中国矿业大学2011届本科生毕业论文 第41页
methods. These are categorized as either construction or improvement-type routines:
1. Construction-type layout routines: This type of routine generates a block layout based on the relationship between different departments. Among the most popular routines within this category are CORELAP (computerized relationship layout planning) (Lee ,and Moore 1967), ALDEP (Automated layout design) (Seehof ,and Evans 1967) and PLANET (Plant layout analysis and evaluation techniques) (Deisenroth ,and Apple 1972).
2. Improvement-type layout routines: This type of routine requires an input of a feasible block layout and aim to reduce movement cost by attempting simultaneous pair-wise (or more) position exchanging among the departments. Among the most popular improvement-type routines are CRAFT (computerized relative allocation of facilities technique) (Armour ,and Buffa 1963) and COFAD (Computerized facilities design) (Tompkins ,and Reed 1973).
Although computationally efficient, FLP heuristics are still far from meeting the constraints commonly found in daily layout tasks (Tam ,and Li 1991). Simulation has been requisitely used to incorporate many of these requirements into the facility layout study. The next section explains the benefits of simulation for facility layout studies.
2 WHY SIMULATION IN LAYOUT STUDIES?
According to Grajo (1996), layout optimization and simulation are two tasks that are crucial to any facility planning and layout study. According to Burgess et al. (1993), simulation is the only methodology robust enough to systematically examine the role and impact of product complexity and other key variables on factory performance. This is especially true because simulation models can capture many of the requirements and attributes of real life problems that are difficult to consider using analytical models for the layout optimization problems (Tam,and Li 1991; Tang ,and Abdel-Malek 1996; Pandey, Janewithayapun ,and Hasin 2000; Castillo ,and Peters 2002). Typically, a simulation study is used in layout studies to estimate system parameters associated with the following tasks: 1. Develop a series of improved layouts that has been generated using traditional facility layout routines or algorithms (Das 1993; Altinkilinc2004).
2. Contrast different layout configurations in terms of operational parameters, such as utilization, flow-time and buffer sizes (Mosier 1989; Morris, and
中国矿业大学2011届本科生毕业论文 第42页
Tersine1990; Sassani 1990; Burgess, Morgan, and Vollmann 1993; Morris,and Tersine1994; Cho, Moon ,and Yun 1996; Hamamoto, Yih, and Salvendy1999; Huq, Hensler ,and Mohamed2001; Adusumilli, and Wright 2004).
3. Evaluate various strategies for the operation of the facility or justify the embracement of manufacturing concepts such as Group technology and flexible manufacturing systems (Pegden, Shannon ,and Sadowski 1995; Taj et al. 1998; Farahmand 2000; Al-Mubarak, Canel ,and Khumawala 2003;Ranky, Morales ,and Caudill 2003).
4. Identify potential problems and bottlenecks in proposed layout structures prior to implementation (Ramirez-Valdivia et al.).
5. Compress or expand time to study the layout in steady state or under specific short-term scenarios such as product mix changes, breakdowns or emergencies.
6. Incorporate stochastic behavior and uncertainty of demand (Shafer,and Charnes 1997; Hamamoto, Yih ,and Salvendy 1999; Kulturel-Konak, Smith,and Norman 2004).
7. Use simulation model to generate random flow volumes to be subsequently supplied to traditional facility layout routines (Gupta 1986). 3 TWO SCHOOLS OF THOUGHT
Despite the wide application of simulation modeling in facility layout studies, the organization of these two tasks has been informally addressed and applied. Specifically, the question of whether a layout study should precede simulation modeling or vice versa to obtain effective layouts has been a point of debate. It seems that two schools of thought exist in this regard. The first school suggests that layout optimization should be conducted prior to a simulation study, while the second one advises that the contrary will result in better layout efficiency. Table 1 compares these two schools of thought. The details of the comparison are provided within the following sections. 3.1 Layout then Simulate
Advocates of the “layout then simulate” paradigm characterize simulation analysis as local and operational, while posting layout optimization as global and strategic. Grajo(1995; 1996) indicate that starting with a layout process with a simulation study will result in around 10%-15% improvement, while missing a much larger opportunity for improvement, had it been the case that layout