Productivity improvement of a small to medium, custom-manufacturing factory

Abstract

This report addresses productivity shortcomings at PG Aluminium (PGA) and how to address those challenges in order to improve productivity. The main problem identi ed through a business process analysis (BPA), value analysis, and work sampling study was labour productivity, as the factory's productivity is at a low of 36% when the time spent on value-adding activities are measured against the total time available for production. Productivity shortcomings can be addressed by implementing Lean principles, such as the Just-In-Time (JIT) tool and techniques. JIT, amongst other things focus on the elimination of wastes. Seven wastes are classi ed by JIT of which three of the seven were identi ed in PGA's factory. They were \waiting", \unnecessary motion", and \unnecessary inventory". These three wastes are directly hampering the ow of the fabrication process. The three wastes are therefore addressed in this study by improving the current hardware picking process, as well as through the design of a scheduling model to increase the ow of the process in the factory. The hardware picking process was analysed in more depth by doing a Business Process Analysis (BPA) which highlighted areas for improvement in the picking process. A simpli ed version of the scheduling model was designed using linear modelling principles and Python software. The model aims to produce as many products as possible in the shortest amount of time. Using the time study data collected for the scheduling model, a hypothetical individual performance measurement tool (IPMT) was designed that PGA can use to compare worker performance to expected performance. The aim is to improve the overall

ow of the factory by improving labour productivity so that ultimately the business process can be optimised. The scheduling model, hardware picking process improvement suggestions, and the IPMT were evaluated using the evaluation methods suggested by Manson (2006). Amongst other evaluation methods, a simulation model was designed using simulation software (AnyLogic), which was used to evaluate the e ect of the scheduling model and the hardware picking process improvements. From the analysis it was determined that the overall hardware picking time can decrease by up to 33.8% if the suggested improvements are implemented. The results of the scheduling model using the simulation indicates that the overall dead time (which translates into work-inprogress) can be reduced by 46.3%. From this study it is therefore clear that the overall productivity at PGA can be improved by implementing an improved hardware picking system as well as a scheduling model.