Simulation-based optimisation using simulated annealing for crew allocation in the precast industry

Numerous different combinations of crew alternatives can be deployed within a
labour intensive manufacturing industry. This can therefore often generate a large
number of possible crew allocation plans. However, inappropriate selection of
these allocation plans tends to lead to inefficient manufacturing processes and
ultimately higher labour allocation costs. Thus, in order to reduce such costs
more sophisticated and innovative allocation systems are required. The main aim
of this study is to develop a Simulation-Based Multi-Layered Simulated
Annealing (“S_MLSA”) system to solve crew allocation problems encountered in
labour-intensive parallel repetitive manufacturing processes. The ‘Multi-Layered’
concept in modelling of crew allocation problems is introduced in response to the
problem-solving requirements of different sets of labour inputs such as multishifted
crews. As part of the methodology used, a process simulation model is
developed to mimic a parallel-repetitive processes layout. A Simulated
Annealing module is proposed and embedded into the developed process
simulation model for a better search for solutions. Also, the Multi-Layered
Dynamic Mutation operator is developed to add more randomness to the
searching mechanism through the solution space. A real industrial case study data
of the precast concrete labour intensive manufacturing systems is used to
demonstrate the applicability and practicability of the developed system. From
the research findings, the proposed system has the potential to produce more cost
effective allocation plans, through reducing process waiting times as compared
with real industrial based plans. Also, the main contribution to knowledge is in
the application of such innovative systems in the precast concrete industry and
the potential impact of reducing production costs and improving process
efficiency.