as the next state;
step 13: calculating an actual rewarding value Q; the actual rewarding value Q is rewarding obtained after the parameter values of the controllable features are adjusted each time;
the actual reward value Q is calculated by the following steps:
Marking a constant set of variable features consisting of actual parameter values of each variable feature as B1;
marking the recommended controllable feature data before adjustment as K0, and marking the recommended controllable feature data after adjustment as K1;
; wherein w1 and w2 are respectively preset proportional coefficients, and C is the power difference value of the transport vehicle in unit time which is required to be consumed after and before adjustment; it will be appreciated that C may be obtained statistically by using a power meter;
Note that in the formula measures the degree of decrease in the deterioration ratio of the commodity before and after adjustment, and/> measures the degree of increase in the power consumption before and after adjustment; the rewarding value Q ensures the balance between optimizing the commodity deterioration proportion and optimizing the power consumption cost;
Step 14: updating the value of the prize value function using an update formula of the Critic network to adjust the estimate of the actual prize value Q for the decision result; it should be noted that the update formula may be a conventional update formula for those skilled in the art, for example: Wherein/> is the current state/> prize value function estimate; the/> is the learning rate, controlling the updated step size; the/> is a discount factor used to measure the importance of future rewards; the next state is/> ;
Step 15: the parameters of the Actor network are updated using the update formula of the Actor network to increase the probability of selecting a high rewards decision result in a given state.
Example 2
As shown in fig. 2, the intelligent supply chain management system implemented by the artificial intelligence+internet of things technology comprises a training data collection module, a model training module, a suggestion controllable feature data generation module and a suggestion controllable feature data adjustment module; the data interaction is carried out among the modules in a data exchange or interface calling mode;
The training data collection module is used for collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance and sending the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data to the model training module;
The model training module is mainly used for taking variable characteristic sample data and controllable characteristic sample data as input, taking metamorphic proportion label sample data as output, training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process, and sending the metamorphic proportion prediction model to the suggestion controllable characteristic data generation module and the suggestion controllable characteristic data adjustment module;
the recommendation controllable feature data generation module is mainly used for collecting variable feature data for each batch of goods to be transported, generating recommendation controllable feature data of the goods to be transported based on the variable feature data and the deterioration proportion prediction model, and sending the recommendation controllable feature data to the recommendation controllable feature data adjustment module;
The suggestion controllable feature data adjusting module is mainly used for transporting commodities to be transported by using the suggestion controllable feature data, collecting adjusting and controlling condition data in real time, judging whether the suggestion controllable feature data need to be adjusted based on the adjusting and controlling condition data, and if so, adjusting the suggestion controllable feature data through the Internet of things by using an Actor-Critic network model.
Example 3
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 3, there is also provided an electronic device 100 according to yet another aspect of the present application. The electronic device 100 may include one or more processors and one or more memories. Wherein the memory has stored therein computer readable code which, when executed by the one or more processors, may perform the intelligent supply chain management method implemented by the artificial intelligence + internet of things technology as described above.
The method or apparatus according to embodiments of the present application may also be implemented by means of the architecture of the electronic device shown in fig. 3. As shown in fig. 3, the electronic device 100 may include a bus 101, one or more CPUs 102, a ROM103, a RAM104, a communication port 105 connected to a network, an input/output component 106, a hard disk 107, and the like. A storage device in the electronic device 100, such as the ROM103 or the hard disk 107, may store the intelligent supply chain management method implemented by the artificial intelligence+internet of things technology provided by the present application. The intelligent supply chain management method implemented by the artificial intelligence+the internet of things technology can comprise the following steps: step one: collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance; step two: taking variable characteristic sample data and controllable characteristic sample data as input and taking metamorphic proportion label sample data as output, and training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process; step three: for each batch of goods to be transported, collecting variable characteristic data, and generating recommended controllable characteristic data of the goods to be transported based on the variable characteristic data and a deterioration proportion prediction model; step four: transporting the commodity to be transported by using the recommended controllable characteristic data, and collecting regulation and control condition data in real time; step five: judging whether the recommended controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the recommended controllable feature data through the Internet of things by using an Actor-Critic network model; otherwise, not processing; step six: repeatedly executing the fourth step to the fifth step until the goods to be transported reach the destination;
further, the electronic device 100 may also include a user interface 108. Of course, the architecture shown in fig. 3 is merely exemplary, and one or more components of the electronic device shown in fig. 3 may be omitted as may be practical in implementing different devices.
