Adaptive combination forecasting model for China’s logistics freight volume based on an improved PSO-BP neural network

Kybernetes ◽  
2015 ◽  
Vol 44 (4) ◽  
pp. 646-666 ◽  
Author(s):  
Zhou Cheng ◽  
Tao Juncheng
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Forecasting Model ◽  
Content Type ◽  
Forecasting Models ◽  
Combination Forecasting ◽  
Freight Volume ◽  
Convergence Performance ◽  
Combination Forecasting Model ◽  
Improved Pso

Purpose – To accurately forecast logistics freight volume plays a vital part in rational planning formulation for a country. The purpose of this paper is to contribute to developing a novel combination forecasting model to predict China’s logistics freight volume, in which an improved PSO-BP neural network is proposed to determine the combination weights. Design/methodology/approach – Since BP neural network has the ability of learning, storing, and recalling information that given by individual forecasting models, it is effective in determining the combination weights of combination forecasting model. First, an improved PSO based on simulated annealing method and space-time adjustment strategy (SAPSO) is proposed to solve out the connection weights of BP neural network, which overcomes the problems of local optimum traps, low precision and poor convergence during BP neural network training process. Then, a novel combination forecast model based on SAPSO-BP neural network is established. Findings – Simulation tests prove that the proposed SAPSO has better convergence performance and more stability. At the same time, combination forecasting models based on three types of BP neural networks are developed, which rank as SAPSO-BP, PSO-BP and BP in accordance with mean absolute percentage error (MAPE) and convergent speed. Also the proposed combination model based on SAPSO-BP shows its superiority, compared with some other combination weight assignment methods. Originality/value – SAPSO-BP neural network is an original contribution to the combination weight assignment methods of combination forecasting model, which has better convergence performance and more stability.

A Short-Term Combination Forecasting Model for Traffic Flow Based on the BP Neural Network

2012 ◽  
Vol 253-255 ◽  
pp. 1339-1344
Author(s):  
Tie Xin Cheng ◽  
Wen Bin Du ◽  
Jing Zhu Chen
Keyword(s):  
Neural Network ◽  
Traffic Flow ◽  
Bp Neural Network ◽  
Continuous Time ◽  
Nonlinear Problems ◽  
Forecasting Model ◽  
Short Term ◽  
Combination Forecasting ◽  
Combination Forecasting Model ◽  
Time Point

The forecasting for short-term traffic flow has always been one important and difficult research focus in the traffic forecasting areas. Based on the BP Neural Network, which was applied to nonlinear problems, the independent short-term forecasting models for the different traffic flow of the continuous time point series in one day and the constant date series at same time point were set up respectively, then, a short-term combination forecasting model for traffic flow, in which the regular fluctuations in the traffic flow data of the continuous time point series in one day and the constant date series at same time point were fully considered, was established, and can be applied to the complex spatio-temporal features of short-term traffic flow. With the sample of traffic flow dada, the forecasting results of the different models showed that the combination forecasting model provided a better forecast accuracy than the independent models.

scholarly journals Combination Forecasting Model for Predicting the Shelf Life of Two-state Materials Based on Support Vector Machine

2015 ◽  
Vol 9 (1) ◽  
pp. 124-129
Author(s):  
Li Zhiwei ◽  
Gao Qi ◽  
Liu Shenyang ◽  
Li Zhen
Keyword(s):  
Support Vector Machine ◽  
Prediction Method ◽  
Weighted Average ◽  
Support Vector ◽  
Forecasting Model ◽  
Forecasting Models ◽  
Optimal Weights ◽  
Combination Forecasting ◽  
Neural Network Prediction ◽  
Combination Forecasting Model

A combination forecasting model based on Support Vector Machine (SVM) whose objective is to minimize the structure risk, is proposed. The storage failure of two-state materials tends to fail immediately without any recognizable defeats prior to the failure, which increases the difficulty of forecasting, so the combination forecasting model is often used to optimize the prediction effect. The core ideas of previous combination forecasting models such as those based on forecasting error and those based on nonlinear weighted average are finding the optimal weights, but the structure of forecasting model is fixed. In this paper, three single forecasting models, Weibull distribution statistic method, BP neural network prediction method and SPFM (Sliding Polynomial Fitting Method) are chosen in which their forecast mechanisms are completely different. The results of single forecasting methods are used as training set of SVM. By using libsvm toolbox, we can get the nonlinear mapping functions that have the minimum structure risk. At last, a simulation is conducted to verify this model by using the data from Petroleum Center.

