Neuro-fuzzy modeling for level prediction for the navigation sector on the Magdalena River (Colombia)

Abstract

The variations associated with level changes and the state of the channel in the Magdalena River in Colombia (South America) frequently affect the navigation possibilities for boats and ferries, which cause high costs for their users. For this reason, this work presents a bio-inspired model to support the decision-making concerning the navigation using a neuro-fuzzy approach developed in previous works with a novel application. Considering the average daily levels of the river registered by the limnigraphical stations from the Colombian Institute for Hydrological, Meteorology and Environmental Studies (IDEAM), during the time period between 1998 and 2003 for the Puerto Salgar, Puerto Berrio, El Banco and Calamar locations, it was possible to design and establish a neuro-fuzzy hydrological model to predict with great precision the level of water in the river for the route of navigation, allowing the appropriate decision-making for Magdalena River operators to pre-determine the weight of shipment for any boat or ferry on their route. The developed model showed better performance for the forecasting than the previously established deterministic models for this specific application. © IWA Publishing 2010.