Example 4
FIG. 4 is a schematic diagram of a computer-readable storage medium according to one embodiment of the present application. As shown in fig. 4, is a computer-readable storage medium 200 according to one embodiment of the application. The computer-readable storage medium 200 has stored thereon computer-readable instructions. When the computer readable instructions are executed by the processor, the intelligent supply chain management method implemented by the artificial intelligence+internet of things technology according to the embodiment of the present application described with reference to the above drawings may be performed. Computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and/or nonvolatile memory. Volatile memory can include, for example, random Access Memory (RAM), cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like.
In addition, according to embodiments of the present application, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions executable by a processor to perform instructions corresponding to the method steps provided by the present application, which when executed by a Central Processing Unit (CPU), perform the functions defined above in the method of the present application.
The methods and apparatus, devices of the present application may be implemented in numerous ways. For example, the methods and apparatus, devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, firmware. The above-described sequence of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the sequence specifically described above unless specifically stated otherwise. Furthermore, in some embodiments, the present application may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present application. Thus, the present application also covers a recording medium storing a program for executing the method according to the present application.
In addition, in the foregoing technical solutions provided in the embodiments of the present application, parts consistent with implementation principles of corresponding technical solutions in the prior art are not described in detail, so that redundant descriptions are avoided.
The purpose, technical scheme and beneficial effects of the invention are further described in detail in the detailed description. It is to be understood that the above description is only of specific embodiments of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The above preset parameters or preset thresholds are set by those skilled in the art according to actual conditions or are obtained by mass data simulation.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (8)
1. The intelligent supply chain management method realized by the artificial intelligence and the Internet of things technology is characterized by comprising the following steps of:
step one: collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance;
Step two: taking variable characteristic sample data and controllable characteristic sample data as input and taking metamorphic proportion label sample data as output, and training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process;
Step three: for each batch of goods to be transported, collecting variable characteristic data, and generating recommended controllable characteristic data of the goods to be transported based on the variable characteristic data and a deterioration proportion prediction model;
Step four: transporting the commodity to be transported by using the recommended controllable characteristic data, and collecting regulation and control condition data in real time;
Step five: judging whether the recommended controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the recommended controllable feature data through the Internet of things by using an Actor-Critic network model; otherwise, not processing;
The mode of collecting the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data in advance is as follows:
N sections of test transportation routes are selected in advance, different controllable characteristic parameter values are adopted for transportation of commodities in each section of test transportation route, the controllable characteristic parameter values in each transportation process are collected to be used as a group of controllable characteristic sets, and all the controllable characteristic sets form controllable characteristic sample data; collecting variable characteristic parameter values in each transportation process and metamorphic proportion labels of commodities after each transportation is completed; the variable characteristic parameter values in each transportation process form a group of variable characteristic sets, all the variable characteristic sets form variable characteristic sample data, and all the metamorphic proportion labels form metamorphic proportion label sample data; wherein N is the number of selected test transportation routes;
the controllable characteristic is a characteristic capable of being remotely and automatically controlled and adjusted in the process of transporting the commodity;
the variable feature is a feature that the pointer changes for different haul roads;
The spoilage proportion label of the commodity is the ratio of the spoiled commodity quantity to the total quantity of the commodity transported after the transportation is completed;
the method for generating the recommended controllable feature data of the commodity to be transported comprises the following steps:
Marking a functional expression corresponding to the trained deterioration ratio prediction model as H (B, K), wherein B represents a constant set of variable characteristics and K represents a variable set of controllable characteristics;
