An on-demand fixture manufacturing cell for mass customisation production systems.

Abstract

Increased demand for customised products has given rise to the research of mass customisation production systems. Customised products exhibit geometric differences that render the use of standard fixtures impractical. Fixtures must be configured or custom-manufactured according to the unique requirements of each product. Reconfigurable modular fixtures have emerged as a cost-effective solution to this problem. Customised fixtures must be made available to a mass customisation production system as rapidly as parts are manufactured. Scheduling the creation/modification of these fixtures must now be treated together with the production scheduling of parts on machines. Scheduling and optimisation of such a problem in this context was found to be a unique avenue of research. An on-demand Fixture Manufacturing Cell (FxMC) that resides within a mass customisation production system was developed. This allowed fixtures to be created or reconfigured on-demand in a cellular manufacturing environment, according to the scheduling of the customised parts to be processed. The concept required the research and development of such a cell, together with the optimisation modelling and simulation of this cell in an appropriate manufacturing environment. The research included the conceptualisation of a fixture manufacturing cell in a mass customisation production system. A proof-of-concept of the cell was assembled and automated in the laboratory. A three-stage optimisation method was developed to model and optimise the scheduling of the cell in the manufacturing environment. This included clustering of parts to fixtures; optimal scheduling of those parts on those fixtures; and a Mixed Integer Linear Programming (MILP) model to optimally synchronise the fixture manufacturing cell with the part processing cell. A heuristic was developed to solve the MILP problem much faster and for much larger problem sizes – producing good, feasible solutions. These problems were modelled and tested in MATLAB®. The cell was simulated and tested in AnyLogic®. The research topic is beneficial to mass customisation production systems, where the use of reconfigurable modular fixtures in the manufacturing process cannot be optimised with conventional scheduling approaches. The results showed that the model optimally minimised the total idle time of the production schedule; the heuristic also provided good, feasible solutions to those problems. The concept of the on-demand fixture manufacturing cell was found to be capable of facilitating the manufacture of customised products.