Unlock instant, AI-driven research and patent intelligence for your innovation.

LSTM model-based express quantity prediction method, device and equipment based on , and storage medium

A prediction method and model technology, applied in the field of logistics management, can solve problems such as RNN cyclic neural network gradient disappearance, inaccurate express delivery volume prediction results, and inaccurate prediction data, so as to improve prediction results, simplify calculations, and reduce data The effect of large fluctuations

Pending Publication Date: 2021-03-02
SHANGHAI DONGPU INFORMATION TECH CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, in practical applications, if the input sensitivity of output Y3 and time 0 is high, when the traditional RNN network structure is used for training, the sensitivity between the input of Y3 and time 0 is weakened, and the problem of gradient disappearance occurs, resulting in poor network performance.
The essential reason for the gradient disappearance problem is that when the previous moment propagates to the next moment, it is affected by the input of the previous moment, so that the relevant information of the previous moment is lost.
[0005] Due to the problem of gradient disappearance in the RNN cyclic neural network, it cannot accurately predict data for a long period of time, making the prediction of the express delivery volume inaccurate

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • LSTM model-based express quantity prediction method, device and equipment based on , and storage medium
  • LSTM model-based express quantity prediction method, device and equipment based on , and storage medium
  • LSTM model-based express quantity prediction method, device and equipment based on , and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Please refer to figure 2 , the method for predicting the volume of express delivery based on the LSTM model in this embodiment includes:

[0050] Step S1: Obtain the historical data of the package quantity, preprocess the historical data, and select the target data set of at least one historical period.

[0051]In this embodiment, the piece quantity historical data refers to the piece quantity data stored in the logistics industry, and may also be the piece quantity data in the logistics industry within a certain period of time published by a statistical agency. The package quantity includes the package quantity, and of course it can be the package quantity. This embodiment takes the package quantity as an example. In the database, whether it is online or offline, information on the number of packages will be stored. Such information may include, but is not limited to: the type, time, and quantity of the quantity. Time can be stored by day, or by week, or by the spe...

Embodiment 2

[0120] This embodiment provides an LSTM model-based express shipment forecasting device, please refer to Figure 5 , the courier volume forecasting device includes:

[0121] The data processing module 1 is used to obtain the historical data of the amount of express parcels, preprocess the historical data, and select the target data set of at least one historical period;

[0122] Model creation module 2 is used to create an LSTM-based piece quantity forecasting model, input the target data set into the piece quantity forecasting model, and train the piece quantity forecasting model;

[0123] The piece quantity prediction module 3 is used to predict the package quantity of the express delivery in the next cycle based on the trained piece quantity prediction model, and output the predicted value.

[0124] The specific content and implementation methods of the above-mentioned data processing module 1, model creation module 2, and piece quantity prediction module 3 are as describe...

Embodiment 3

[0126] The second embodiment above describes in detail the LSTM model-based express delivery volume prediction device of the present invention from the perspective of modular functional entities. The following describes the express delivery volume prediction device based on the LSTM model of the present invention in detail from the perspective of hardware processing.

[0127] Please see Image 6, the LSTM model-based courier volume forecasting device 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 510 (for example, one or more processors ) and memory 520, one or more storage media 530 (such as one or more mass storage devices) for storing application programs 533 or data 532. Wherein, the memory 520 and the storage medium 530 may be temporary storage or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown in t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an LSTM (Long Short Term Memory) model-based express quantity prediction method, device and equipment and a storage medium, and aims to solve the problem that the express quantity prediction is inaccurate due to the fact that the gradient of an RNN (Recurrent Neural Network) disappears and long-time data cannot be accurately predicted.The method includes: acquiring historical data of the express collection amount, and preprocessing the historical data; selecting a target data set of at least one historical period; creating an LSTM-based part quantity prediction model, inputting the target data set into the part quantity prediction model, and training the part quantity prediction model; and based on the trained parcel quantity prediction model, predicting the expressparcel collection quantity of the next period, and outputting a prediction value. The problem of gradient disappearance of the RNN recurrent neural network is solved, so that the accuracy of expressquantity prediction is improved.

Description

technical field [0001] The invention belongs to the technical field of logistics management, and in particular relates to an LSTM model-based method, device, equipment and storage medium for predicting the volume of express delivery. Background technique [0002] With the development of e-commerce and the logistics industry, the volume of packages has increased rapidly, especially during statutory holidays, shopping festivals and other holidays, the express delivery business volume has increased significantly, and "explosions" have occurred from time to time. Accurate piece quantity forecasting will help logistics companies allocate resources rationally and make preparations in advance. [0003] With the development of science and technology, the prediction of express delivery volume has evolved from manual experience to machine deep learning. Machine deep learning usually uses RNN (Recurrent Neural Networks) cyclic neural network to predict the volume of express delivery. ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06Q10/08G06N3/04G06N3/08
CPCG06Q10/04G06Q10/0838G06N3/084G06N3/048G06N3/044G06N3/045
Inventor 夏扬陈玉芬李斯李培吉
Owner SHANGHAI DONGPU INFORMATION TECH CO LTD