Marking the variable characteristic data as B0, solving the parameter values of each controllable characteristic corresponding to the function H (B0, K) reaching the minimum value by using a derivative method or a gradient descent algorithm, and forming the recommended controllable characteristic data by the solved parameter values of each controllable characteristic;
The method for adjusting the recommended controllable feature data through the Internet of things by using the Actor-Critic network model comprises the following steps:
Initializing network parameters of an Actor network and a Critic network;
When each time it is judged that adjustment is required, the following steps are executed:
Step 11: taking the parameter value of the current controllable feature as the current state;
step 12: the Actor network outputs the adjusted parameter values of each controllable feature;
Taking the parameter values of the adjusted controllable features as the next state;
step 13: calculating an actual rewarding value Q; the actual rewarding value Q is rewarding obtained after the parameter values of the controllable features are adjusted each time;
step 14: updating the value of the prize value function using an update formula of the Critic network to adjust the estimate of the actual prize value Q for the decision result;
Step 15: updating parameters of the Actor network by using an updating formula of the Actor network so as to improve the probability of selecting a high rewarding decision result in a given state;
the actual reward value Q is calculated by the following steps:
Marking a constant set of variable features consisting of actual parameter values of each variable feature as B1;
marking the recommended controllable feature data before adjustment as K0, and marking the recommended controllable feature data after adjustment as K1;
; wherein w1 and w2 are preset proportional coefficients respectively, and C is the power difference value of the unit time required to be consumed by the transport vehicle after and before adjustment.
2. The intelligent supply chain management method implemented by the technology of artificial intelligence and internet of things according to claim 1, wherein the training of the spoilage ratio prediction model for predicting the spoilage ratio in the commodity transportation process is as follows:
forming a group of metamorphic proportion feature vectors by a controllable feature set and a variable feature set generated by each transport commodity of each test transport route;
Taking each group of metamorphic proportion feature vectors as input of a metamorphic proportion prediction model, taking a predicted value of metamorphic proportion of the transportation corresponding to the metamorphic proportion feature vectors as output, taking a metamorphic proportion label of the transportation corresponding to the metamorphic proportion feature vectors as a prediction target, taking a difference value between the predicted value of the metamorphic proportion and the metamorphic proportion label as a prediction error, and taking the sum of minimized prediction errors as a training target; training the deterioration ratio prediction model until the sum of prediction errors reaches convergence, and stopping training; the deterioration ratio prediction model is a polynomial regression model.
3. The method for intelligent supply chain management implemented by artificial intelligence+internet of things according to claim 2, wherein the manner of collecting variable feature data for each batch of goods to be transported is:
Collecting expected transportation routes of goods to be transported;
Collecting parameter values of various variable features each time the variable features are transported on the expected transportation route in the past as historical parameter values of the variable features;
for each variable feature, calculating an average of all corresponding historical parameter values as an expected parameter value for the variable feature;
all expected parameter values constitute variable characteristic data.
4. The intelligent supply chain management method implemented by the technology of artificial intelligence+internet of things according to claim 3, wherein the manner of collecting the regulation and control condition data in real time is as follows:
In the process of transporting goods to be transported, collecting actual parameter values of various variable features in a driven road section in real time through a vibration sensor, and collecting real-time concentration of metamorphic gas in a storage space in real time through a metamorphic gas sensor;
The actual parameter values of the various variable features and the real-time concentration of the metamorphic gas constitute regulation condition data.
5. The method for intelligent supply chain management implemented by artificial intelligence+internet of things technology according to claim 4, wherein the determining whether the recommended controllable feature data needs to be adjusted is:
An abnormal regulation threshold value is preset for each variable feature, and if any variable feature exists, if the difference value between the actual parameter value and the expected parameter value is larger than the abnormal regulation threshold value or the real-time concentration of metamorphic gas is larger than the preset metamorphic gas concentration threshold value, the recommended controllable feature data is judged to be required to be regulated.
