Constraint programming model for resource-constrained assembly line balancing problem

Mixed-integer linear programming models for the type-II resource-constrained assembly line balancing problem

“…In a MILP formulation, such an assumption can be integrated by appending either Constraints (13) or Constraints (14) with γ = 1. The former is equivalent to the ones used in Alakaş et al (2020), but the latter is more flexible. While Constraints (13) explicitly prohibit the assignment of tasks that require different resources ( i , j ∈ R i ∩ R j = ∅) to the same station, one can easily modify the constant on the right-hand side of Constraints (14) to reflect a more realistic limitation on the number of resources that are allowed in each station.…”

“…Even though the RM permits only one resource per station, it does not oblige each available station to take a resource as in the Alakaş et al (2020) CP model (hereafter called AM) used for comparison. With both proposed models defined, this advantage can be demonstrated in a sample scenario provided by Alakaş et al (2020). In this comparative instance, there are three types of resources (| R | = 3), and some tasks can be performed by two or more resources (alternative).…”

“…Becker and Scholl (2006) state that the RCALBP is a generalisation of the previous SALBP and surveys it alongside various practical extensions. Recently, Alakaş et al (2020) proposed a variant of the RCALBP with the goal of minimising both the cycle time and the resource usage for a given number of stations. Sample scenarios were solved with a constraint programming (CP) approach, and optimal solutions were found for all tested instances.…”

“…Such an approach had been successfully applied in a previous CP model that dealt with the ALBP with assignment restrictions (Pinarbasi et al , 2019). However, the CP method modelled by Alakaş et al (2020) has one constrictive assumption: the objective of minimising cycle time is achieved by compulsorily assigning one resource at each station. In practice, by allowing resources to be distributed between stations more freely, it may be possible to achieve the same cycle time with fewer resources or lower cycle times with the same number of resources.…”

“…Section 4 presents new MILP models, which contemplate different extensions regarding multiple-resource types and availability, as well as budget-oriented constraints, taking into account characteristics found in real-world assembly lines. Computational experiments use the data set proposed by Alakaş et al (2020) as a benchmark. Furthermore, a data set expansion is proposed to fulfil more practical concerns.…”