Performance Analysis of a Real-Time Optimization Model for a Mixed Model Stochastic Assembly Line

Rangith Baby Kuriakose (University of Technology & Central University of Technology, South Africa); Hermanus Vermaak (Central University of Technology, Free State, South Africa)

Mixed Model Stochastic Assembly lines are critical in the age of the 4th Industrial revolution. They are at the fore front of the shifting trend in manufacturing which sees moving from a Make-To-Stock approach to a Make-To-Order approach. Make-to-order systems are by nature stochastic as their production cannot be pre-planned and complicated further by the fact that variants of the same product need to be manufactured on the same assembly line. This paper looks at the results of a real-time optimization model for a Mixed Model Stochastic Assembly Line. The model was tested on a water bottling plant that needs to produce 500 ml and 750 ml bottles based on client orders that are sourced from a cloud with external constraints such as the date of delivery and internal constraints like available clean water and number of bottles. The primary results show that in all instances, the optimization model was able to meet the delivery date set by the client. However, there are some instances where an anomaly was seen, especially when completing the first order. The aim of this paper is to do an in-depth analysis of the performance of the model to ascertain its veracity and robustness.

Conference: UKSim-AMSS 22nd International Conference on Computer Modelling and Simulation, UKSim2020

Published: Mar 25, 2020

DOI: 10.5013/IJSSST.a.21.02.15