6. An intelligent supply chain management system realized by an artificial intelligence+internet of things technology for realizing the intelligent supply chain management method realized by the artificial intelligence+internet of things technology according to any one of claims 1 to 5, which is characterized by comprising a training data collection module, a model training module, a suggestion controllable feature data generation module and a suggestion controllable feature data adjustment module; the data interaction is carried out among the modules in a data exchange or interface calling mode;
The training data collection module is used for collecting variable characteristic sample data, controllable characteristic sample data and metamorphic proportion label sample data in advance and sending the variable characteristic sample data, the controllable characteristic sample data and the metamorphic proportion label sample data to the model training module;
The model training module is used for taking the variable characteristic sample data and the controllable characteristic sample data as input, taking the metamorphic proportion label sample data as output, training a metamorphic proportion prediction model for predicting metamorphic proportion in the commodity transportation process, and sending the metamorphic proportion prediction model to the suggestion controllable characteristic data generation module and the suggestion controllable characteristic data adjustment module;
the recommendation controllable feature data generation module is used for collecting variable feature data for each batch of goods to be transported, generating recommendation controllable feature data of the goods to be transported based on the variable feature data and the deterioration proportion prediction model, and sending the recommendation controllable feature data to the recommendation controllable feature data adjustment module;
And the suggestion controllable feature data adjusting module is used for transporting the commodity to be transported by using the suggestion controllable feature data, collecting regulation and control condition data in real time, judging whether the suggestion controllable feature data need to be adjusted based on the regulation and control condition data, and if so, adjusting the suggestion controllable feature data by using an Actor-Critic network model through the Internet of things.
7. An electronic device, comprising: a processor and a memory, wherein:
The memory stores a computer program which can be called by the processor;
The processor executes the artificial intelligence + internet of things-implemented intelligent supply chain management method of any one of claims 1-5 in the background by invoking a computer program stored in the memory.
8. A computer readable storage medium having stored thereon a computer program that is erasable;
when the computer program is run on a computer device, the computer device is caused to perform the artificial intelligence+internet of things-implemented intelligent supply chain management method of any of claims 1-5 in the background.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410161877.2A CN117709830B (en) | 2024-02-05 | 2024-02-05 | Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410161877.2A CN117709830B (en) | 2024-02-05 | 2024-02-05 | Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117709830A CN117709830A (en) | 2024-03-15 |
| CN117709830B true CN117709830B (en) | 2024-04-16 |
Family
ID=90161083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410161877.2A Active CN117709830B (en) | 2024-02-05 | 2024-02-05 | Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117709830B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113099729A (en) * | 2018-10-26 | 2021-07-09 | 陶氏环球技术有限责任公司 | Deep reinforcement learning for production scheduling |
| CN113125657A (en) * | 2021-04-02 | 2021-07-16 | 中国肉类食品综合研究中心 | Refrigerated meat product spoilage monitoring and early warning system and application thereof |
| WO2022138829A1 (en) * | 2020-12-25 | 2022-06-30 | 三井化学株式会社 | Freshness estimation device, freshness estimation method, freshness estimation program, and data structure |
| CN116050971A (en) * | 2023-02-01 | 2023-05-02 | 安徽农业大学 | Multi-temperature co-distribution platform based on Internet of things technology |
| CN117236825A (en) * | 2023-11-15 | 2023-12-15 | 南京龟兔赛跑软件研究院有限公司 | Agricultural product supply chain transportation management system and method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160196527A1 (en) * | 2015-01-06 | 2016-07-07 | Falkonry, Inc. | Condition monitoring and prediction for smart logistics |
