Development of Mathematical Models for Estimation of the Quantity of Biomass Residues

Abstract

Biomass residues derived from agricultural and forest-based operations and processes in Kenya have been increasing rapidly over the years without effective interventions for their efficient and economic utilization, leading to environmental and human health and safety problems. This is due to limited and unreliable documented data on the quantity of various biomass residues as well as lack of a common methodology for estimating the recoverable quantities. In this paper linear regression models of the form Y = b0 + b1X were developed for estimation of bagasse and sawdust residues using computer-based conceptual models. The models were statistically tested for their suitability in application using 95% confidence limits. The residue indices of bagasse and sawdust were found to be 0.364 and 0.528 respectively at 0.996 and 0.857 coefficient of determination,(R2) respectively. Currently there are enormous quantities of bagasse in Kenya of about 1.6 million tones annually with a potential of 2.6 million tonnes compared to between 800,000 m3 and 1.3 million m3 of sawdust. Although bagasse contribute almost 95 % of the total amount of residue generated, only 25% is economically utilised, leaving the bulk of it unexploited. Bagasse had enormous potential for utilization in modern commercial applications. In conclusion, regression models were found to be useful in estimation of biomass residues and since these residues are available in enormous quantities, they are recommended for use in various industrial applications.