Benefits of robust multiobjective optimization for flexible automotive assembly line balancing

Abstract

Changing conditions and variations in the demand are frequent in real industrial environments. Decision makers have to take into account this uncertainty and manage it properly. One clear example is the automotive industry where manufacturers have to assume an uncertain and heterogeneous demand. For instance, automotive manufacturers must adapt their decisions when balancing the assembly line by considering different flexible solutions. Our proposal is using robust multiobjective optimization and simulation techniques to provide managers with a set of robust and equally-preferred solutions for assembly line balancing. We study a Nissan case where the demand of each product family is uncertain. The problem is addressed by considering a robust multiobjective model for assembly line balancing based on a high number of production plans. After the selection of six different assembly line configurations, we study the implications of robustness metrics based on workstations' overload. We show that the adverse managerial effects of not having flexible line configuration when demand changes are alleviated. For the real Nissan automotive case, our analysis and conclusions show the managerial and industrial benefits of using robust assembly lines. We also encourage decision makers to use robust multiobjective optimization methods for selecting the most flexible decisions.

Robust optimization for U-shaped assembly line worker assignment and balancing problem with uncertain task times

“…While many studies conducted to cope with the uncertainties in task time for assembly line balancing used stochastic programming and fuzzy logic [1,2,7,8,16], relatively few studies investigated practical applications of robust optimization for dealing with uncertainties in assembly line balancing, in particular U-shaped assembly line, see for example [6,9,11,13,18,20].…”

“…The cost of protecting a solution in case of uncertainty was investigated through the experiments. Chica et al [6] used robust multi-objective optimization and simulation method to obtain robust solutions for assembly line balancing problem. A real case study was presented through six different configurations to show the effectiveness robust approach.…”

“…While p and q are employed as auxiliary variables, Γ is used as a robust parameter determining the number of tasks with uncertain task times. Equations (5)- (6) are also added to the model to develop the robust counterpart of the nominal model proposed by [15,23]. In these equations, the interval and polyhedral uncertainty set proposed by [4] is considered to obtain tractable and non-conservative solutions.…”

Mixed integer linear programming models for minimizing ergonomic risk dispersion in an assembly line at the Nissan Barcelona factory

“…− Uncertainty in the input attributes of the tasks, such as operation time, caused by defining interval values or by setting different plausible scenarios with a set of possible values for the input attributes depending on historical data (Simaria et al 2009;Xu and Xiao 2011;Dolgui and Kovalev 2012;Gurevsky et al 2012Gurevsky et al , 2013) − The robustness of the assembly line configuration to mitigate the uncertainty defined by a set of possible demand scenarios or different demand plans (Chica et al 2013(Chica et al , 2016Li and Gao 2014;Papakostas et al 2014;Chica et al 2018) − Human resources, such as the ergonomic risks or the comfort of the production line (Otto and Scholl 2011;Bautista et al 2013Bortolini 2017;Otto and Battaïa 2017;Bautista-Valhondo and Alfaro-Pozo 2018b) In our framework, an ergonomically comfortable assembly line involves setting the maximum risk to a minimum level for any operator from the assembly line, as well as achieving a balanced sharing of ergonomic risks between the set of workstations.…”

A case study at the Nissan Barcelona factory to minimize the ergonomic risk and its standard deviation in a mixed-model assembly line

“…The set of workstations on the line, which can be finite or infinite 3. The set of sequencing constraints, such as the precedence relationships between tasks, incompatibility between tasks, and restrictions that may affect the workstations with respect to their assignable time, their available area, and their admissible risk Like the SALBP (Baybars, 1986;Scholl and Becker, 2006) and TSALBP families (Bautista and Pereira, 2007;Chica et al, 2010Chica et al, , 2013Chica et al, , 2016Chica et al, , 2018, the TSALBP_erg family focuses on assigning all tasks to workstations in order to achieve maximum efficiency regarding some of the considered attributes, while all constraints imposed are fulfilled. Accordingly, this family of problems also comprises a set of problem types in accordance with the optimization criteria.…”