As the U-shaped assembly lines are concerned with manual assembly operations, workers' performance plays a major role. However, the past studies assume that all workers can carry out all tasks with an equal performance level, and actual task times do not depend on the worker who performs it. On the other hand, this is unrealistic. In this study, maximization of the line efficiency for the U-shaped assembly line worker assignment and balancing problem by considering the worker performance is aimed. First, a non-linear model of the problem is formulated, and then it is linearized. Later, an Artificial Bee Colony Algorithm and a Genetic Algorithm are proposed to solve it, in short computational times. A hypothetical data set (Jackson, 1956), a real data set (Cevikcan et al., 2009) and data sets of assembly line worker assignment and balancing problem (Araujo et al., 2015) were solved by using the model and the meta heuristics. The meta-heuristics were able to reach solutions that have equal line efficiency with the optimum solution found by the proposed mathematical model. The contribution of this study is that the mathematical model for the U-shaped line assembly line worker assignment and balancing type-E problem that considers the workers' performance is formulated, and an Artificial Bee Colony Algorithm is proposed for the first time for this problem type. (C) 2017 Elsevier Ltd. All rights reserved.