Assembly lines balancing with disabled workers

Author(s)

Mayron César de Oliveira Moreira

• Document type: Master's Dissertation
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB)
• Defense date: 2011-04-15

Examining board members

Alysson Machado Costa; Vitória Maria Miranda Pureza; Maristela Oliveira dos Santos

• Advisor: Alysson Machado Costa

Abstract

Disabled workers face enormous difficulties when trying to enter to the labor market. At the present moment, in particular in developing countries, this group constitutes a small portion of the labor force in productive processes. Among the initiatives that attempt to reverse this situation, we highlight the creation of sheltered work centers for the disabled (referred to as SWD henceforth), which are non-profit companies that employ people with disabilities, often in assembly lines. The organization and planning of the operation of a SWD involves a number of challenges. Issues related to ergonomy or production quality management, for instance, acquire particular characteristics in this environment. Likewise, classic assembly lines balancing modeling and solving techniques have to be modified, due to the significant heterogeneity among workers. In this context, we are concerned with problems related to the assembly line balancing with disabled workers, which attempts to achieve the higher production efficiency as possible, given the specific skills of each worker. More precisely, the assembly line balancing problem in SWD, known in the literature as the assembly line worker assignment and balancing problem (ALWABP), consists in assigning tasks and workers to workstations, in order to minimize the bottleneck of the production line while considering that each task duration time depends on the worker chosen for its execution. This double assignment structure leads to a much more complex problem. In this dissertation, we study a variety of techniques for solving the ALWABP. The goals of this study are, first of all, the development of a number of efficient techniques for solving the problem, both in terms of computational time required for their use and in terms of the quality of the obtained solutions. Among the techniques proposed and tested, we have versions of algorithms with different complexities, ranging from constructive heuristics and monotonic neighborhood search strategies to metaheuristics such as Tabu Search and GRASP. The diversity of the developed techniques allowed the resolution of a problem even more complex than the ALWABP, which consists of programming the line for a set of periods, taking into account the rotation of tasks among workers. The objective of this new problem is to propose a solution for a given production period that considers the fact that it might be positive to expose the workers to as many tasks as possible (for training, therapeutical and motivational reasons). In order to solve this job rotation problem, the techniques developed were integrated into a hybrid optimization scheme that uses a pool of solutions (obtained with the heuristic methods) which become inputs of mixed integer linear optimization models. The results suggest that the techniques developed are efficient and flexible to the ALWABP and their integration allows the obtention of efficient solutions to the job rotation problem. Thus, this dissertation proposes a complete scheme for the resolution of the balancing problem in SWD production lines (AU)

• FAPESP's process: 09/03128-2 - Job rotation in assembly lines with disabled workers
• Grantee: Mayron César de Oliveira Moreira
• Support Opportunities: Scholarships in Brazil - Master