Prediction of johor river water quality parameters using artificial neural networks

Ali Najah, Ahmed Elshafie, Othman A. Karim, Othman Jaffar

Research output: Contribution to journalArticle

73 Citations (Scopus)


Water is a vital for all aspects of human and ecosystem survival and health. Thus, its quality is also important. Water quality refers to the composition of a water sample. The interpretation of data may be difficult and lengthy. Evaluations of water quality parameters are necessary to enhance the performance of an assessment operation and develop better water resources management and plan. Water quality modeling involves the prediction of water pollution using mathematical simulation techniques. In fact, classical process-based modeling approach could provide relatively good prediction for water quality parameters; however, those models rely on lengthy data and required lot of input data that often unknown. New approach such as Artificial Intelligence techniques has proven their ability and applicability for simulating and modeling various physical phenomena in the water engineering field. In additional, Artificial Neural Network (ANN) captures the embedded spatial and unsteady behavior in the investigated problem using its architecture and nonlinearity nature compared with the other classical modeling techniques. Johor River Basin located in Johor state, Malaysia which is significantly degrading due to human activities as well as urbanization in and within the area. The present study attempted to predict water quality parameters at Johor River Basin utilizing ANN modeling. This study proposed a prediction model for total dissolved solids, electrical conductivity, and turbidity. The results show that the proposed ANN prediction model has a great potential to simulate and predict the total dissolved solids, electrical conductivity, and turbidity with absolute mean error 10% for different water bodies.

Original languageEnglish
Pages (from-to)422-435
Number of pages14
JournalEuropean Journal of Scientific Research
Issue number3
Publication statusPublished - 01 Jan 2009


All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Mathematics(all)
  • Materials Science(all)
  • Agricultural and Biological Sciences(all)
  • Engineering(all)
  • Earth and Planetary Sciences(all)

Cite this