|In developing countries, solid waste levels have continued to climb so quickly that municipalities have been unable to handle the increasing quantity of waste. Amidst the bleak setting of uncollected waste littering the streets, an informal solid waste processing chain based on ragpickers has evolved. Ragpickers earn income scavenging mountains of waste and organizing their pickings for sale. Understanding the complex interrelationships within the ragpicker community could help ragpickers increase their income. Researchers have yet to look at nonlinear interactions among group size, literacy, receptiveness to support from nongovernmental organizations, and resource level as factors of the productivity of processing of various types of solid waste, including recyclable, biodegradable, and inert waste. To capture these nonlinear interactions, I used an artificial neural network (ANN). An ANN does not operate with preconceived assumptions about data patterns. Instead, an ANN freely models any data pattern, capturing complex nonlinear relationships among variables. As is the case in many nontraditional business areas, it was challenging to collect rich information regarding ragpickers, so data were limited. Bootstrapping, oversampling, and the ANN were used to capture nonlinear interactions and overcome limitations of the small data set. Capturing nonlinear interactions in a small data set in this way was novel and could be a model for researchers in other nontraditional business domains who wish to uncover new relationships.