Proceedings of the 1st International Multi-Disciplinary Conference Theme: Sustainable Development and Smart Planning, IMDC-SDSP 2020, Cyperspace, 28-30 June 2020

Research Article

PSO Swarm intelligence technique to optimized ANN for demand forcasting

Download494 downloads
  • @INPROCEEDINGS{10.4108/eai.28-6-2020.2297944,
        author={Daham  Matrood and Muna  Sedeeq},
        title={PSO Swarm intelligence technique to optimized ANN for demand forcasting},
        proceedings={Proceedings of the 1st International Multi-Disciplinary Conference Theme: Sustainable Development and Smart Planning, IMDC-SDSP 2020, Cyperspace, 28-30 June 2020},
        publisher={EAI},
        proceedings_a={IMDC-SDSP},
        year={2020},
        month={9},
        keywords={swarm intelligence optimization artificial neural network backpropagation},
        doi={10.4108/eai.28-6-2020.2297944}
    }
    
  • Daham Matrood
    Muna Sedeeq
    Year: 2020
    PSO Swarm intelligence technique to optimized ANN for demand forcasting
    IMDC-SDSP
    EAI
    DOI: 10.4108/eai.28-6-2020.2297944
Daham Matrood1,*, Muna Sedeeq2
  • 1: Technical College of Management /Mosul, Northern Technical University, Iraq
  • 2: Computer Science Department, College of Computer and Mathematics Science, University of Mosul, Mosul/Iraq
*Contact email: daham.stat@gmail.com

Abstract

With the expansion and development of artificial intelligence algorithms and the use of swarm intelligence and artificial neural networks extensively in solving many complex problems recently. In this research, one of the swarm intelligence algorithms was used, which is represented by the algorithm of the Particle swarm optimization (PSO), and then combined it with one of the algorithms of artificial neural networks, that represented by the algorithm of error back propagation neural network EBPNN, in order to solve the problem of forecasting the demand. And the data that was used in this research was prepared by the general company for prepared clothes and the northern general company for cement represented by the weekly demand data for cement and towels. And the method of the combined PSO with EBPNN obtained the better performance than the standard back propagation neural network algorithm.