Abstract
Plant layout is an important and long-term field in industrial research and practice, which can reduce capital cost, shorten production time and increase productivity. At present, most researchers studied on the layout design of facilities in a plant but not plants in an area. Industrial area-wide layout problem is more complex. The relationship between plants and the conditions around the industrial area have a significant impact on area-wide layout design, but few researchers considered these aspects. A plant is a basic production unit in the petrochemical industrial area. There are many material connections between plants. Besides, the land cost accounts for a large proportion of the total infrastructure cost. A new methodology is proposed in this work to consider both aspects. The objective function of the proposed mathematical model in this paper is to minimize the land cost and piping cost. In this paper, each plant is simplified as a rectangle with fixed area, which can be placed horizontally or vertically in the area. As some plants may have great influence on the whole occupied area, these plants are regarded as the key plants whose aspect ratios are adjusted to optimize the layout while whose areas are constant. Considering the requirement on the natural conditions, transport conditions and other factors around the industrial area, the locations of some plants are fixed. The consideration makes the layout more reasonable and practical. A new algorithm which combines surplus rectangle fill algorithm and genetic algorithm is proposed to optimize the problem. A case study is described to demonstrate the effectiveness of the proposed methodology.