| US20160203422A1 (en) * | 2015-01-14 | 2016-07-14 | Nextop Italia Srl Semplificata | Method and electronic travel route building system, based on an intermodal electronic platform |
| US20180053140A1 (en) * | 2016-08-18 | 2018-02-22 | Intel Corporation | Food storage method and apparatus with sensors |
| US20190387375A1 (en) * | 2018-06-14 | 2019-12-19 | Candibell, Inc. | System and method for food quality monitoring and intelligent restocking |
-
2024
- 2024-02-05 CN CN202410161877.2A patent/CN117709830B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113099729A (en) * | 2018-10-26 | 2021-07-09 | 陶氏环球技术有限责任公司 | Deep reinforcement learning for production scheduling |
| WO2022138829A1 (en) * | 2020-12-25 | 2022-06-30 | 三井化学株式会社 | Freshness estimation device, freshness estimation method, freshness estimation program, and data structure |
| CN113125657A (en) * | 2021-04-02 | 2021-07-16 | 中国肉类食品综合研究中心 | Refrigerated meat product spoilage monitoring and early warning system and application thereof |
| CN116050971A (en) * | 2023-02-01 | 2023-05-02 | 安徽农业大学 | Multi-temperature co-distribution platform based on Internet of things technology |
| CN117236825A (en) * | 2023-11-15 | 2023-12-15 | 南京龟兔赛跑软件研究院有限公司 | Agricultural product supply chain transportation management system and method |
Non-Patent Citations (3)
| Title |
|---|
| Reinforcement Learning to Optimize Lifetime Value in Cold-Start Recommendation;Luo Ji 等;《Proceedings of the 30th ACM International Conference on Information & Knowledge Management》;20211030;第782-791页 * |
| 基于供应链管理的冷链物流发展模式研究;陈彬斌;《中国集体经济》;20220125(第3期);第121-122页 * |
| 基于强化学习的生鲜库存策略优化研究;翁丽珊;《万方》;20221031;第1-72页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117709830A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Luo et al. | Dynamic multi-objective scheduling for flexible job shop by deep reinforcement learning | |
| Luo | Dynamic scheduling for flexible job shop with new job insertions by deep reinforcement learning | |
| He et al. | Prediction modelling of cold chain logistics demand based on data mining algorithm | |
| CN111738532B (en) | Method and system for acquiring influence degree of event on object | |
| CN115146764A (en) | Training method and device of prediction model, electronic equipment and storage medium | |
| CN115983485A (en) | Prediction method and system for carbon emission of power plant | |
| CN114356641B (en) | Incremental software defect prediction method, system, equipment and storage medium | |
| Berling et al. | Dynamic speed optimization in supply chains with stochastic demand | |
| CN117709830B (en) | Intelligent supply chain management system and method realized by artificial intelligence and Internet of things technology | |
| CN115454162A (en) | Intelligent temperature control method, equipment and medium for grain storage | |
| CN117495019B (en) | Agricultural product cooperative scheduling method and system based on agricultural product supply chain | |
| Quevedo et al. | Using reinforcement learning for tuning genetic algorithms | |
| CN112288484A (en) | Commodity automatic pricing method and device, electronic equipment and storage medium | |
| CN116880612A (en) | Temperature control method and system for cold chain transportation and electronic equipment | |
| CN116522065A (en) | Coal mill health degree assessment method based on deep learning | |
| US20210027109A1 (en) | Evaluation system, evaluation method, and program for evaluation | |
| CN111507499A (en) | Construction method, test method, device and system of model for prediction | |
| CN108985526B (en) | Transportation capacity prediction method and device, computer readable storage medium and terminal | |
| CN112686416A (en) | Wind power probability prediction method and system | |
| CN118095605B (en) | Logistics route planning method | |
| Wang et al. | Robust dynamic bus control: A distributional multi-agent reinforcement learning approach | |
| CN115456168B (en) | Training method of reinforcement learning model, energy consumption determining method and device | |
| CN117973635B (en) | Decision prediction method, electronic device, and computer-readable storage medium | |
| CN118092195B (en) | Multi-agent cooperative control method for improving IQL based on cooperative training model | |
| Tian et al. | The Research of Building Logistics Cost Forecast Based on Regression Support Vector Machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |