Design and Selection of Facility Layout Using Simulation and Design of Experiments

摘要

In the global industrial competition, layout design and planning is becoming more and more critical due to shorter product life cycles and highly dynamic demand conditions. In this context, traditional layouts such as product and process/functional layouts are considered inferior. Cellular layouts have been proposed as an alternative to these layouts but they need a complete reallocation of resources, which consumes time and money. Hence, it is necessary to verify the performance measures before the design and selection of a certain type of layout. In this context, this paper makes an attempt to study the suitability of a virtual cellular layout (VCL) along with an existing functional layout (FL) of an industry and a classical cellular layout (CL), if considered for implementation. A Genetic algorithm (GA) based intra-cell formation procedure is used in the cellular layout design. To identify the suitability of a particular layout in a given environment, a typical manufacturing system is modeled using the WITNESS 2006 simulation software. Design of experiments (DOE) is used to plan the simulation experiments. The performance of each of the three layouts is analyzed statistically by means of operational parameters such as machine utilization, throughput, average distance traveled by parts and average work-in-process. The results from the simulation experiments indicate that the performance of virtual cellular manufacturing falls between that of functional and cellular manufacturing. Also, we find that the performance of a virtual cellular layout is often relatively superior to that of a functional layout and marginally inferior to a cellular layout.