The analysis of the energy consumption of Chinese manufacturing based on the combination forecasting model

2015 ◽  
Vol 5 (1) ◽  
pp. 41-53 ◽  
Author(s):  
Tianxiang Yao ◽  
Wenrong Cheng
Keyword(s):  
Energy Consumption ◽  
Manufacturing Industry ◽  
Prediction Models ◽  
Grey System Theory ◽  
Forecasting Model ◽  
Single Model ◽  
Content Type ◽  
Advantages And Disadvantages ◽  
Combination Forecasting ◽  
Combination Forecasting Model

Purpose – The purpose of this paper is to find a method that has high precision to forecast the energy consumption of China’s manufacturing industry. The authors hope the predicted data can provide references to the formulation of government’s energy strategy and the sustained growth of economy in China. Design/methodology/approach – First, the authors respectively make use of regression prediction model and grey system theory GM(1,1) model to construct single model based the data of 2001-2010, analyze the advantages and disadvantages of single prediction models. The authors use the data of 2011 and 2012 to test the model. Second, the authors propose combination forecasting model of manufacturing’s energy consumption in China by using standard variance to allocate the weight. Finally, this model is applied to forecast China’s manufacturing energy consumption during 2013-2016. Findings – The result shows that the combination model is a better one with higher accuracy; the authors can take the model as an effective tool to predict manufacturing’s energy consumption in China. And the energy consumption of China’s manufacturing industry continued to show a steady incremental trend. Originality/value – This method takes full advantages of the effective information reflected by the single model and improves the prediction accuracy.

A preliminary study for selecting the appropriate AI-based forecasting model for hospital assets demand under disasters

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Sara Jebbor ◽  
Chiheb Raddouane ◽  
Abdellatif El Afia
Keyword(s):  
Neural Network ◽  
Mass Casualty ◽  
University Hospital ◽  
Forecasting Model ◽  
Real Time Control ◽  
Framework Analysis ◽  
Inference System ◽  
Everyday Activity ◽  
Content Type ◽  
Forecasting Models

PurposeHospitals recently search for more accurate forecasting systems, given the unpredictable demand and the increasing occurrence of disruptive incidents (mass casualty incidents, pandemics and natural disasters). Besides, the incorporation of automatic inventory and replenishment systems – that hospitals are undertaking – requires developed and accurate forecasting systems. Researchers propose different artificial intelligence (AI)-based forecasting models to predict hospital assets consumption (AC) for everyday activity case and prove that AI-based models generally outperform many forecasting models in this framework. The purpose of this paper is to identify the appropriate AI-based forecasting model(s) for predicting hospital AC under disruptive incidents to improve hospitals' response to disasters/pandemics situations.Design/methodology/approachThe authors select the appropriate AI-based forecasting models according to the deduced criteria from hospitals' framework analysis under disruptive incidents. Artificial neural network (ANN), recurrent neural network (RNN), adaptive neuro-fuzzy inference system (ANFIS) and learning-FIS (FIS with learning algorithms) are generally compliant with the criteria among many AI-based forecasting methods. Therefore, the authors evaluate their accuracy to predict a university hospital AC under a burn mass casualty incident.FindingsThe ANFIS model is the most compliant with the extracted criteria (autonomous learning capability, fast response, real-time control and interpretability) and provides the best accuracy (the average accuracy is 98.46%) comparing to the other models.Originality/valueThis work contributes to developing accurate forecasting systems for hospitals under disruptive incidents to improve their response to disasters/pandemics situations.

Application of a Traffic Flow Prediction Model Based on Neural Network in Intelligent Vehicle Management

2019 ◽  
Vol 33 (03) ◽  
pp. 1959009 ◽  
Author(s):  
Yang Guo ◽  
Lu Lu
Keyword(s):  
Neural Network ◽  
Traffic Flow ◽  
Forecasting Model ◽  
Intelligent Vehicle ◽  
Traffic Flow Prediction ◽  
Model Based ◽  
Flow Prediction ◽  
Combination Forecasting ◽  
Combination Forecasting Model ◽  
Vehicle Management

The ultimate direction of intelligent vehicle management is to achieve artificial intelligence (AI), and data mining is an important supporting technology for AI. The adoption of new AI technology can effectively improve operational efficiency and safety, especially in terms of performance. This paper takes the researches on traffic jam as an example and proposes one algorithm for combination forecasting model based on a segmentation algorithm for traffic flow sequence and BP neural network prediction. In this paper, it also introduces the traffic flow clustering analysis and mining algorithms for congestion events at the intersections. The blocking point algorithm is improved, and experimental analysis is performed through samples. Experimental results show that the algorithm use for combination forecasting model can greatly improve the real-time performance of short-term traffic flow prediction and significantly reduce the prediction error rate. Therefore, this algorithm has practical and innovative significance in the field of intelligent vehicle management.

Sign in / Sign up

Export Citation Format

Share Document