Forecasting Model for the Number of Breeding Sows Based on Pig’s Months of Age Transfer and Improved Flower Pollination Algorithm-Back Propagation Neural Network

Regulating the number of breeding sows (NBS) is crucial for pork supply–demand balance. Current forecasting methods for NBS fail to consider the principle of pig’s months of age (MOA) transfer and the impact of factors like diseases and policies on NBS fluctuations, leading to unsatisfactory accurac...

Full description

Saved in:
Bibliographic Details
Published in:Applied intelligence (Dordrecht, Netherlands) Vol. 54; no. 7; pp. 5826 - 5858
Main Authors: Song, Haohao, Zhang, Hongyu, Yang, Jingnan, Wang, Jiquan
Format: Journal Article
Language:English
Published: New York Springer US 01.04.2024
Springer Nature B.V
Subjects:
Accuracy
Algorithms
Artificial Intelligence
Artificial neural networks
Back propagation
Back propagation networks
Computer Science
Forecasting
Hogs
Machines
Manufacturing
Mechanical Engineering
Neural networks
Principles
Processes
Swine
IFPA-BPNN
Pork price
Forecasting
The principle of pig’s MOA transfer
The number of breeding sows
ISSN:0924-669X, 1573-7497
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Regulating the number of breeding sows (NBS) is crucial for pork supply–demand balance. Current forecasting methods for NBS fail to consider the principle of pig’s months of age (MOA) transfer and the impact of factors like diseases and policies on NBS fluctuations, leading to unsatisfactory accuracy. To bridge the research gap, a two-part forecasting model for the NBS was developed. In the first part, a recurrence forecasting model was established according to the growth characteristics of pigs and the principle of pig’s MOA transfer. In the second part, the random disturbance term was introduced to consider the influence of plague, policy and other factors on the NBS, and a forecasting method for random disturbance term based on Improved Flower Pollination Algorithm-Back Propagation Neural Network (IFPA-BPNN) was given. Subsequently, the proposed IFA and other newer optimization algorithms were evaluated on CEC 2017 test suite to verify the effectiveness and superiority of IFA. Lastly, the proposed model was employed to forecast the NBS in Heilongjiang Province and Anhui Province of China from 2009 to 2021. Compared to other time series forecasting models, the proposed model showed superior accuracy, confirming its scientific and effective nature. Relevant managerial insights were provided at the end of this paper.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-024-05413-1