Addressing uncertainty in sugarcane harvest planning through a revised multi-choice goal programming model

Abstract

In this paper a new revised multi-choice goal programming (RMCGP-LHS) model is proposed to deal with uncertainty in sugar cane harvest scheduling for sugar and ethanol milling companies. The RMCGP-LHS model uses a weekly decision-making horizon and takes into account the time and condition of land management, cane cutting decisions, and agricultural logistics. Its objective is to obtain information in order to harvest sugar cane plots in the period closest to the highest saccharose levels, while also minimizing agro-industrial costs. The RMCGP-LHS model was applied to a real case sugar and ethanol mill, and its optimization has provided harvesting policies that were validated by the company's managers. Besides that the RMCGP-LHS model is a very practical tool for simulating in a fast way different scenarios involving uncertainties on model parameters and helping the managers in decision making process in real time. (C) 2015 Elsevier Inc. All rights reserved.